The question of whether Hume is a moral skeptic is difficult to answer, since there seems to be no generally accepted account of what exactly "skepticism" is.⁽¹⁾ Often the notion of skepticism is little more than an abusive label, and at other times it is taken to mean only that one denies knowledge of some fact which most people accept. To say that Hume is a moral skeptic in either of these senses would, of course, be uninteresting because similar charges could be brought against almost any moral theory. The question is whether he is a moral skeptic in a more interesting sense, and what this sense might be. This is the question I want to address here.

I will approach this issue by seeing how Hume's moral theory compares to a contemporary standard of moral skepticism. Using J.L. Mackie's analysis of moral skepticism as a point of reference, I will argue that, as a normative theory, Hume's account of morality is not at all skeptical since he is offering a relatively optimistic consequentialist theory of right and wrong action. As a metaethical theory, however, I will argue that Hume is a weak metaethical skeptic insofar as he denies that morality is independent of the existence and character of human beings. He should not be considered a thorough or strong metaethical skeptic, though, since he advances a moral theory which is firmly grounded in human instinct.

It is important to note that the question of Hume's moral skepticism is often approached in an entirely different manner than my approach here. For, often it is seen as part of a larger question involving Hume's philosophy as a whole, specifically whether his epistemological and moral theories are part of the same plan. To illustrate briefly, Donald Livingston has recently argued that Hume's overall project is to advance a philosophy of common life. On Livingston's interpretation, skepticism is only a tool Hume uses to show the pitfalls of erroneous philosophical theories. Far from being skeptical, then, Hume's final views on epistemology as well as morality are based on the discourse of common life.⁽²⁾

Alasdair Macintyre also concludes that Hume's moral theory is not skeptical. He does so, however, only because he thinks Hume divorced his moral theory in Book III from his skeptical epistemology of Book I. This indicates for him a certain tension in Hume:
 

... Book III of the Treatise rests squarely upon the argument of Book II which seems to presuppose the argument of Book I, with its outcome in the problems of the concluding section of Book I and of the Appendix. The conclusion, drawn by Reid among others, that if you began where Hume began, you were bound to end with moral skepticism, was far from absurd.⁽³⁾

To avoid moral skepticism, MacIntyre argues, Hume in his moral writings maintained a "hostile" stance towards his own skepticism in his epistemological writings.

David Fate Norton agrees that Hume's epistemology is skeptical while his moral theory is not, but he gives different reasons than MacIntyre does. According to Norton, Hume developed his philosophy in response to two distinguishable kinds of skepticism, one epistemological and one moral. The former involves questions of truth and falsehood regarding matters of fact and relations of ideas, whereas the latter involves questions of the objectivity of moral distinctions. Because Hume knowingly responded to these separate kinds of skepticism in different ways, his metaphysics and morals differ significantly in method and substance. The result for Hume was a skeptical metaphysics and a nonskeptical morality.⁽⁴⁾

All three of these interpretations, which view the issue of Hume's moral skepticism within a larger philosophical context, deserve more attention than can be given here.⁽⁵⁾ The point to be noted is that a highly sophisticated discussion of Hume's moral skepticism may take place within the context of Hume's philosophical writings alone, and a complete evaluation of Hume's moral skepticism must take this into account.⁽⁶⁾ However, my narrower analysis of Hume's moral skepticism will be based only on a comparison of Hume's moral theory to a contemporary standard of moral skepticism.

Hume's moral theory (as presented in Book III of the Treatise and the second Enquiry) is for the most part a descriptive psychological account of the phenomenon of moral approval. His account contains a number of intricately connected features, such as virtues, pleasure, agreeability and sympathy. Often commentators stress only portions his theory, ignoring others. Harrison, for example, discusses only Hume's theory of moral approval (moral pleasure and pain), ignoring all else.⁽⁷⁾ Mackie stresses mainly Hume's account of virtue and moral approval, ignoring the consequentialist aspect of Hume's theory.⁽⁸⁾ Ayer simply lists eleven key features of Hume's moral theory, making no attempt to tie them together.⁽⁹⁾ Thus, before discussing whether Hume is a moral skeptic, it is important to first be clear about what Hume's moral theory is.

