MORAL FACULTY THEORY AND ITS FORGOTTEN VIRTUES

WHAT IS PROBATIVE ABOUT OUR MORAL INTUITIONS?

Philosophers love to dream up examples, to reflect upon what would be true were various possibilities real. They do not regard this as aimless speculation, but rather an important way to test philosophical principles. For example, were the moon really green cheese, everyone would agree that you couldn't know that it is only paper. Many would take such agreement to be conclusive evidence in favor of the familiar principle that knowledge implies truth. They thereby make these assumptions: firstly, that those to whom such examples are addressed share a conception of knowledge which allows them recognize instances of it, and secondly, that it is epistemology's task to describe and explain this common conception. It is otherwise hard to see how shared intuitions about examples could have any probative force for questions of philosophical principle.[1]

Some philosophers believe that the method of examples is also legitimately employed in normative theory. Its use here is controversial,[2] but a brief glance through the journals shows that many rely upon it. What propositions do they presuppose?

Philosophers who employ the method of examples in normative theory intend more than merely to register their opinions. They believe that their intuitions are true; and they take agreement in their intended audiences to be important confirmation of what they maintain. Their practice therefore presupposes that they and their intended audiences share an ability to discover what is morally true under hypothetical circumstances. Resuscitating an old label, let us call the "moral faculty" whatever capacity is thereby supposed.

A variety of moral faculty theories may be found in the history of western ethics; and undoubtedly there are many possible others waiting to be articulated.[3] One dimension on which these theories differ is defined by their sundry metaphysical views of the nature of the moral faculty and of moral truth. Historically, there have been two leading candidates: firstly, that the moral faculty is Reason, which apprehends by direct insight the synthetic necessary truth of fundamental moral principles;[4] and secondly, that it is a particular kind of Sentiment, which both discerns and determines the contingent truth of such principles.[5] A third candidate, albeit historically unpopular, is ethical naturalism of the kind dubbed a Fallacy by Moore.[6] Naturalists hold that substantive moral principles are necessarily true because of the meanings of the sentences that express them. A naturalist moral faculty theorist might identify it as being the capacity to employ a moral vocabulary without semantic deviance.[7]

Which of these candidates could be true, or whether a metaphysically more adequate moral faculty theory is yet to be formulated, are questions of the first importance for the metaphysics of morals, but are irrelevant to normative theory.[8] Regardless of their metaphysics, moral faculty theorists will give probative force to exactly the same kinds of evidence when they test normative principles: they will rely upon their own intuitions and those of their intended audiences. The moral rationalist will call these the deliverances of Reason; the sentimentalist will think of them as an nonrational causal product of nonmoral belief; & etc. But the content of what they intuit and the evidentiary force which they ascribe to their intuitions will be unaffected by their metaphysical differences. If the rationalist and the sentimentalist agree that breaking a serious promise cannot be justified solely on the ground that a slight surplus in utility will result, they must agree that act utilitarianism cannot be a wholly adequate theory of the right.

What matters for normative theory is not the metaphysical nature of the moral faculty but rather how widely it is distributed. Philosophers’ differences here have had profound consequences for their views about the proper goals of normative theory, which have in turn shaped the form of the theories they have devised.[9] Some philosophers, whom we might call Platonic elitists, believe that nonphilosophers cannot hope to obtain moral knowledge without philosophers’ assistance, and that nonphilosophers' moral intuitions cannot be deemed correct until confirmed by philosophical argument.[10] Hence, they ascribe a revisionary task to normative theory, that of framing a catechism of moral principles that humanity ought to accept and put into practice. For example, Richard Hare distinguished between “levels” of moral thinking. He recommends to philosophers that they engage in “critical thinking,” which proceeds “in accordance with canons established by philosophical logic and thus [is] based upon linguistic intuitions only;” and he argues that such thinking reveals the rational preferability of a variant of rule utilitarianism.[11] But philosophers should also construct a system of principles for “intuitive thinking,” which will promote utility when generally followed and which are sufficiently simple for nonphilosophers to understand and to apply. We may understand Hare and many other elitists as espousing a restricted moral faculty theory, one that supposes extensive philosophical education to be necessary for developing an adequate capacity to perceive moral truth.[12]

For the present, let us set elitist theories aside to focus upon what I shall call commonalism, which holds that the capacity for moral judgment is universal and fungible, one of the incidents of our common humanity.[13] Commonalism implies that nonphilosophers’ moral beliefs are no less authoritative than philosophers’, and hence, that the task of normative theory is not to revise, but rather to discern and explain the shared moral conception which we all apply in our ordinary moral lives. What can be said for it?

