William K. Frankena, Ethics, second edition, 1973.

CHAPTER FOUR

Moral Value and Responsibility

We have been a long time considering the central question of normative ethics, namely, that of the basic principles, criteria, or standards by which we are to determine what we morally ought to do, what is morally right or wrong, and what our moral rights are. We saw earlier, however, that there are other moral judgments besides deontic judgments in which we say of actions or kinds of action that they are right, wrong, or obligatory, namely, aretaic judgments in which we say of persons, traits of character, motives, intentions, etc., and also of "deeds," that they are morally good or bad, responsible, blameworthy or praiseworthy, admirable or despicable, heroic or saintly, virtuous or vicious, etc. We must, then, say something about the question on what basis or by what standard we should make such judgments. In other words, we must have a normative theory of moral value to supplement our normative theory of moral obligation, even though we can give relatively little space or time to working one out.

MORAL AND NONMORAL SENSES OF "GOOD"

Moral value (moral goodness and badness) must be distinguished, not only from moral obligatoriness, rightness, and wrongness, but also from nonmoral value. Moral values or things that are morally good must be distinguished from nonmoral values or things that are good in a nonmoral sense. We must, therefore, say a little more than we did in Chapter 1 about the latter distinction. Partly, it is a matter of the difference in the objects that are called good or bad. The sorts of things that may be morally good or bad are persons, groups of persons, traits of character, dispositions, emotions, motives, and intentions -- in short, persons, groups of persons, and elements of personality. All sorts of things, on the other hand, may be nonmorally good or bad, for example: physical objects like cars and paintings; experiences like pleasure, pain, knowledge, and freedom; and forms of government like democracy. It does not make sense to call most of these things morally good or bad, unless we mean that it is morally right or wrong to pursue them. Partly, the distinction between judgments of moral and nonmoral value is also a matter of the difference in the grounds on or reasons for which they are made. When we judge actions or persons to be morally good or bad we always do so because of the motives, intentions, dispositions, or traits of character they manifest. When we make nonmoral judgments it is on very different grounds or reasons, and the grounds or reasons vary from case to case, depending, for example, on whether our judgment is one of intrinsic, instrumental, or aesthetic value.

Of course, the same thing may be both morally good and nonmorally good. Love of fellow man is a morally good disposition or emotion; it is normally also a source of happiness and so is good in a nonmoral sense. But the ground or reason for its being good is different in the two judgments. Consider also the expressions "a good life" and "the good life." We sometimes say of a man that he "had a good life"; we also sometimes say that he "led a good life." In both cases we are saying that.his life was good; but in the second case we are saying that it was morally good, or useful, or virtuous, while in the first we are saying, in effect, that it was happy or satisfying, that is, that it was good but in a nonmoral sense (which, again, is not to say that it was immoral) . It will, therefore, be convenient for our purposes to speak of "the morally good life" on the one hand, and of "the nonmorally good life" on the other. Since the latter expression seems rather odd, I shall hereafter use the phrase "the good life" to mean the nonmorally good life, especially in Chapter 5.

MORALITY AND CULTIVATION OF TRAITS

Our present interest, then, is not in moral principles nor in nonmoral values, but in moral values, in what is morally good or bad. Throughout its history morality has been concerned about the cultivation of certain dispositions, or traits, among which are "character" and such "virtues" (an old-fashioned but still useful term) as honesty, kindness, and conscientiousness. Virtues are dispositions or traits that are not wholly innate; they must all be acquired, at least in part, by teaching and practice, or, perhaps, by grace. They are also traits of "character," rather than traits of "personality" like charm or shyness, and they all involve a tendency to do certain kinds of action in certain kinds of situations, not just to think or feel in certain ways. They are not just abilities or skills, like intelligence or carpentry, which one may have without using.

In fact, it has been suggested that morality is or should be conceived as primarily concerned, not with rules or principles as we have been supposing so far, but with the cultivation of such dispositions or traits of character. Plato and Aristotle seem to conceive of morality in this way, for they talk mainly in terms of virtues and the virtuous, rather than in terms of what is right or obligatory. Hume uses similar terms, although he mixes in some nonrnoral traits like cheerfulness and wit along with moral ones like benevolence and justice. More recently, Leslie Stephen stated the view in these words:

. . . morality is internal. The moral law. . . has to be expressed in the form, "be this," not in the form, "do this," . . . the true moral law says "hate not," instead of "kill not.". . . the only mode of stating the moral law must be as a rule of character.1