Hume's descriptive moral theory involves a complex chain of events involving a moral agent or actor (who does an action) and a moral spectator (who observes that action).⁽¹⁰⁾ It begins with motives or virtuous character traits in the mind of an agent, extends through actions and social consequences of that action, and ends with sympathetic feelings of pleasure in the mind of a spectator:
 
 

AGENT         |           PUBLIC PHENOMENA               |            SPECTATOR
                      |                                                                 |
MOTIVE      >    VISUAL CLUE > CONSEQUENCE  >    SYMPATHY > PLEASURE
                      |                                                                 |
   artificial       |                     overt utility                            |           moral
      or            |                     action or                               |           approval
   natural        |                       agreeability                         |

To explain this chain of events more closely, Hume argues that a spectator's feelings of moral approval are in response to an agent's motive: "'Tis evident, that when we praise any actions, we regard only the motives that produce them" (T 477).⁽¹¹⁾ Hume frequently argues that a spectator's moral judgments are made about some permanent feature of an agent since actions or external performances in themselves have no permanent merit. To find permanent merit, "we must look within to find the moral quality," that quality being the agent's motive. A problem, though, is that the spectator cannot look directly into an agent's mind and see his qualities, so we must "fix our attention on actions, as on external signs. But these actions are still considered as signs; and the ultimate object of our praise and approbation is the motive, that produced them" (T 477). Hume argues that an agent's motive is either an instinctive quality (natural virtue) or an acquired intention (artificial virtue), implying that an agent's action only gives the spectator a sign or a clue about the existence of the agent's natural or artificial motive.

As most actions have some effect on society, Hume finds four irreducible classes of effects stemming from morally approvable actions. These consequences are useful to others (general utility), useful to oneself (private utility), immediately agreeable to others or immediately agreeable to oneself.⁽¹²⁾ Whether an action has useful or agreeable consequence, all actions and all consequences are publicly observable events. For example, the spectator can observe that an act of charity has useful consequences for the recipients or that an agent's wit has an agreeable effect at social gatherings.

Once the spectator observes these consequences he will feel sympathy for the party involved (either the agent or society) insofar as the spectator's idea of a pleased agent (or society) sympathetically produces a resembling passion of pleasure within himself. For Hume, sympathy is not itself a passion but is instead a natural propensity (or instinct) whereby person A's passion is communicated to person B:
 

... when we sympathize with the passions and sentiments of others, these movements appear at first in our mind as mere ideas, and are conceiv'd to belong to another person, as we conceive any other matter of fact. 'Tis also evident, that the ideas of the affections of others are converted into the very impressions they represent, and that the passions arise in conformity to the images we form of them. (T 319)

Without the connecting link of sympathy, Hume believes that the spectator can never have feelings of pleasure or pain (i.e. moral approval or disapproval) in response to the agent's action. He also stresses that a spectator's sympathy is necessary not only for the approval of useful actions, but agreeable ones as well (T 588, 590).

A few illustrations will help tie these elements together. If an agent donates a dollar to the Salvation Army (a useful act), the spectator will feel a sympathetic pleasure for the needy who will receive the money. In other words, the pleasure which the needy experiences is transferred to the spectator. Similarly, if the agent is a comedian and performs an action which is immediately agreeable to others, the spectator will feel a sympathetic pleasure for the audience which is laughing at the agent's jokes. In both of these cases, the spectator's feelings of pleasure constitute his moral approval.⁽¹³⁾ That is, by feeling this sympathetic pleasure, the spectator thereby deems the agent's motive to be virtuous. For the Salvation Army donor, the virtue is called charity; for the comedian, the virtue is called wit.⁽¹⁴⁾

The above account of Hume's moral theory, though brief, will be sufficient to determine whether Hume is a moral skeptic.

In contemporary discussions, the term "moral skepticism" carries several meanings. Singer, for example, argues that it can mean any of the following: (1) moral subjectivism, were ethical judgments are reduced to statements about a person's or a society's approval, (2) emotive theory, were ethical judgments are only expressions of feelings, (3) ethical relativism, were moral standards are said to vary from place to place, or time to time, (4) egoism, the view that everyone is actuated by self-interest, and (5) determinism, the view that a person's actions are determined by factors beyond his control.⁽¹⁵⁾ Unfortunately, each of these meanings would require extensive analysis which cannot be done here.

There is, however, a more general understanding of "moral skepticism" proposed by Mackie which is appropriate to Hume's theory. In Ethics: Inventing Right and Wrong, Mackie makes a convincing argument for distinguishing between moral skepticism at the normative level and at the metaethical level. A normative moral skeptic would argue that all established morality should be rejected. A metaethical skeptic, by contrast, would argue that there is no objective basis for morality. Mackie argues further that the two are not mutually inclusive:
 

A man could hold strong moral views, and indeed ones whose content was thoroughly conventional, while believing that they were simply attitudes and policies with regard to conduct that he and other people held. Conversely, a man could reject all established morality while believing it to be an objective truth that it was evil or corrupt.⁽¹⁶⁾

On the surface, we may be puzzled by Mackie's suggestion that a normative theory (such as utilitarianism) could be compatible with metaethical skepticism (or the rejection of any objective basis for that view). Mackie's suggestion, though, runs parallel to the commonly recognized view of religious fideism, such as that of Kierkegaard. For, the fideist (a) believes in God, (b) thinks everyone should believe in God, yet, (c) believes there is no objective basis for the truth of the proposition "God exists." Here, the fideist would hold a normative belief in God, yet also be a meta-religious skeptic. Our search for moral skepticism in Hume's writings will begin at the normative level.