I shall not attempt any elaborate theory about faculties of judgment. What I think commonalists meant (or should have meant) in positing a faculty of Reason, or moral Sentiment or Sense, is a model of moral judgment involving (in contemporary jargon) a mentalistic "black box," into which nonmoral beliefs are fed as stimuli, and from which moral conclusions issue. Since normative theory seeks to discern the truth conditions of our moral judgments, I shall suppose that causal influence runs only from nonmoral belief to moral belief. But a psychologically accurate model would posit reciprocal causal relations. Many of our moral beliefs are unreflectively taken from our families, friends, cultures, religions. When moral beliefs are so acquired we (often unconsciously) adopt a set of nonmoral beliefs to justify our holding them..

The "black box" model of the moral faculty supposes that there is an important distinction to be drawn between moral and nonmoral belief, and also that, in tracing causal and inferential relationships between these kinds of belief, philosophers may rely upon their pretheoretical ability to distinguish clear cases in each category.[14] This methodological assumption is recommended by the familiar metaethical principle that moral properties supervene upon “factual” ones: when we judge that something is good or that an act is right, we so suppose because it is of a certain kind or nature, or has certain results, etc. The distinction between the moral and nonmoral is also supported by the principle of universalization, which in its sundry formulations requires us to treat relevantly similar cases in a morally similar fashion.[15] However, the moral/nonmoral distinction and the principles of supervenience and universalization do not establish commonalism, because they are consistent with the possibility that individuals process nonmoral beliefs in radically different ways. Commonalism makes an additional assumption, viz., that the moral judgment of humanity tends to converge under ideal conditions, i.e., when we perfectly agree upon all relevant propositions of nonmoral fact and when our considered judgments are unaffected by bias. Let us call this the convergence hypothesis.

Some readers may be tempted to respond that there is an obvious fact of experience which immediately falsifies the hypothesis, viz., the profound moral disagreement that exists both between individuals within societies, and between the dominant ideologies of myriad cultures. But it's silly to suppose this a modern discovery. The great commonalist philosophers were as aware of moral disagreement as anyone is today. They explained it by appealing to the phenomena of nonmoral factual disagreement and of bias. We must explore this defensive maneuver at some length.

Let us first cast their point in another useful modern idiom. Borrowing the terms from Chomskian linguistics, we may distinguish between "surface" and "deep" moral opinion. Surface opinion is that which we can discover by conducting surveys, e.g., as to the propriety of enforcing surrogate motherhood contracts, where our sole interest is to determine what percentage of the population approves. But surface moral opinion always supervenes upon nonmoral belief. For example, brief reflection reveals that our surface beliefs about the morality of enforcing surrogate motherhood contracts depends upon such factors as these: what we believe the predominant effect will be upon surrogate mothers, the contracting families, and the offspring of such unions; what we believe its effect will be generally upon the structure of families, and also whether we believe the present structure promotes happiness or produces misery; and finally, too, upon our theological beliefs, upon whether we think that God has issued relevant commands. This list of factors is neither precise nor exhaustive, but it does give some idea as to the complexity of the nonmoral issues upon which surface moral opinion frequently depends.

Surface moral opinion is a product of relevant nonmoral beliefs about this and other possible worlds and of our deep moral opinions, which are hypothetical in formal structure and which link nonmoral beliefs to determinate conclusions expressed in fundamental moral concepts, such as rightness or goodness. Deep opinions are more often revealed by moral argument than consciously relied upon. (Racists who spout preposterous beliefs about the capacities of those whom they hate and disdain thereby show that they too believe it unjust to visit serious disabilities upon people for possessing morally insignificant properties.) Moreover, the various kinds of bias, e.g., personal interest, race, gender, class and cultural associations, and even one's philosophical commitments,[16] also play an important causal role. Bias skews nonmoral judgment, causing racists and sexists to hold patently false beliefs. But often too it skews deep moral opinion; and at times it even prevents interaction between deep opinion and nonmoral belief, as when one recognizes that one's adult child has willfully caused grave injury, but yet fails to condemn her act or even to perceive its wrong.

We may now cast the convergence hypothesis anew: that virtually everyone holds the same deep moral opinions. This hypothesis is evidently consistent with an enormous diversity in surface moral opinion, particularly between people of different religious belief.[17] It could be falsified only by a showing that there is widespread disagreement among humanity's deep moral opinions. How might we test it?

