ETHICAL SUBJECTIVISM

A subjectivist ethical theory is a theory according to which moral judgments about men or their actions are judgments about the way people react to these men and actions—that is, the way they think or feel about them. It follows that moral predicates are not possessed by actions or actors in the absence of people who pass judgments upon them or who respond to them with such feelings as admiration, love, approval, detestation, hate, or disapproval. It follows from this definition that nonpropositional or noncognitive ethical theories are not subjective, for according to them there are no moral propositions at all, and thus moral judgments cannot be propositions about people's feelings.

This definition is also intended to exclude views according to which moral judgments are judgments about how people behave; hence, views such as that "wrong" means "contrary to the accepted code of the society in which the action is performed" will not count as subjective, even if, as seems likely, statements about moral codes are statements about the prohibitory or permissory behavior of the communities possessing these codes. Although this distinction is not hard and fast (for statements about moral codes might often turn out to be statements about how the people possessing these codes think or feel), we shall maintain it in order not to trespass too far into the subject of ethical relativism, which is treated elsewhere. Elements of subjectivism can be found in so many ethical theories that it is almost impossible to give an account of them. The greatest, though not the most consistent, subjectivist was the eighteenth-century Scottish philosopher David Hume. The theory has also been popular among anthropologists, of whom Edward Westermarck was probably the most outstanding.

Subjectivist theories can provisionally be classified according to whether moral judgments are alleged to be about the speaker's thoughts or feelings, about the thoughts or feelings of some group of people, or about the thoughts or feelings of men as such.

Moral Judgments State What the Speaker Feels

The view that moral judgments are about the feelings of the person making the judgment—that what I mean when I say that an action is right or that a man is good is that the thought of that man or action evokes in me, personally, at this moment, a feeling of approval—has been subjected to an enormous number of objections. It has been argued that, so far from its being possible to identify moral judgments as those judgments that are about feelings of approval, it is in fact only possible to identify feelings of approval as those feelings that are evoked by the judgment that an action is right, and argued that if we feel approval of an action because we judge it to be right, our thinking that it is right cannot be identical with our approving of it. It has also been objected that if the theory is true, all I need do to settle any doubt I have concerning the rectitude of someone else's action is to introspect, and that it is very difficult, if not impossible, to be mistaken about one's feelings, although it is very easy to make a mistake about whether an action is right. More plausibly, it has been alleged that the theory implies that one can only criticize someone else's moral judgment on the ground that the other person is mistaken about how he himself feels.

Some of the worst difficulties for the theory arise from the fact that sentences offered as definitions of moral judgments contain such words as I, now, and here, whose reference depends upon who uses them, at what time, and in what place. From this it follows that one person's moral judgments can never be incompatible with any other person's moral judgments; the sentence "I do not feel disapproval of divorce," when used by one speaker, does not express a judgment incompatible with that expressed by the sentence "I do feel disapproval of divorce," uttered by a different speaker. It would appear, however, that when one person says, "Divorce is wrong," he does mean to say something incompatible with what someone else means when he says, "Divorce is not wrong." From the fact that moral judgments are alleged to have a covert reference to the feelings the speaker now has, it follows that if at one time he judges an action (say, Brutus's assassination of Caesar) to be right, his judgment is not incompatible with the judgment he may make at a later time when he judges Brutus's assassination of Caesar to be wrong. For according to this theory, what he meant on the first occasion was that he did not then feel disapproval of Brutus's assassination of Caesar, and what he meant on the second occasion was that he now does. Clearly there is no reason why both judgments should not be true.