The importance and even presence of a normative side to Hume's account of morality has for the most part been overlooked. Mackie, for example, describes Hume's moral project as being entirely descriptive:
 

Hume's moral theory is not primarily an attempt to answer the first order practical question, "what ought we to do?" He is not asserting particular obligations or duties, nor putting forward a general normative doctrine like utilitarianism.... Rather, his question is a demand for an explanation of the sort typically given by the empirical sciences: "Here is this curious phenomenon, human morality, a cluster of attitudes, dispositions, practices, behavioral tendencies, and so on that we find almost universally among men, even in different societies and at different times; why is it there, and how did it develop?" (Mackie 1980, 6)

Mackie is correct in his analysis of Hume's moral theory as it appears in Book III of the Treatise. In that work, Hume's purpose is to explain the cognitive factors which give rise to ethical judgments, such as an agent's character traits, social consequence, and a spectator's sympathetic pleasure.

In the moral Enquiry, though, Hume's purpose is not just to present a descriptive account of morality. This he clearly forecasts at the outset: "the end of all moral speculations is to teach us our duty; and, by proper representations of the deformity of vice and beauty of virtue, beget correspondent habits, and engage us to avoid the one, and embrace the other" (E 172; emphasis is mine). Hume's point is that once we see which character traits are deemed virtuous by society (a descriptive project) that should tell us which character traits we ought to cultivate or "engage" in. The bulk of the moral Enquiry is a descriptive search for all character traits deemed virtuous. In Part II of the Conclusion, though, Hume returns to the theme of our normative duty:
 

Having explained the moral approbation attending merit or virtue, there remains nothing but briefly to consider our interested obligation to it, and to inquire whether every man, who has any regard to his own happiness and welfare, will not best find his account in the practice of every moral duty. If this can be clearly ascertained from the foregoing theory, we shall have the satisfaction to reflect, that we have advanced principles, which not only, it is hoped, will stand the test of reasoning and inquiry, but may contribute to the amendment of men's lives, and their improvement in morality and social virtue. (E 278)

Hume continues by giving two reasons for why we should prescribe the types of virtues discovered in the preceding chapters of his book. First, the virtues listed have advantages for society and, second, these virtues are desirable in all social spheres.

That Hume's descriptive account of morality is meant to have normative applications is, therefore, apparent. There may be, however, some confusion when deciding which descriptive elements in Hume's moral theory are foundational for a normative principle. Norton, for instance, argues that Hume is a "moral realist" in the sense that the crux of morality involves objective character traits (virtues and vices) of an agent. A spectator's moral sentiments are only "effects" of (or merely report the presence of) an agent's moral character traits. Following Norton (although Norton does not suggest this himself), the normative import of Hume's moral theory might be that we should act in accord with these objective character traits.⁽¹⁷⁾

This normative principle is clearly inadequate, though, since the mere presence of an agent's character traits (objective as they may be) provides no clue about their moral worth. Without the observation of consequences by the spectator there is neither a psychological nor practical class distinction between virtues such as benevolence and vices such as malevolence. A normative rule focusing on objective character traits would then be too broad, since it would entail an obligation to act in accord with some character traits which we normally classify as vices.

A more popular interpretation of Hume's normative theory would have us focus mainly on the spectator's moral sentiments. On Rawls's interpretation of Hume, the spectator plays a central role in normative matters: "an institution is right, let us say, if an ideally sympathetic and impartial spectator would approve of it more strongly than any other institution feasible in the circumstances."⁽¹⁸⁾ Glossop has argued similarly for a central role of the ideal spectator in determining the rightness of rules which regulate property distribution.⁽¹⁹⁾ Ayer, too, emphasizes the spectator portion of Hume's moral theory when he casts Hume as an emotivist:
 

... if we did insist on extracting from Hume a reformulation of our moral statements, we should come nearer the mark by crediting him with the modern "emotive" theory that they serve to express our moral sentiments rather than with the theory that they are statements of fact about one's own or other people's mental condition. (Ayer, 85)⁽²⁰⁾

On this interpretation, then, Hume's normative principle would be "act in such a way that a spectator will experience moral pleasure in response to your motives." The problem with this principle is that it is not informative: it tells us how to find the highest good, but it does not tell us what the highest good is (in other words, it does not tell us which things are of moral value each time they occur).⁽²¹⁾ To elicit an informative normative principle from Hume's descriptive account of morality, it is necessary to take into account the consequences of the motives which a spectator responds to, for only then will the highest good become apparent.