III

Here we must rely upon an informal canvass of our individual and collective experience. The hypothesis evidently concerns a matter of contingent fact: possibly our moral judgments converge; possibly they do not; neither alternative can be ruled out a priori. But we don't meet with ideal conditions in everyday experience.[18] Hence, we must proceed indirectly. We must find relevant facts about the human practice of making moral judgments, and then ask whether the convergence hypothesis best explains these facts. I believe that it will but I won’t offer to prove this. My aim is more modest: to make the convergence hypothesis appear reasonable—something worthy of belief in the absence of decisive reason to the contrary. [19]

But before we consider any evidence we must first distinguish between deep moral disagreement and what we may call competition indeterminacy. Commonalists typically suppose that there are universally valid moral standards, but that these state moral reasons (i.e., considerations that count for or against particular moral conclusions), which can compete with one another in particular circumstances, rather than exceptionless general rules of rightness, goodness, etc. When reasons compete, it is often easy to tell which wins out: e.g., if I have promised to meet with a student about her term paper which is due in a week, but on my way to the meeting I happen across a badly injured person whom only I can get to hospital, my choice is simple. But suppose that the student needs my signature within the next ten minutes else she will lose a hefty scholarship and that the injury is painful but transitory—what then? We can set out many similar examples, adjusting the harm done by the broken promise and the severity of the injury. We shall probably disagree about some of these cases without disagreeing over the facts. (We’ve stipulated these.) We thereby can reach points of moral indeterminacy. But there’s no deep disagreement between us. We all agree that our promises have a moral claim upon us, as does injury to a stranger that we can assuage without cost or risk. We disagree in some cases only as to the comparative weight of these reasons. Let us reserve the term ‘deep moral disagreement’ for those intractable controversies wherein parties disagree because they sincerely fail to recognize the moral standards that others rely upon.

Perhaps the strongest evidence in favor of convergence is the mundane fact that we fruitfully engage with one another in political discussion and moral argument. Were deep moral disagreement common, we would often find each other’s reasoning unintelligible. We should find ourselves saying such things as, “Let me get this straight. You really think that the pleasure you gain from others’ pain makes it morally right for you to inflict it upon them?” Or, “You think that the fact that Sally helped Mary get in her crop justifies Mary’s present refusal to help Sally in the same task—is that it?” We don’t find one another surd in exactly this way. We are often baffled by others’ surface moral opinions, but virtually always because we can’t understand how anyone could believe their supporting factual premises. I have never understood how anyone can believe that a loving God hates gay people and punishes severely any society that tolerates them. But we all can understand how someone who holds this belief would disapprove of gay liberation.

Moral understanding transcends particular societies and cultures.

We find lots of wickedness, self-deception, insincerity and hypocrisy at the United Nations. But we find too that diplomats understand each other’s arguments, and that they often quarrel about each other’s factual premises but rarely about pure moral principle. No one defends acknowledged unprovoked aggression. Hence, it appears that there is a transnational understanding of moral reasoning. This fact is more plausibly explained by positing a common moral faculty than by any historically contingent explanation, e.g., that there is a Western moral conception that now reigns in international discussion because of imperialism and Western prowess at arms.

The fact of universal moral understanding implies that deep moral disagreement cannot be widespread, but not that it never exists. For example, it may well show up in debates about the role of luck in the moral assessment of action. If I attempt to murder another but am prevented by something outside my control, it seems intuitively obvious that I am as blameworthy and as culpable as had I succeeded. Nonetheless, in most systems of criminal justice, my punishment will be much less severe.[20] Most theorists of the criminal law share an intuition that the traditional attempts rule is reasonable. But they also find it paradoxical, since it conflicts with the intuition that the severity of punishment ought be proportioned to culpability. However, others reject the traditional rule altogether. They point out that an act’s actual result is often (perhaps always) beyond an agent’s control; and they contend that it is unjust to have the severity of someone’s punishment turn upon her good or bad fortune.[21] Since this clash in principle seems not to stem from disputed factual premises or from competition indeterminacy, it appears to be an instance of deep moral disagreement. Hence, it seems that the moral faculty is indeterminate on this issue of moral luck. Still, it is not wholly silent. Indeterminacy means that society may permissibly retain the traditional rule or may instead punish attempts and completed crimes equally severely. But we shall agree that societies are not morally free to adopt other possible alternatives, e.g., that attempts are to be punished more severely than completed crimes or that attempts may not be punished at all.