ETHICS OF VIRTUE

Those who hold this view are advocating an ethics of virtue or being, in opposition to an ethics of duty, principle, or doing, and we should note here that, although the ethical theories criticized or defended in Chapters 2 and 3 were all stated as kinds of ethics of duty, they could also be recast as kinds of ethics of virtue. The notion of an ethics of virtue is worth looking at here, not only because it has a long history but also because some spokesmen of "the new morality" seem to espouse it. What would an ethics of virtue be like? It would, of course, not take deontic judgments or principles as basic in morality, as we have been doing; instead, it would take as basic aretaic judgments like "That was a courageous deed," "His action was virtuous," or "Courage is a virtue," and it would insist that deontic judgments are either derivative from such aretaic ones or can be dispensed with entirely. Moreover, it would regard aretaic judgments about actions as secondary and as based on aretaic judgments about agents and their motives or traits, as Hume does when he writes:

. . . when we praise any actions, we regard only the motives that produced them. . . . The external performance has no merit. . . . all virtuous actions derive their merit only from virtuous motives.2
For an ethics of virtue, then, what is basic in morality is judgments like "Benevolence is a good motive," "Courage is a virtue," "The morally good man is kind to everyone" or, more simply and less accurately, "Be loving!" -- not judgments or principles about what our duty is or what we ought to do. But, of course, it thinks that its basic instructions will guide us, not only about what to be, but also about what to do.

It looks as if there would be three kinds of ethics of virtue, corresponding to the three kinds of ethics of duty covered earlier. The question to be answered is: What dispositions or traits are moral virtues? Trait-egoism replies that the virtues are the dispositions that are most conducive to one's own good or welfare, or, alternatively, that prudence or a careful concern for one's own good is the cardinal or basic moral virtue, other virtues being derivative from it. Trait-utilitarianism asserts that the virtues are those traits that most promote the general good, or, alternatively, that benevolence is the basic or cardinal moral virtue. These views may be called trait-teleological, but, of course, there are also trait-deontological theories, which will hold that certain traits are morally good or virtuous simply as such, and not just because of the nonmoral value they may have or promote, or, alternatively, that there are other cardinal or basic virtues besides prudence or benevolence, for example, obedience to God, honesty, or justice. If they add that there is only one such cardinal virtue, they are monistic, otherwise pluralistic.

To avoid confusion, it is necessary to notice here that we must distinguish between virtues and principles of duty like "We ought to promote the good" and "We ought to treat people equally." A virtue is not a principle of this kind; it is a disposition, habit, quality, or trait of the person or soul, which an individual either has or seeks to have. Hence, I speak of the principle of beneficence and the virtue of benevolence, since we have two words with which to mark the difference. In the case of justice, we do not have different words, but still we must not confuse the principle equal treatment with the disposition to treat people equally.

On the basis of our earlier discussions, we may assume at this point that views of the first two kinds are unsatisfactory, and that the most adequate ethics of virtue would be one of the third sort, one that would posit two cardinal virtues, namely, benevolence and justice, considered now as dispositions or traits of character rather than as principles of duty. By a set of cardinal virtues is meant a set of virtues such that (1) they cannot be derived from one another and (2) all other moral virtues can be derived from or shown to be forms of them. Plato and other Greeks thought there were four cardinal virtues in this sense: wisdom, courage, temperance, and justice. Christianity is traditionally regarded as having seven cardinal virtues: three "theological" virtues -- faith, hope, and love; and four "human" virtues -- prudence, fortitude, temperance, and justice. This was essentially St. Thomas Aquinas's view; since St. Augustine regarded the last four as forms of love, only the first three were really cardinal for him. However, many moralists, among them Schopenhauer, have taken benevolence and justice to be the cardinal moral virtues, as I would. It seems to me that all of the usual virtues (such as love, courage, temperance, honesty, gratitude, and considerateness), at least insofar as they are moral virtues, can be derived from these two. Insofar as a disposition cannot be derived from benevolence and justice, I should try to argue either that it is not a moral virtue (e.g., I take faith, hope, and wisdom to be religious or intellectual, not moral, virtues) or that it is not a virtue at all.

ON BEING AND DOING: MORALITY OF TRAITS VS. MORALITY OF PRINCIPLES

We may now return to the issue posed by the quotation from Stephen, though we cannot debate it as fully as we should.3 To be or to do, that is the question. Should we construe morality as primarily a following of certain principles or as primarily a cultivation of certain dispositions and traits? Must we choose? It is hard to see how a morality of principles can get off the ground except through the development of dispositions to act in accordance with its principles, else all motivation to act on them must be of an ad hoc kind, either prudential or impulsively altruistic. Moreover, morality can hardly be content with a mere conformity to rules, however willing and self-conscious it may be, unless it has no interest in the spirit of its law but only in the letter. On the other hand, one cannot conceive of traits of character except as including dispositions and tendencies to act in certain ways in certain circumstances. Hating involves being disposed to kill or harm, being just involves tending to do just acts (acts that conform to the principle of justice) when the occasion calls. Again, it is hard to see how we could know what traits to encourage or inculcate if we did not subscribe to principles, for example, to the principle of utility, or to those of benevolence and justice.