G. E. Moore, in a famous argument, attempted to deduce that the subjectivist theory led to the paradoxical conclusion that the same action could be both right and wrong, and that one and the same action could change from being right to being wrong. (It is important to remember that Moore thought classes of actions—for example, marrying one's deceased wife's sister—could change from being right to being wrong, that is, that an instance of a class of actions, performed at one time, might be right, while another instance of the same class of actions, performed at a later time, might be wrong.) First, Moore argued as follows. If Jones approves of Brutus's assassination of Caesar and says Brutus was right, it follows from the theory that Brutus was right. Similarly, if Smith disapproves of Brutus's assassination of Caesar and says Brutus was wrong, then Brutus was wrong. Hence Brutus was both right and wrong to assassinate Caesar. Second, to show that Brutus's assassination of Caesar can change from being right to being wrong, all Moore thought he need do was to point out that if Jones says (at a time when he approves of Brutus's action) that Brutus was right, then according to the theory, Brutus was right; if he later comes to disapprove of Brutus's action, then, if he says Brutus was wrong, according to the theory, Brutus was wrong. If Jones can truly judge at one time that Brutus was right and at a later time that Brutus was wrong, it must follow that Brutus's action has changed from being right to being wrong.

C. L. Stevenson has criticized Moore's argument in the following manner. Although Jones can truly say that Brutus was right to assassinate Caesar and Smith can truly say that Brutus's action was wrong, neither Jones nor Smith nor anyone else can say that this action is both right and wrong. For anyone to be able to say it is both right and wrong, someone would have both to approve of it and disapprove of it. Hence, although Jones, who approves of Brutus's action, can say it is right, and Smith, who disapproves of it, can say it is wrong, neither can say it is both right and wrong. Moore's mistake, perhaps, consisted in construing the theory we are considering as maintaining that "right" is a predicate like "disapproved of by someone " (from which it would follow that the same action can be both right and wrong), whereas "right" is alleged to be a predicate like "disapproved of by me, the speaker," from which it follows that the same action cannot be both right and wrong, since the speaker cannot both approve and disapprove, on the whole, of one and the same action.

A free exposition of Stevenson's criticism of Moore's argument (that if the view that moral judgments are statements about the speaker's feeling is true, one and the same action can change from being right to being wrong) might take the following form. Moore supposes that if ten years ago I could truly say Brutus was right to assassinate Caesar (because at that time the thought of this action did arouse approval in me) and now I can truly say that Brutus was wrong to do this (because at this time the thought of his action arouses disapproval in me), it follows that the action must have changed from being right to being wrong. Stevenson, however, points out that the statement "Brutus's action has changed from being right to being wrong" is equivalent to the conjunction of statements "Brutus's action was right a while ago" and "Brutus's action is now wrong." Although the truth of the second of these statements is entailed by the fact that I now feel disapproval of Brutus's assassination of Caesar, the first of them is not entailed by the fact that I earlier felt approval of Brutus's assassination of Caesar. Although Moore supposes "Brutus's assassination of Caesar was right" to mean "I once approved of Brutus's assassination of Caesar," what it actually means is "I now approve of Brutus's erstwhile assassination of Caesar," and, ex hypothesi, "I do not now approve of Brutus's action, I disapprove of it." Moore's mistake is to suppose that the word was in the sentence "Brutus was right to assassinate Caesar" shows that the sentence is about my past approval, whereas in fact its function is to show that it is the action I am disapproving of, not my disapproval, that is past.

Although Stevenson's detailed criticisms of Moore are valid, it is possible to restate Moore's arguments in a way that avoids them. To take the second of Moore's arguments first, it would plainly be absurd for me to say that Brutus was wrong to assassinate Caesar and at the same time to say that I was correct many years ago when I judged that Brutus was right to assassinate Caesar. If I now say he was wrong, I am bound to say that when, earlier, I said he was right, I was mistaken. In regard to the first argument, although it does not follow from this subjectivist theory that anyone can say that an action is both right and wrong, it does follow that if Jones says that Brutus was right to assassinate Caesar, he is not saying anything incompatible with what Smith is saying when he says that Brutus was wrong to assassinate Caesar.