Are there other instances of deep moral disagreement? Almost invariably, when people are challenged to produce an example, they seize upon abortion: they say, “Pro-life and pro-choice people don't differ over any nonmoral fact about human conception. Of course, they do disagree about when life begins, but that is really the disguised moral question about the stage at which products of human conception enjoy a right to life.” This is a plausible account of the debate. And, if there really be deep moral disagreement about abortion, the commonalist must concede that the moral faculty is indeterminate with respect of this issue as well. Some readers may find this disappointing. Nonetheless, since all classificatory schemes are afflicted by some indeterminacy and vagueness, a few instances of deep moral disagreement is insufficient reason to abandon the convergence hypothesis, if it proves faithful to the rest of our moral experience. So I invite the reader to check her own experience, to see if the convergence hypothesis fits generally.

Still, we may have conceded too much to the foes of convergence, for it is far from clear that there is deep moral disagreement about abortion. Those who debate the issue typically disagree about all sorts of non-moral facts: e.g., about whether God disapproves, about the psychological and physiological effects upon women who choose to abort, about the life possibilities of children born to teenage mothers (or what these could be were society to make “a commitment to life”—and whether such a commitment would require us to sacrifice other valued social goals), about the life possibilities of teen age parents (or what these could be. . . ), and even about whether a society that tolerates abortion must form calluses upon its conscience, thus raising the probability that it will adopt a policy of exterminating the incurably ill or developmentally disabled. Also, the heat of the controversy makes it fertile ground for bias, inclining the disputants to slight the weight of countervailing considerations. Despite the fervency of disagreement over abortion, it simply isn't at all clear that there is deep moral disagreement about it, at least not in the technical sense that we use here.

Here those viscerally opposed to convergence and commonalism are likely to have their doubts crystallize into a methodological objection: they will say, “Your strategy is now clear: You're going to reject any putative counterexample to convergence by claiming that it's merely an instance of surface moral disagreement. But this makes the convergence hypothesis unfalsifiable, and so a trivial tautology. Your case is no better than that of the psychological hedonist, who rejects falsifying examples such as soldiers who save comrades by diving upon grenades, by ascribing to them (seemingly on a priori grounds) a belief that any action other than self-sacrifice must prove intolerably painful.”

But the case for convergence is very different from that usually offered for psychological hedonism. The empirically plausible counterexamples to the latter view are legion; and there is no reason to suppose, e.g., that heroic soldiers must foresee intolerable pain from eschewing self-sacrifice, apart from the fact that the hypothesis is saved by ascribing this belief to them. Psychological hedonists often do defend what must be an empirical hypothesis in a manner more appropriate to defense of a necessary truth.[22] This is beyond doubt a methodological error; but it is not found here. There are very few empirically plausible counter-examples to the convergence hypothesis—as we should expect if (as commonalists believe) it expresses a fundamental truth about our common humanity. Still, the hypothesis concerns empirical reality, not metaphysical necessity. There is no difficulty in conceiving circumstances that falsify it: for example, humanity might have had nothing like a morality at all.

Suppose with me that, subject to instances of indeterminacy, the convergence hypothesis is worthy of reasonable belief. Since convergence is an essential part of commonalism, this result gives reason to believe that some variant of this theory must be true. Let us now gauge the import of convergence for other metaethical theories.

Four important positions concerning the possibility of attaining moral knowledge have been articulated in the history of ethics. There is commonalism, which holds that objective moral truths exist and that virtually everyone who is unbiased and properly informed about relevant nonmoral fact has the capacity to know them. Then there is cognitivist Platonic elitism, which also allows that objective moral truths exist but that only those with extensive philosophical training can discern them. Thirdly, there is relativism, which admits of moral truth and knowledge, but insists that this is always relative to the diverse conventions of different societies and cultures.[23] And lastly, there is moral scepticism, which denies moral truth and a fortiori the possibility of moral knowledge.

The convergence hypothesis counts most evidently against moral scepticism, inasmuch as its most powerful argument has always been that humanity cannot agree upon what is true in morals.[24] But convergence also undercuts the various forms of relativism. Relativism purports to explain how moral constraints can be real despite the absence of an objective morality binding upon everyone; and its deep rationale is to permit its proponents to escape moral nihilism though they be persuaded by sceptical argument.[25] And, like commonalism, relativism takes what people agree upon as its criterion of moral truth. But if a deep moral consensus among humanity really does lie behind surface moral diversity, sceptical arguments are defanged, nihilism poses no threat, and there is no point in hovering at a relativistic surface plurality of “truths.” Relativists persuaded of convergence should find no theoretical obstacle to joining commonalists in looking to this deep agreement for objective principles of moral truth.