From my discussion in the previous section, the morally relevant consequences of an agent's motives are useful or immediately agreeable to oneself or others. This is exactly Hume's point when he concludes that "personal merit consists altogether in the possession of mental qualities, useful or agreeable to the person himself or others" (E 268, cf. E 336). Thus, the highest good is public and private usefulness and agreeableness. Hume's normative principle, then, is this: "act only from motives which produce useful and agreeable consequences."⁽²²⁾ On this principle Hume is a consequentialist. Further, Hume is a fairly optimistic consequentialist since, like Bentham and Mill, he would argue that the highest good is a desirable or beneficial consequence of actions. However, unlike Bentham and Mill, he is not a thorough-going utilitarian.⁽²³⁾ In many cases Hume may be seen as an egoist, particularly with actions that are found useful to oneself. Bravery, for example, is approved of at least in part because of its useful consequences for the possessor. Similarly, Hume might be considered an altruist when "altruism" is construed according to how good or bad the consequences are for everyone except the agent. This is so particularly with actions that are found useful to others, such as with charity.

Whether the beneficiary is society at large, the agent himself, or everyone except the agent, right actions are determined consequentially depending on whether they are useful or agreeable to the agent or others. Normative moral skepticism, as defined earlier, involves the rejection of all established morality. By this definition Hume is not at all a normative moral skeptic since he is advocating that we act in ways which promote utility and agreeableness. If there is a skeptical element to Hume's moral theory, it would therefore be only at the metaethical level. Thus, we turn next to examine whether Hume is a metaethical skeptic.

Mackie argues that metaethics involves two distinct issues. One is a conceptual issue which concerns the linguistic meaning of key moral terms, such as "good," "right," and "ought." Here, the linguist may take either a cognitivist or noncognitivist approach to the issue. The second metaethical issue is a factual one, and concerns whether there exists an objective standard of value. The issue here is ontological, and Mackie argues that it is independent of the linguistic issue. For, regardless of what we ordinarily mean by moral terms such as "good," there is an independent issue of whether there is any objective standard for moral goodness. So, to see whether Hume was a metaethical skeptic in this ontological sense, we must see whether Hume believed there is an objective foundation for his normative theory. The idea of an "objective foundation of morality" will be clarified later. Until then, we may consider it as a foundation of morality which is not merely personal preference and which has a point of reference independent of a given moral agent and spectator.

Contemporary ethicists often directly address the issue of an objective foundation of morality. Nozick, for example, specifies unambiguously five philosophical alternatives one has when dealing with this issue.⁽²⁴⁾ In Hume's day, the approach was not so direct, and the issue was not as clearly defined. In A Letter from a Gentleman, Hume replies to several criticisms launched against his recently published Treatise. Among the charges (which included universal skepticism), Hume notes that he was accused of
 

...sapping the foundations of morality, by denying the natural and essential difference betwixt right and wrong, good and evil, justice and injustice; making the difference only artificial, and to arise from human conventions and compacts. (LG 18)

The key moral issue here is clearly whether Hume denied an objective foundation of morality in the Treatise. From polemical works such as this, three issues emerge where Hume may have denied an objective value: (1) the rejection of a natural basis for justice, (2) the rejection of common rational agreement in matters of morality, and (3) the rejection of the necessity or certainty of morality. In the above order, these are the three possible areas of metaethical skepticism that I will focus on.

Hume's account of the artificiality of justice is well known. Unlike instinctively based virtues, such as benevolence, Hume argues in both the Treatise and the Enquiry that the basis for motivations of justice are instilled through education and social influence (T 477-484, E 201-204). Thus, in A Letter to a Gentleman he concedes denying an instinctive sense of justice (LG 31). Although Hume denies an objective origin of some virtues such as justice, this does not necessarily taint his entire moral theory.⁽²⁵⁾ If Hume is a consequentialist, as I have argued, the highest moral goods are the consequences of utility and agreeableness. This means that all morally commendable traits (natural or artificial) are ultimately deemed good because of their good results. So, whether justice is natural or artificial, its goodness, too, rests entirely on the goodness of the results. This would be the sole basis for the goodness of justice on any consequentialist theory, and not just on Hume's.⁽²⁶⁾ Hence, an artificial foundation of justice for Hume does not in itself commit him to metaethical skepticism. Such skepticism would result only if Hume denied an objective foundation for his claim that utility and agreeableness are the highest moral goods.