Clearly, however, Jones and Smith think they are saying something incompatible, and it is unlikely that they have such a poor understanding of how to use their own language as to be mistaken on this point. In other words, according to the subjectivist theory, if Jones says Brutus was right to assassinate Caesar, Smith can say to Jones that Brutus was wrong and at the same time agree that Jones is making a true statement when he says that Brutus was right. This is absurd. Stevenson has tried to overcome this particular difficulty by saying that although insofar as Jones and Smith are making assertions, their assertions are not incompatible with one another, they are doing something over and above asserting things—namely, Jones is trying to evoke in Smith an attitude of approval toward Brutus's action, and Smith is trying to evoke in Jones an attitude of disapproval toward Brutus's action. Hence, although their beliefs are compatible, their interests clash. Jones aims to achieve something that is incompatible with what Smith aims to achieve. A consideration of this view of Stevenson's, however, would take us away from subjectivism to a consideration of nonpropositional ethical theories.

The difficulties already mentioned may well be fatal to the type of subjectivist theory we are considering, at any rate as long as it is not bolstered by the nonpropositional theory. Moreover, there is a further difficulty that seems to settle the issue. Suppose Jones says that the death penalty for murder ought to be retained in Great Britain, and he says this because he wrongly supposes that abolishing the death penalty would lead to an increase in the number of murders. According to this kind of subjectivism, all Jones means when he says that the death penalty ought to be retained is that the thought of retaining it arouses in him feelings of approval, and since it does do this, his statement that it ought to be retained is true, however mistaken he may be about the facts of the situation. He is under no obligation to withdraw his statement, therefore, when he discovers his mistake. Again, this is absurd.

Moral Judgments State the Speaker's Thoughts

So far we have considered the view that moral judgments are judgments to the effect that the action under judgment arouses certain feelings in the person making the judgment. It is possible, however, to think that what someone making a moral judgment is asserting is that the action or person being judged arouses in the person making the judgment certain thoughts or beliefs. The most natural view is that someone who asserts that an action is wrong is saying that the thought of the action arouses in him personally the belief that it is wrong, or in other words, that all we mean when we say that an action is wrong is that we personally think it is wrong. This view is circular because even though it offers a definition of "wrong" (that is, it maintains that "X is wrong" just means "I think X is wrong"), the word wrong still occurs in the definition (compare "'thoroughbred horse' means 'horse both of whose parents are thoroughbred horses'"). It is obviously impossible to get rid of the circularity by again substituting "thought wrong" for "wrong" in the definition, however many times we do it.

Objective and Subjective Senses of "Right"

It is quite commonly held that whenever one thinks one is acting rightly, one is acting rightly. Philosophers who hold this view, however, are not properly regarded as subjectivists. Clearly, if the word right is being used twice in the same sense when it is asserted that one is doing rightly if one does what one thinks is right, the view is contradictory. According to it, one would be acting rightly even if one did what one mistakenly thought was right: From the fact that one mistakenly thought it was right it follows that it is wrong, and from the fact that one is acting rightly if one does what one thinks is right, it follows that the act is right. Those philosophers who have held this view, however, have generally distinguished two senses of the word right, sometimes called an objective sense and a subjective sense, and have held that an action is right, in the subjective sense, if it is thought to be right in the objective sense. This removes both the contradiction and the suggestion that the property of being right depends on being thought to be right; the property of being subjectively right depends on the different property of being thought to be objectively right.

Moral Judgments State What a Community Feels

Next to be considered is the view that when individuals make moral judgments they are talking not about the way they themselves think or feel about the things they are judging, but of the way some group of people thinks and feels about these things. Presumably the group of people might be named by a proper name—for example, "Englishmen" or "Melanesians," or more plausibly (to avoid the difficulty that Englishmen cannot be supposed to be talking about the feelings of Melanesians and vice versa) by a descriptive phrase such as "my group" or "the community to which I belong." The theory that moral judgments are about the feelings of the speaker's own community is open to a large number of the objections to which the private reaction theory is open and a few more besides. Although two people, one of whom says a given action is right and the other of whom says that the same action is wrong, will be saying incompatible things if they belong to the same community, if they belong to different communities their statements will be perfectly compatible. Again, if a man says at one time that an action is right and at a later time that the same action is wrong, there is no need for him to withdraw his first assertion, provided that the attitude of the community to which he belongs has changed during the interval. In that case there is no reason why both his first assertion and his second assertion should not be true. It follows, too, that however ignorant or mistaken a given community may be concerning the nature of the action being morally assessed, the statement by a member of that community that the action is right will be true as long as his community does approve of it. For example, if a community disapproves of giving eggs to pregnant women because it believes that this will cause them to give birth to chickens, the statement by a member of the community that it is morally wrong to give eggs to pregnant women will be a true one, because this community really does have the feelings it is alleged to have.