Finally, as for Platonic elitists, one might initially suppose convergence to be partial confirmation of their views, since they often claim that ordinary folk are in the grip of some imperfect morality which they then criticize. However, their folk “moralities” are formulated on the basis of only cursory investigation. For example, Derek Parfit criticizes “Common-Sense Morality,” which he identifies as a theory: 1) that posits special obligations towards those who stand in certain relationships to us (e.g., parents, children, clients, etc.) which are more stringent than obligations to strangers;[26] and 2) that “[w] e successfully follow this theory when each does what, of the acts that are possible for him, best achieves his [special] moral aims.”[27] His criticism is that the theory is “self-defeating,” by which he means that in some circumstances (those where collective action better benefits those to whom special obligations are owed) the moral aims of each are best achieved when the theory's second condition is not satisfied, i.e., when everyone cooperates in collective action rather than attempting individually to maximize achievement of her special moral aims.[28] This undoubtedly is an unhappiness in the theory, which we all will recognize when it is brought to our attention. But then, this latter fact alone shows that the theory cannot be that upon which we all deeply agree.[29] Parfit’s argument is therefore misconceived—or at least mislabeled. It is not an argument against anything that could reasonably be called a Common-Sense Morality, i.e., one that we all share. Rather it shows that this morality (if it exists) is not captured by a particular candidate theory.

This last point suggests a deep difficulty for anyone who undertakes to criticize the moral conception upon which deep moral agreement converges, viz., that of finding an Archimedean point apart from common agreement from which such criticism could proceed. Our elitist critic's task has two parts: to specify the principles of our common moral conception; and then to provide arguments that show these principles to be mistaken. In its first part, he must use the intuitionistic method of examples that is appropriate for discovering the details of a commonly held conception, whether this be in ethics, epistemology or any other part of philosophy: i.e., he will test intuitively plausible principles against intuitions about examples, and vice versa, until the total body of intuitions settles (in Rawls' useful phrase) into “reflective equilibrium.”[30] Because convergence implies that each of us has the same moral conception somehow installed within, for the most part he will rely only upon his own intuitions; but he will also check these against those of others, in order to correct for his own biases and also for limitations in his ability to imagine perspicuous testing examples. Along the way he will be forced to discard many principles that at first seemed very plausible, but that upon reflection proved inconsistent with his secure intuitions about examples or with other principles that he is confident have already been established. He will find this first part of his critic's task to be exceptionally difficult: we employ our common moral conception with great facility; and it is not until we attempt to set it out in principled array that we appreciate its immense complexity.[31] But suppose that the elitist philosopher succeeds in formulating a theory that she believes captures our common moral conception: what criticism might she then make of it? She cannot find it morally defective, for it is the theory that best fits her own moral intuitions. Perhaps only were it radically inconsistent or incoherent might she reasonably seek to supplant it. But then she must question whether her candidate theory really does capture our common moral conception, for such a defect is hard to square with the hypothesis that our moral beliefs do converge. Our argument against Platonic elitism may thus be summed: If the convergence hypothesis is true, then the elitist's opinions converge along with everyone else's. Hence, if she rejects the convergent moral conception, she thereby rejects her own best reflective moral judgment. What possible incentive could anyone have to do this?

Before 1800, virtually every important secular moral theorist held some variant of commonalism; today virtually none do. This is an extraordinary transformation in intellectual fashion, which cries for an explanation. But contemporary ethicists and historians of ethics are mostly oblivious of it. One exception is Alisdair MacIntyre, a relativist, who scorns the older view:

Ever since the Enlightenment our culture has been far too hospitable to the all too plainly self-interested view that whenever we succeed in discovering the rationality of other and alien cultures and traditions, by making their behavior intelligible and by understanding their languages, what we will also discover is that in essentials they are just like us.[32]

MacIntyre's complaint is hard to understand. Similarity being a symmetrical relation, if we come to believe that people of other cultures are essentially like us, we must also believe that we are essentially like them. Hence, the “discovery” MacIntyre disapproves is not in the least self-interested: to the contrary, it implies that we cannot give any automatic priority to our own culture's interests or opinions, and that where we differ from others we must be alive to the possibility that it is we who are mistaken. The unsettling possibilities of commonalism are confirmed, e.g., by the fact that many prominent American critics of slavery explicitly grounded their condemnation of this once widely approved social practice upon natural law theory.[33] Indeed, this variant of commonalism has been moral philosophy’s most potent weapon against injustice and oppression. The magnificent rhetorical tones struck by Martin Luther King, Jr.’s Letter From Birmingham Jail rise from its natural law foundation.