I propose therefore that we regard the morality of duty and principles and the morality of virtues or traits of character not as rival kinds of morality between which we must choose, but as two complementary aspects of the same morality. Then, for every principle there will be a morally good trait, often going by the same name, consisting of a disposition or tendency to act according to it; and for every morally good trait there will be a principle defining the kind of action in which it is to express itself. To parody a famous dictum of Kant's, I am inclined to think that principles without traits are impotent and traits without principles are blind.

Even if we adopt this double-aspect conception of morality, in which principles are basic, we may still agree that morality does and must put a premium on being honest, conscientious, and so forth. If its sanctions or sources of motivation are not to be entirely external (for example, the prospect of being praised, blamed, rewarded, or punished by others) or adventitious (for example, a purely instinctive love of others), if it is to have adequate "internal sanctions," as Mill called them, then morality must foster the development of such dispositions and habits as have been mentioned. It could hardly be satisfied with a mere conformity to its principles even if it could provide us with fixed principles of actual duty. For such a conformity might be motivated entirely by extrinsic or nonmoral considerations, and would then be at the mercy of these other considerations. It could not be counted on in a moment of trial. Besides, since morality cannot provide us with fixed principles of actual duty but only with principles of prima facie duty, it cannot be content with the letter of its law, but must foster in us the dispositions that will sustain us in the hour of decision when we are choosing between conflicting principles of prima facie duty or trying to revise our working rules of right and wrong.

There is another reason why we must cultivate certain traits of character in ourselves and others, or why we must be certain sorts of persons. Although morality is concerned that we act in certain ways, it cannot take the hard line of insisting that we act in precisely those ways, even if those ways could be more clearly defined. We cannot praise and blame or apply other sanctions to an agent simply on the ground that he has or has not acted in conformity with certain principles. It would not be right. Through no fault of his own, the agent may not have known all the relevant facts. What action the principles of morality called for in the situation may not have been clear to him, again through no fault of his own, and he may have been honestly mistaken about his duty. Or his doing what he ought to have done might have carried with it an intolerable sacrifice on his part. He may even have been simply incapable of doing it. Morality must therefore recognize various sorts of excuses and extenuating circumstances. All it can really insist on, then, except in certain critical cases, is that we develop and manifest fixed depositions to find out what the right thing is and to do it if possible. In this sense a person must "be this" rather than "do this." But it must be remembered that "being" involves at least trying to "do." Being without doing, like faith without works, is dead.

At least it will be clear from this discussion that an ethics of duty or principles also has an important place for the virtues and must put a premium on their cultivation as a part of moral education and development. The place it has for virtue and/or the virtues is, however, different from that accorded them by an ethics of virtue. Talking in terms of the theory defended in Chapter 3, which was an ethics of duty, we may say that, if we ask for guidance about what to do or not do then the answer is contained, at least primarily, in two deontic principles and their corollaries, namely, the principles of beneficence and equal treatment. Given these two deontic principles, plus the necessary clarity of thought and factual knowledge, we can know what we morally ought to do or not do, except perhaps in cases of conflict between them. We also know that we should cultivate two virtues, a disposition to be beneficial (i.e., benevolence) and a disposition to treat people equally (justice as a trait). But the point of acquiring these virtues is not further guidance or instruction; the function of the virtues in an ethics of duty, is not to tell us what to do but to ensure that we will do it willingly in whatever situations we may face. In an ethics of virtue, on the other hand, the virtues play a dual role -- they must not only move us to do what we do, they must also tell us what to do. To parody Alfred Lord Tennyson:

Theirs not (only) to do or die,
Theirs (also) to reason why.

MORAL IDEALS

This is the place to mention ideals again, which are among what we called the ingredients of morality. One may, perhaps, identify moral ideals with moral principles, but, more properly speaking, moral ideals are ways of being rather than of doing. Having a moral ideal is wanting to be a person of a certain sort, wanting to have a certain trait of character rather than others, for example, moral courage or perfect integrity. That is why the use of .exemplary persons like Socrates, Jesus, or Martin Luther King has been such an important part of moral education and self-development, and it is one of the reasons for the writing and reading of biographies or of novels and epics in which types of moral personality are portrayed, even if they are not all heroes or saints. Often such moral ideals of personality go beyond what can be demanded or regarded as obligatory, belonging among the things to be praised rather than required, except as one may require them of oneself. It should be remembered, however, that not all personal ideals are moral ones. Achilles, Hercules, Napoleon, and Prince Charming may all be taken as ideals, but the ideals they represent are not moral ones, even though they may not be immoral ones either. Some ideals, e.g., those of chivalry, may be partly moral and partly nonmoral. There is every reason why one should pursue nonmoral as well as moral ideals, but there is no good reason for confusing them.