Over and above these quite fatal objections, the theory that moral judgments state how members of a community feel about the actions under judgment is exposed to two difficulties, to which the view that they state how the speaker feels is not exposed. Although one might just accept the conclusion, implied by the latter theory, that one discovers a given action is right by introspection (in Hume's language, by looking inside one's breast and discovering there a sentiment of approval or disapproval), it is quite impossible to accept the view that one discovers a given action is right simply by asking other members of one's community whether they approve of it. The theory also leads to the quite unacceptable consequence that anyone who believes, for example, that most of his community approves of retaining a law against homosexuality and at the same time believes that the law against homosexuality ought to be abolished believes two logically incompatible things; for, according to this theory, what he is believing when he believes that the law against homosexuality ought to be abolished is just that most of his community would feel approval of its abolition.

A variant of this last theory would be the view that what we mean when we say that an action is right is that it is approved of by the agent's community, not the speaker's. This is more plausible because it means that instead of condemning actions performed in distant places or times because they do not accord with the moral attitudes of our community, we may praise them because they do accord with the moral attitudes of the community to which the man who performed them belonged. Hence, it appeals to a tendency in some modern moral philosophers to be rather absurdly uncritical of the moral attitudes of communities other than their own. They may be less willing to accept this variant theory, however, when they realize that if it is true, they must be equally uncritical of the morals of their own community. It must be pointed out that according to this theory, if one says, "Introducing racial segregation into University X was wrong," one means that introducing segregation into the university was disapproved of by the community in which the action was performed. It is perfectly obvious, however, that one might perfectly well think that it was wrong and at the same time know that it was not disapproved of by the community in which it was performed. Again, this theory has the difficulty that even when a community disapproves of some practice only because the community is grossly mistaken concerning its nature (as in the case of the women and the eggs), we are still bound to say that the practice is wrong, providing the community does in fact disapprove of it.

Moral Judgments State What Most People Feel

The objections that have already been raised against earlier types of subjectivism can be applied without much difficulty to the view that what we mean when we say an action is right is that most people approve of it. This theory does imply that any two individuals (whoever they are), one of whom condemns an action and the other of whom judges it to be right, really will be saying incompatible things, for it cannot be that most people approve of an action and at the same time disapprove of it. However, since people may change from approving of something to disapproving of it, the theory does entail that a man may judge an action to be right and later judge the same action to be wrong without having to retract his first judgment. The theory means that an action is wrong if most people feel disapproval of it, however ignorant or mistaken they may be about the nature of the action. It also means that it is impossible for a man to make up his mind concerning the rectitude of any action unless he has decided whether humankind in general would approve of it; it is obvious, however, that we can make up our minds on such questions without having the least idea what the attitude of most men would be. In view of what has already been said about the theory that what we mean when we say that an action is right is that the speaker thinks it is right, nothing need be said about the analogous views that an action is right if the speaker's community thinks that it is right, or that an action is right if most people think that it is right.

It is fairly obvious from what has been said that all subjectivist theories need to be amended, at least to exclude the possibility that the attitude of the people we are alleged to be describing when we make moral judgments is not based on ignorance or mistake. Hence, a consideration of subjectivism may lead to the view that an action is right not if it is approved of by any actual person or group of people, but only if it would be approved of by a person of a very special kind—for instance, one who, at the very least, is never ignorant of or mistaken about any relevant matter of fact concerning the action toward which his attitude of approval is directed. Hence, a consideration of subjectivism must inevitably lead to a consideration of ideal observer theories; these, however, are best treated as a variety of ethical objectivism.

See also Emotive Theory of Ethics; Error Theory in Ethics; Ethical Naturalism; Ethical Relativism; Metaethics; Moral Skepticism; Noncognitivism.