Commonalism is a family of theories whose members offer great explanatory and moral advantages. It justifies philosophers’ reliance upon the method of examples in normative theory—a research program that has made substantial progress despite two centuries of derision from utilitarians, noncognitivists and relativists. (For example, Judith Jarvis Thompson’s analysis of moral claims in The Realm of Rights significantly advanced our understanding of our common moral conception.[34])

Secondly, commonalism can undergird a robust natural law theory, giving an objective content to the many legal norms that are couched in moral terms,[35] and founding a perspective from which positive law can be criticized. Commonalism also palliates the greatest fear of American constitutional jurisprudence. It much less smacks of judicial tyranny that life-tenured appointed judges frequently decide cases by means of their moral opinions, if it be supposed that they share a common moral conception with the rest of us.

Thirdly, commonalism fosters virtuous attitudes in those who accept it. Since it presumes us to be equally morally competent, it implies that one can’t lightly dismiss divergent moral opinions. Hence, commonalists will seek to understand the perspective of others and be willing to re-examine their own beliefs. But they will also believe that there are right answers to moral questions that are important to discover. (Indeterminacy—

whether deep disagreement or competition indeterminacy—is a kind of right answer.) If commonalists attend to the implications of their theory, they will be more likely to assess moral issues correctly.

Lastly, far better than relativism, scepticism or elitism, commonalism explains how reflective persons of good will, but of different languages and cultures, can intelligibly debate with one another over moral and political issues, and how they often find substantial ground for agreement. Commonalism also offers a secure moral foundation for human rights and democracy: if all humanity shares a moral conception and so the capacity for attaining moral knowledge, there plainly can be no good moral reason for reserving political advantages and opportunities to only a portion of it.

Commonalism's particular variants deserve to be back in fashion, deserve vigorous defense and debate. But I wonder whether they will again find it. Once fixed, philosophical fashion is hard to fight.

M. B. E. Smith

Smith College

[1]. I here adopt John Rawls' account of "considered judgments," as best explicating the concept of intuition. He characterizes these as judgments made when “the more common excuses and explanations for making a mistake do not obtain,” i.e., when we are confident in our opinion, when we are not emotionally upset, when our interests are not directly affected, and when we have a desire to reach a correct determination. J. Rawls, A Theory of Justice 47-8 (1971).

[2]. For vehement criticism, see, e.g., R. M. Hare, Moral Thinking, 47f., 131-40 (1981).

[3]. Aristotle and Aquinas, among others, were moral faculty theorists. But the theory-type had its heyday in eighteenth century England, when such a prominent philosopher as Joseph Butler could confidently write "It is manifest [that a] great part of common language, and of common behavior over the world, is formed upon supposition of such a moral faculty; whether called conscience, moral reason, moral sense, or divine reason; whether considered as a sentiment of the understanding, or as a perception of the heart; or, which seems the truth, as including both." Dissertation II: Of The Nature of Virtue, in L. Selby-Bigge, British Moralists (1897)(hereinafter BM).

[4]. The most sophisticated presentation of moral rationalism is found in R. Price, Review of the Principal Questions in Morals, in BM, esp. §§ 584-627. Twentieth century moral rationalist theories may be found in H. H. Prichard, Moral Obligation, 7f. (1949), W. D. Ross, The Right and the Good, 29f (1930) and J. J. Thomson, The Realm of Rights, ch. 1 (1990).

[5]. The foremost sentiment theorists were Frances Hutchinson, Adam Smith, and David Hume. C.f., F. Hutchinson, An Inquiry Concerning the Original of Our Ideas of Virtue and Moral Good, in BM, esp. §§ 169-174; A. Smith, The Theory of the Moral Sentiments, in BM, esp. §§ 299-306; D. Hume, Enquiry Concerning the Principles of Morals (1777), esp. Conclusion and Appendix I: Of Moral Sentiment.

[6]. G. Moore, Principia Ethica 10 (1903).

[7]. For exploration and defense of such a theory, see my Ethical Intuitionism and Naturalism: a Reconciliation, 9 Can, J Phil. 609-29 (1979).