A second candidate for metaethical skepticism may be found in Hume's denial that there is common moral agreement among all rational spectators. That is, according to Hume, there is no basis for saying that superior rational beings (such as God or angels) have the same concept of morality that we do. Unlike his attack on justice, this would seem to directly undermine an objective foundation of his spectator dependent consequentialism. This line of reasoning first appears in Book III of the Treatise. Hume argues that we cannot prove a priori that universal moral relations, "if they existed and were perceiv'd, would be universally forcible and obligatory" to all rational beings, particularly to God (T 455-456). He implies further that we have no experience of superior rational beings, hence we have no a posteriori proof that morality applies universally to all rational beings (T 456). Another version of this argument appears more clearly in an early letter to Hutcheson:
 

If morality were determined by reason, that is the same to all rational beings: but nothing but experience can assure us, that the sentiments are the same. What experience have we with regard to superior beings? How can we ascribe to them any sentiments at all? They have implanted those sentiments in us for the conduct of life like our bodily sensations, which they possess not themselves. (L 40)

And, again, we find a similar argument in the Dialogues:
 

All sentiments of the human mind, gratitude, resentment, love, friendship, approbation, blame, pity, emulation, envy, have a plain reference to the state and situation of man, and are calculated for preserving the existence and promoting the activity of such a being in such circumstances.... And as the ideas of internal sentiment, added to those of the external senses, compose the whole furniture of human understanding, we may conclude that none of the materials of thought are in any respect similar in the human and in the divine intelligence. (D 156)

Hume's point in all of these is that if God exists, we have no way of knowing if he morally approves of the same character traits that we do. For, the whole fabric of moral approval is dependent upon human psychology and physiology, which, obviously, God lacks. Thus, Hume is denying that there is common moral agreement among all rational beings.

This argument presents a special problem for the consequentialist ethical theory which attempts to ground morality in the will of God. For, we are without a basis for deciding what God's will would be, or at least we cannot appeal to what humans approve of as a reference. This strikes particularly against Hutcheson's spectator dependent consequentialism which is grounded in instincts implanted in us by God:
 

...if the Deity be really benevolent, or delights in the happiness of others, he could not rationally act otherwise, or give us a moral sense upon another foundation, without counteracting his own benevolent intentions.⁽²⁷⁾

Hume would reject Hutcheson's theistic approach arguing that we simply do not know if "the Deity be really benevolent, or delights in the happiness of others." A further implication of Hume's argument is that morality loses its divine sanction. This is particularly unacceptable to those, such as Locke,⁽²⁸⁾ who feel that morality requires just punishments and rewards either in this life or the next. Paradoxically, when discussing this issue in the first Enquiry, Hume himself argues that the belief in a divine judge has an important sanctioning effect on morality, hence those who try to undermine the concept of a divine judge do a disservice to society (E 147).⁽²⁹⁾

In spite of the social benefits of extending moral approval to God, Hume's official formulation of the concept of common moral agreement excludes God or superior rational beings:
 

The notion of morals implies some sentiment common to all mankind, which recommends the same object to general approbation, and makes every man, or most men, agree in the same opinion or decision concerning it. (E 272)

Hume is advocating that for any human (as opposed to rational) spectators S1 and S2, S1's moral judgment in circumstance C will be the same as S2's. Hume's justification for this notion is empirical: from experience we know that all or most people will make the same judgments when in the same circumstances (cf. L 40). What underlies this, Hume believes, is that mental faculties are alike in every human being (E 80).

Does Hume's narrow notion of common moral agreement (which excludes superior rational beings) deny an objective basis of his consequentialism, and thereby make him a metaethical skeptic? Mackie would seem to say no. For, according to him, the issue of an objective foundation of morality is distinct from the issue of common agreement:
 

To say that there are objective values would not be to say merely that there are some things which are valued by everyone, nor does it entail this. There could be agreement in valuing even if valuing is just something that people do, even if this activity is not further validated. Subjective agreement would give intersubjective values, but intersubjectivity is not objectivity. (Mackie, 1977, 22)

Mackie would hold that common agreement in morality is a good thing, and perhaps necessary for society, but lack of common agreement does not imply that there is no objective foundation for morality. It may only mean that people simply ignore this objective foundation or even rebel against it. By analogy, there is disagreement about scientific law, yet no one reasonably doubts the existence of such law. Lack of common agreement would imply moral skepticism only if we could further establish that our plurality of moral beliefs mimic the culture we live in. This, then, would be an argument from cultural relativism (see Mackie, 1976, 36). But without causally linking our conflicting moral beliefs to our various cultures, moral skepticism would not follow. And Hume does not attempt this. Instead, Hume's notion of common agreement, even though excluding superior beings, denies that there is significant moral disagreement (which means he also rejects arguments from cultural relativism). Thus, Hume's narrow notion of common agreement does not deny an objective foundation of his consequentialism.

The final issue in Hume's moral theory which appears to deny an objective foundation of morality is his claim that morality does not involve the necessity or certainty essential to mathematics or logic. This is found in his attack on the rationalistic views of Clarke and Wollaston who held that "there are eternal fitnesses and unfitnesses of things, which are the same to every rational being that considers them; that the immutable measures of right and wrong impose an obligation, not only on human creatures, but also on the Deity himself..." (T 456).⁽³⁰⁾ The exact intention of the rationalists is far from clear, and Hume appropriately shows his frustration on this matter (E 288).