Bibliography

Boyd, Richard. "How to Be a Moral Realist." In Essays on Moral Realism, edited by Geoffrey Sayre-McCord, Ithaca: Cornell University Press, 1988.

Brandt, Richard. A Theory of the Good and the Right. Oxford: Clarendon Press, 1979.

Brink, David. Moral Realism and the Foundations of Ethics. Cambridge, U.K.: Cambridge University Press, 1989.

Carritt, E. F. Ethical and Political Thinking. Oxford: Clarendon Press, 1947.

Carritt, E. F. The Theory of Morals. London: Oxford University Press, 1928.

Edwards, Paul. The Logic of Moral Discourse. Glencoe, IL: Free Press, 1955. An acute discussion of various kinds of subjectivism.

Ewing, A. C. The Definition of Good. New York: Macmillan, 1947. Ch. 1.

Firth, Roderick. "Ethical Absolutism and the Ideal Observer," Philosophy and Phenomenological Research 12 (1952): 317–345.

Foot, Philippa. "Does Moral Subjectivism Rest on a Mistake?" Oxford Journal of Legal Studies 15 (1995): 1–14.

Harman, Gilbert. "What Is Moral Relativism?" In Values and Morals, edited by Alvin Goldman and Jaegwon Kim. Dordrecht: Reidel, 1978.

Harman, Gilbert, and Judith Thomson. Moral Relativism and Moral Objectivity. Oxford: Blackwell, 1996.

Hume, David. Inquiry concerning the Principles of Morals, edited by C. W. Hendel. New York: Liberal Arts Press, 1957. Ch. 1, Appendix I. The classical criticism of objectivism.

Hume, David. Treatise of Human Nature, edited by L. A. Selby-Bigge. Oxford: Clarendon Press, 1888; reprinted 1955. Bk. III, Pt. I, Secs. I–II.

Johnston, Mark. "Dispositional Theories of Value." Proceedings of the Aristotelian Society Suppl. (1989): 139–174.

Kant, Immanuel. The Moral Law, or Kant's Groundwork of the Metaphysics of Morals. Translated by H. J. Paton. London: Hutchinson, 1948.

Lewis, David K. "Dispositional Theories of Value." Proceedings of the Aristotelian Society Suppl. (1989): 113–137.

McDowell, John. "Values and Secondary Qualities." In Morality and Objectivity, edited by Ted Honderich. London: Routledge & Kegan Paul, 1985.

Moore, G. E. Ethics. London: Williams and Norgate, 1912. Chs. 3–4. An important criticism of subjectivism.

Plato. The Republic. Translated, with an introduction, by A. D. Lindsay. New York, 1907; paperback ed., 1958. Bk. I.

Price, Richard. A Review of the Principal Questions in Morals, edited, with an introduction, by D. Daiches Raphael. Oxford: Clarendon Press, 1948. Excellent defense of objectivism.

Ross, W. D. The Foundations of Ethics. Oxford: Clarendon Press, 1939. This work and the following are two modern versions of Price's theory.

Ross, W. D. The Right and the Good. Oxford: Clarendon Press, 1930.

Shafer-Landau, Russ. Moral Realism. Oxford: Clarendon, 2003.

Smith, Michael. "Dispositional Theories of Value." Proceedings of the Aristotelian Society Suppl. (1989): 89–111.

Smith, Michael. The Moral Problem. Oxford: Blackwell, 1994.

Stevenson, C. L. "Moore's Arguments against Certain Forms of Ethical Naturalism." In The Philosophy of G. E. Moore, edited by P. A. Schilpp. Evanston, IL: Northwestern University Press, 1942. A reply to some of Moore's criticisms.

Westermarck, Edward. Ethical Relativity. New York: Harcourt Brace, 1932. A well-known version of subjectivism.

Wiggins, David. "A Sensible Subjectivism." In Needs, Values, Truth. Oxford: Basil Blackwell, 1987.

Jonathan Harrison (1967)

Bibliography updated by Mark van Roojen (2005)