[8]. Butler hints at this point in the passage quoted supra in n. 3. See also, Adam Smith, in supra n. 5 at BM § 335.

[9]. For example, as I show in The Duty to Obey the Law, in D. Patterson, ed., Companion to the Philosophy of Law and Legal Theory, 465-74 (1996), Platonic elitists tend to think that ordinary citizens have a prima facie obligation to obey the law, whereas those who employ the method of examples tend to doubt this.

[10]. Such philosophers include Plato, but also the utilitarians, Bentham, Mill and Austin, as well as Nietzsche. (C.f., Plato, The Republic of Plato, 175-235 (trans. F. Cornford 1941); Bentham, Principles of Morals and Legislation, esp. ch. 2, n. 7 (1789); J. Austin, The Province of Jurisprudence Determined, 87-105 (ed. H. Hart 1954); J. S. Mill, Utilitarianism, ch. 1 (1863), Nietzsche, Thoughts on the Philosophy of Morals, Ethical Theories 435, 442-444 (ed. A. Melden, 2d. ed. 1967) Contemporary Platonic elitists include John Rawls, Derek Parfit, Norman Daniels and Richard Hare. (C.f., Rawls, Justice as Fairness: Political Not Metaphysical, 14 Phil.&Pub.Affairs 223 (1985); Parfit, Reasons and Persons 453 (1986); Daniels, Wide Reflective Equilibrium and Theory Acceptance in Ethics, 76 J. Phil. 256 (1979); Hare, supra n. 2 at v, 1-24.)

[11]. Id., supra n. 2 at 40.

[12]. I say "many elitists" because some deny the existence of moral truth or knowledge, e.g., J. Mackie, Ethics: Inventing Right and Wrong 15-49 (1977); and others decline on principle to speculate about the metaphysics of morals, e.g., J. Rawls, supra n. 10. Despite this, they yet regard the proper goal of normative theory as that of devising principles for nonphilosophers to put into practice. However, they supply no answer to the obvious question of why nonphilosophers should take any interest in principles devised by philosophers, if they are also told that they may not regard these as encapsulating moral truth.

[13]. E.g., Hume wrote in the Conclusion, supra n. 5, “The notion of morals implies some sentiment common to all mankind, which recommends the same object to general agreement, and makes every man, or most men, agree in the same opinion or decision concerning it.” See also, J. Butler, supra n. 3 at BM § 244.

[14]. A complication: Some beliefs are “mixed” between the moral and the nonmoral: e.g., the truth conditions for the belief that Sally murdered John comprise the nonmoral proposition that Sally killed John and the moral proposition that her act was wrong. (Similar analyses may be made of “John promised. . .” or “Sally was rude. . .”) But the existence of mixed beliefs does not threaten the distinction between moral and nonmoral belief. Rather, our ability so to analyze them presupposes an ability to distinguish between the ingredients in the blend. I have explored the logic of mixed beliefs in Foot and Hare on Naturalism, 5 Metaphilosophy 187-197 (1974)

[15]. See, e.g., Henry Sidgwick's version: “We cannot judge an action to be right for A and wrong for B, unless we can find in the natures or circumstances of the two some difference which we can regard as a reasonable ground for difference in their duties. If therefore I judge any action to be right for myself, I implicitly judge it to be right for any other person whose nature and circumstances do not differ from my own in some important respects.” Methods of Ethics 209 (7th ed. 1907)

[16]. The extensive literature on act utilitarianism richly shows the power of this last factor. Act utilitarians hold that we are obligated to perform whatever act maximizes utility (which usually is identified as either pleasure or happiness). But virtually everyone agrees that slight gains in utility cannot justify breaking promises. In the face of this fact, act utilitarians waver between insisting that promises should be broken whenever utility is even marginally maximized, and insisting that the disutility of breaking promises must be so great that one must doubt whether utility could ever be so maximized. They therefore vacillate between letting their theory dictate consequences that they recognize to be morally disquieting, and letting it force them into improbable a priori legislation as to what we must believe will be the factual consequences of certain kinds of action. See, e.g., G. Moore, supra n. 6 at 157-64.

[17]. From the perspective of secular commonalism, religious belief can be a kind of moral “wild card.” To the extent that it permits revelatory factual findings (e.g., that God disapproves of this or that practice and will punish any society which tolerates them), it potentially enables the faithful to reach any surface moral opinion which their collective unconscious desires strongly recommend.