There are two ways the eternal fitness thesis may be understood. One way, suggested by W.D. Hudson, is that moral fitness is a relation between an agent's action and the person toward whom the act is directed (this interpretation may be best associated with Clarke).⁽³¹⁾ To illustrate, if A picks B's pocket, the relation that A bears to B is one which we naturally disapprove of. A second interpretation of the eternal fitness thesis (which may be best associated with Wollaston) is that there are universal ethical standards to which actions either conform or do not conform. Further, the relation actions bear to the ethical standards (whether conforming or nonconforming) is as self-evident as mathematical relations (cf. E 288-289). On either of these interpretations, then, Hume is arguing that moral approval is not a judgment about the proposition that "X bears relation R to Y." Accordingly, he concedes denying knowledge of "the eternal difference of right and wrong in Book III of the Treatise in the sense in which Clark and Woolaston maintained them, viz. that the propositions of morality were of the same nature with the truths of mathematicks and the abstract sciences, the objects merely of reason, not the feelings of our internal tastes and sentiments" (LG 30).

Hume's argument against the eternal fitness thesis is that morality would be eternal (or necessary) only if it is founded on relations of ideas which yield certainty (cf. T 70). Since this is not so, then morality is not eternal. The key premise in this argument is the denial that morality is founded on relations of ideas. His reasoning on the point is as follows (T 463-467, E 292):
 

(1) There are only four relations involved in relations of ideas.
(2) Suppose moral approval was a judgment about one of these relations.
(3) Then anything which exhibited such a relation would be judged good or bad accordingly.
(4) A young tree overgrowing and killing its parent exhibits the same relation as Nero killing his mother.
(5) Since Nero's act is morally bad, then so too is that of the young tree.
(6) Clearly, this is absurd, hence, moral approval is not a judgment about relations.

The success of this argument cannot be examined here, but Hume is clearly rejecting the view of rationalists such as Clarke, particularly because no absolute moral relation can be detected.

In place of the rationalistic concept of the necessity of moral standards, Hume argues for a morality which involves impartiality. The notion of morals, he writes,
 

...implies some sentiment, so universal and comprehensive as to extend to all mankind, and render the actions and conduct, even of the persons the most remote, an object of applause or censure, according as they agree or disagree with that rule of right which is established. (E 272)

His point is that for any human spectator S and any human agents A1 and A2, S's moral judgment in circumstance C would be the same whether C involved A1 or A2. Hume believes that we can be assured of impartiality (as opposed to necessity) in our moral judgments insofar as we instinctively pronounce similar judgments on similar actions, whether the agent is one's friend or enemy, neighbor or foreign inhabitant (T 582-583, 603, E 227).

Does Hume deny an objective foundation of morality when he denies moral necessity (in favor of a more narrow notion of impartiality)? Given the vague notion of "objective" we are working with, it is hard to tell. Until now in our discussion, a clear definition of "objective" was not necessary since Hume's theories of artificial justice and common human agreement posed no threat. To address this problem, perhaps we can understand morality to be objectively grounded if and only if,
 

a spectator's moral approval is not uniquely dependent on himself or on the time and place of his approval.

Although this definition is some improvement, the vague concept of "unique dependence" is still problematic. On one reading, Hume might not deny an objective foundation of morality, for his notions of common human agreement and impartiality, grounded in instinct, assure us that a spectator's approval is not unique to himself, or to the time and place of its occurrence.

On another reading of "unique dependence," though, it seems that Hume does deny an objective foundation of morality. For, a spectator's approval, although not unique to himself, is still dependent on his having certain human instincts which assure impartiality and common agreement. The very existence and nature of instinctive moral standards is a contingent matter, so it would seem strange to call such instincts "objective." Further, a purely instinctively grounded morality would not be nearly as objective as either a moral dictate by God (according to divine command theory) or a natural order to the universe (such as Tillich's notion of the ground of being or Clarke's notion of eternal fitness).

The question of Hume's metaethical skepticism hinges on whether an instinctively grounded morality should count as "objective." We may come to some resolution on this topic if we distinguish between a strong and a weak sense of an "objective foundation of morality." I will use the phrase "objectively strong foundation of morality" to refer to morality not at all dependent upon the existence or constitution of human beings (a la Clarke). By contrast, an "objectively weak foundation of morality" will refer to morality grounded in human instincts alone.⁽³²⁾ From this distinction, a further distinction would arise between (1) weak metaethical skepticism which entails the acceptance of an objectively weak foundation of morality and the denial of an objectively strong foundation, and (2) strong metaethical skepticism, which entails the denial of both objectively strong and weak foundations of morality. In this light, Hume should be seen as a weak metaethical skeptic given his adherence to consequentialist morality firmly grounded in instinct and his rejection of the eternal fitness hypothesis. This would place Hume in the same class with ethicists such as Ross who ground morality in human instincts (although Ross takes a nonconsequentialist approach).