[18]. Many true and useful theories similarly rely upon a counterfactual appeal to ideal conditions: e.g., linguistics posits ideal speakers and physics frictionless surfaces.

[19]. I also rely upon—and commend the reader’s attention to—John Mikhail’s defense of moral universals against Richard Posner’s attack upon them, in his Law, Science, and Morality: A Review of Richard Posner’s The Problematics of Moral and Legal Theory, 54 Stanford L.R. 1057, esp. 1088-1110 (2002).

[20]. For example, Massachusetts punishes first degree murder by life imprisonment without possibility of parole. Mass. Gen. L. c. 265 § 1. But attempted murder carries a maximum sentence of ten years in state prison. Id., c. 274 § 6.

[21]. See, e.g., R. Parker, Blame, Punishment and the Role of Result, 21 Am.Phil.Q. 1 (1984).

[22]. See, e.g., J. S. Mill, Utilitarianism, ch. 4, para. 10 (1861). In the course of his notorious “proof” of the principle of utility, Mill identified the question of “whether mankind really do desire nothing for itself but that which is a pleasure to them, or of which the absence is a pain” as being “a question of fact and experience, dependent, like all similar questions, upon evidence.” But he then offered no evidence, but rather an a priori argument: “I believe that these sources of evidence impartially consulted, will declare that desiring a thing and finding it pleasant, aversion to it and finding it painful, are phenomena entirely inseparable, or rather two parts of the same phenomenon—in strictness of language, two modes of naming the same psychological fact; that to think of an object as desirable (unless for the sake of its consequences) and to think of it as pleasant, are one and the same thing; and that to desire anything, expect in proportion as the ideal of it is pleasant, is a physical and metaphysical impossibility.” Great philosopher though he was, Mill was here hopelessly confused.

[23]. Relativism has only in this century become popular, but is now very widely held. See, e.g., G. Harman and J. J. Thomson, Moral Relativism and Moral Objectivity, esp. pp. 1-64 (1996); C. Elgin, The Relativity of Fact and the Objectivity of Value, in M. Krausz ed., Relativism: Interpretation and Confrontation 86 (1989).

[24]. A widespread recognition of convergence might even convert sceptics to commonalism, by the process Hume recommended: “Let a man's insensibility be ever so great, he must often be touched with the images of Right and Wrong; and let his prejudices be ever so obstinate, he must observe that others are susceptible of like impressions. The only way, therefore, of converting an antagonist of this kind is to leave him to himself. For, finding that nobody keeps up the controversy with him, it is probable that he will at last, of himself, from mere weariness come over to the side of common sense and reason.” Hume, supra n. 5, § 1.

[25]. “The pluralism and relativism I favor thus do not lead to the conclusion that anything goes. If many things are right, many more remain wrong.” C. Elgin, supra n. 22 at 98.

[26]. D. Parfit, supra n. 10 at 95.

[27]. Id. at 99.

[28]. Id. at 98-108.

[29]. There is other evidence that Parfit's Common-Sense Morality is not commonly held. Condition 2) implies that our special obligations towards others are unaffected by what third parties will do for them. But we don't believe this is true: e.g., I may have a special obligation to care for my ill parent, but no one would say that I must pay for his medical treatment when his insurer or Medicare will pick up the tab.

[30]. For detailed description of this program, see my Rawls and Intuitionism, Can. J. Phil., Sup. Vol. III: New Essays on Contractarianism, 163-178 (1977). For argument showing its extensional equivalence to Chomskian linguistic methodology as applied to the semantics of our moral vocabulary, see my Ethical Intuitionism and Naturalism: a Reconciliation, supra.

[31]. For difficulties in formulating these principles, see R. Nozick, Moral Complications and Moral Structures, 13 Natural Law Forum 1 (1968).

[32]. A. MacIntyre, Relativism, Power and Philosophy, in Krausz ed., supra n. 22 at 201.

[33]. For references, see R. Cover, Justice Accused: Antislavery and the Judicial Process (1977); R. Fogel, Without Consent or Contract 332 (1989).

[34]. Id, supra n. 4. For argument supporting the above assessment of Thompson’s analysis of claims, see my Review Essay: The Best Intuitionistic Theory Yet!—Thomson on Rights, 11 Crim. Justice Ethics 85-97 (Summer/Fall 1992).

[35]. E.g., Rule 30(b) of the Massachusetts Rules of Criminal Procedure provides that “The trial judge upon motion in writing may grant a new trial at any time if it appears that justice may not have been done.”