How pessimistic is Hume's weak metaethical skepticism? That is, does Hume's denial of an objectively strong foundation "sap the foundations of morality" as he was accused? My own view is that the issue of an objectively strong foundation of morality is less of a moral issue than it is a religious issue. It is a religious issue when philosophers or theologians ground morality in an objectively strong divine command theory, a Whiteheadian process theory or a Tillichian ground of being. When taken in these ways, an objectively strong foundation of morality is largely a commentary on the nature of God. Denial of these kinds of objectively strong foundations (in favor of the weak variety) would then be tantamount to the denial of a type of theism (e.g. the denial of a God who creates absolute morality). But such a denial would not undermine the nature of morality. Morality grounded in instinct could bring about a cross-cultural, impartial and obligatory normative theory which would be in good moral tradition. This, I believe, is what Hume was striving for. I, therefore, agree with Hume's own assessment of his moral theory that "...tho' the clergy be always enemys to innovations in philosophy, yet I do not think they will find any great matter of offense in this volume. On the contrary I shall be disappointed, if impartial judges be not much pleas'd with the soundness of my morals" (L 37).

In summary, I have argued that Hume is not a normative skeptic since he is advancing a fairly optimistic view of consequentialism and is not suggesting that we reject all established morality. On the issue of metaethical skepticism, I argued that an objective foundation of morality is not undermined either by Hume's account of artificial justice or by his narrow notion of common moral agreement (which excludes superior rational beings). Finally, I have argued that Hume's denial of moral necessity makes him only a weak metaethical skeptic, since his instinctively based morality has an objectively weak foundation. Accordingly, Hume's consequentialist normative theory and its objectively weak foundation suggest a comparatively optimistic view of morality.

1. Quotes from David Hume are from, A Treatise of Human Nature, second edition, ed. Selby-Bigge, (Oxford: Clarendon Press, 1978), hereafter abbreviated T; Enquiries Concerning Human Understanding and Concerning the Principles of Morals, third edition, ed. Selby-Bigge and P.H. Nidditch, (Oxford: Clarendon Press, 1975), hereafter abbreviated E; Dialogues Concerning Natural Religion, ed. Norman Kemp Smith, (Indianapolis: Bobbs-Merrill, 1947), hereafter Abbreviated D; The Letters of David Hume, ed. J.Y.T. Greig, (Oxford: Clarendon Press, 1932), hereafter abbreviated L; A Letter from a Gentleman from his friend in Edinburgh, ed. Ernest C. Mossner and John V. Price, (Edinburgh: University of Edinburgh Press, 1967), hereafter abbreviated LG.

2. Donald Livingston, Hume's Philosophy of Common Life (Chicago: University of Chicago Press, 1985), 3.

3. Alasdair Macintyre, book review of David Fate Norton's David Hume, Nous, (1984), 381.

4. David Fate Norton, David Hume: Common-Sense Moralist, Skeptical Metaphysician (Princeton: Princeton University Press, 1982).

5. All three of these interpretations have problems. Livingston's interpretation (similar to Kemp Smith's) begs the entire question of Hume's skepticism. For, even if Hume's final views on epistemology and morality are grounded in "common life" (or are naturally based), the question remains about whether these views are themselves skeptical. MacIntyre's position that Hume radically divorced his skeptical epistemology from his non-skeptical morality is in direct conflict with Hume's conviction that his system of morals will corroborate or verify "whatever has been said concerning the understanding and the passions" (T 455). Although Norton is correct in separating moral from rational decision making, he fails to prove his claim that the latter is skeptical while the former is not. Most significantly, Norton ignores Hume's own comments admitting to moral skepticism (E 278, cf. LG 30-31).

6. In my Hume's Moral Skepticism (Dissertation, Purdue, 1986), I address the issue of Hume's moral skepticism in the context of his larger philosophical project. There I argue that, given Hume's own conception and definition of "skepticism," key elements of his moral theory are indeed skeptical.

7. Jonathan Harrison, Hume's Moral Epistemology, (Oxford: Clarendon Press, 1976).

8. J.L. Mackie, Hume's Moral Theory, (London: Routledge and Kegan Paul, 1980).

9. A.J. Ayer, Hume, (New York: Hill and Wang, 1980), 80-81.

10. Many 18th century British moral theories rest on this division of labor between a moral agent and a moral spectator, particularly moral sense theories such as those of Shaftesbury, Hutcheson, Hume and Smith. See D.D. Raphael, The Impartial Spectator, (London: Oxford University Press, 1972), 6-7.

11. Hume's use of the terms "motive" and "action" are unusually broad. By "motive," he means an either conscious or unconscious mental quality, character trait or intention which persists in an agent's mind. By "action" he means any behavioral movement, series of behavioral movements or mode of behavioral movements which arises from a motive. Virtues and vices, then, are specific motives of an agent which generate feelings of moral approval in the mind of a spectator. Bentham and Mill distinguish between motives (durable mental habits) and intentions (mental inclinations which immediately precede an action); for Hume, intentions in this sense, if present, arise from more durable motives. J.S. Mill Utilitarianism, Indianapolis: Bobbs-merrill, 1957, 24.

12. Most morally significant actions seem to produce more than one of these results. Also, it is important to note that there are four additional classes of effects stemming from actions which are subject to moral disapproval.

13. Not all sympathetic pleasures felt by a spectator are feelings of moral approval. Pleasures of moral approval must (1) produce additional feelings of love, hate, pride or humility, (2) be focused on an agent's mental quality, and (3) be impartial (T 472-473). These conditions distinguish moral approval from aesthetic approval and selfish approval.

14. Hume has a simplified sequence of moral approval which develops out of the formation of general rules. He observes that interrupted actions or mere intentions without actions are evidently approved of even when they produce no useful results. The explanation is that we form habits or general rules based on past experiences of completed actions which had proven useful. Through sympathy, the spectator will then experience moral pleasure from the idea of a useful result (T 584-585).

15. Marcus Singer, "Moral Skepticism," in Skepticism and Moral Principles, ed. C.L. Carter, (Evanston: New University Press, 1973), 79-83. See also Marcus Singer, Generalization in Ethics, (new York: Atheneum, 1971), 7-8.

16. J.L. Mackie, Ethics: Inventing Right and Wrong, (New York: Penguin Books, 1977), 16.

17. David Fate Norton, "Hume's Moral Ontology" in Hume Studies: Tenth Anniversary Issue, (1985), 189.

18. John Rawls, A Theory of Justice, (Cambridge: Harvard University press, 1971), 185.

19. Ronald J. Glossop's "Is Hume a Classical Utilitarian," Hume Studies, (1976), 1-17.

20. Although Ayer is suggesting a metaethical (linguistic) interpretation of Hume's moral theory, his emphasis on the role of the spectator has implications for Hume's normative theory as well.

21. Cf. G.E. Moore, Principia Ethica, (Cambridge: University Press, 1903), 9-10.

22. Hume would seem to be both an act and a rule consequentialist. For, actions arising from natural virtues are right insofar as they are individually useful and those arising from artificial virtues are right insofar as they are useful as a rule. Although this suggests a mixed stand on the act/rule issue, it is not clear whether the artificial rules are merely generalizations of and logically posterior to particular actions (which Rawls calls the summary view of rules) or whether the rules are logically prior to actions (or the practice view). The difference as Rawls describes it is that, on the summary view, rules can be disregarded on utilitarian (or consequentialist) grounds if a greater good results from particular actions which are contrary to that rule. On the practice view, however, the agent is obligated to forfeit any direct appeal to the utilitarian standard in particular cases. Cf. John Rawls, "Two Concepts of Rules," in Philosophical Review, (1955), 3-32.

23. There are some obvious differences between Hume's consequentialism and classical utilitarianism such as that of Bentham and Mill. For example, Bentham and Mill play down both the role of the agent's motives (i.e. virtues and vices) and the spectator's approval (i.e. moral sentiments). Thus, they focus almost exclusively on the consequences produced by an action (cf. J.S. Mill Utilitarianism, 24).

24. Robert Nozick, Philosophical Explanations, (Cambridge: Belknap Press, 1981), 555-558.

25. Mackie believes that Hume's account of artificial justice is skeptical, at least when compared to the views of his contemporaries, such as those of Locke, Clarke, Wollaston and Hutcheson (Mackie, 1980, 82).

26. The issue of explaining justice, rights and duties within a consequentialist framework is one of the traditional problems of utilitarianism. See Jonathan Harrison's "Utilitarianism, Universalization and the Duty to be Just," in Mill: Utilitarianism, ed. Samuel Gorovitz, (Indianapolis: Bobbs-Merrill, 1971), 151-167.

27. Francis Hutcheson, An Inquiry concerning the Original or our Ideas of Virtue or Moral Good, in British Moralists, ed. Selby-Bigge, (Oxford: Clarendon Press, 1897), vol. I, 175.

28. John Locke, An Essay concerning Human Understanding, (Oxford: Clarendon Press, 1975), 352.

29. The dialogical context of this argument makes it difficult to attribute this view to Hume. Nevertheless, Hume clearly recognized the negative impact on morality of denying a divine judge.

30. Hume's account of the eternal fitness thesis here is a reference to Samuel Clarke's view. Cf. On Natural Religion, in Selby-Bigge's British Moralists, vol. II, 12.

31. W.D. Hudson, Ethical Intuitionism (New York: St. Martin's Press, 1967), 26.

32. An objectively weak foundation of morality might further be extended to values grounded in firm human conventions which are not instinctive, such as Mill's notion of acquired duty.