When one has a moral ideal, wanting to be a certain sort of moral person, one has at least some motivation to live in a certain way, but one also has something to guide him in living. Here the idea of an ethics of virtue may have a point. One may, of course, take as one's ideal that of being a good man who always does his duty from a sense of duty, perhaps gladly, and perhaps even going a second mile on occasion. Then one's guidance clearly comes entirely from one's rules and principles of duty. However, one may also have an ideal that goes beyond anything that can be regarded by others or even oneself as strict duty or obligation, a form or style of personal being that may be morally good or virtuous, but is not morally required of one. An ethics of virtue seems to provide for such an aspiration more naturally than an ethics of duty or principle, and perhaps an adequate morality should at least contain a region in which we can follow such an ideal, over and beyond the region in which we are to listen to the call of duty. There certainly should be moral heroes and saints who go beyond the merely good man, if only to serve as an inspiration to others to be better and do more than they' would otherwise be or do. Granted all this, however, it still seems to me that, if one's ideal is truly a moral one, there will be nothing in it that is not covered by the principles of beneficence and justice conceived as principles of what we ought to do in the wider sense referred to earlier.

DISPOSITIONS TO BE CULTIVATED

Are there any other moral virtues to be cultivated besides benevolence and justice? No cardinal ones, of course. In this sense our answer to Socrates' question whether virtue is one or many is that it is two. We saw, however, that the principles of beneficence and equality have corollaries like telling the truth, keeping promises, etc. It follows that character traits like honesty and fidelity are virtues, though subordinate ones, and should be acquired and fostered. There will then be other such virtues corresponding to other corollaries of our main principles. Let us call all of these virtues, cardinal and non-cardinal, first-order moral virtues. Besides first-order virtues like these, there are in other moral virtues that ought also to be cultivated, which are in a way more abstract and general and may be called second-order virtues. Conscientiousness is one such virtue; it is not limited to a certain sector of the moral life, as gratitude and honesty are, but is a virtue covering the whole of the moral life. Moral courage, or courage when moral issues are at stake, is another such second-order virtue; it belongs to all sectors of the moral life. Others that overlap with these are integrity and good-will, understanding good-will in Kant's sense of respect for the moral law.

In view of what was said in a previous chapter, we must list two other second-order traits: a disposition to find out and respect the relevant facts and a disposition to think clearly. These are not just abilities but character traits; one might have the ability to think intelligently without having a disposition to use it. They are therefore virtues, though they are intellectual virtues, not moral ones. Still, though their role is not limited to the moral life, they are necessary to it. More generally speaking, we should cultivate the virtue Plato called wisdom and Aristotle practical wisdom, which they thought of as including all of the intellectual abilities and virtues essential to the moral life.

Still other second-order qualities, which may be abilities rather than virtues, but which must be cultivated for moral living, and so may, perhaps, best be mentioned here, are moral autonomy, the ability to make moral decisions and to revise one's principles if necessary, and the ability to realize vividly, in imagination and feeling, the "inner lives" of others. Of these second-order qualities, the first two have been referred to on occasion and will be again, but something should be said about the last.

If our morality is to be more than a conformity to internalized rules and principles, if it is to include and rest on an understanding of the point of these rules and principles, and certainly if it is to involve being a certain kind of, person and not merely doing certain kinds of things, then we must somehow attain and develop an ability to be aware of others as persons, as important to themselves as we are to ourselves, and to have a lively and sympathetic representation in imagination of their interests and of the effects of our actions on their lives. The need for this is particularly stressed by Josiah Royce and William James. Both men point out how we usually go our own busy and self-concerned ways, with only an external awareness of the presence of others, much as if they were things, and without any realization of their inner and peculiar worlds of personal experience; and both emphasize the need and the possibility of a "higher vision of an inner significance" which pierces this "certain blindness in human beings" and enables us to realize the existence of others in a wholly different way, as we do our own.

What then is thy neighbor? He too is a mass of states, of experiences, thoughts and desires, just as concrete, as thou art. . . . Dost thou believe this? Art thou sure what it means? This is for thee the turning-point of thy whole conduct towards him.4
These are Royce's quaint old-fashioned words. Here are James's more modern ones.
This higher vision of an inner significance in what, until then, we had realized only in the dead external way, often comes over a person suddenly; and, when it does so, it makes an epoch in his history.5
Royce calls this more perfect recognition of our neighbors "the moral insight" and James says that its practical consequence is "the well-known democratic respect for the sacredness of individuality." It is hard to see how either a benevolent (loving) or a just (equalitarian) disposition could come to fruition without it. To quote James again,
We ought, all of us, to realize each other in this intense, pathetic, and important way.6
Doing this is part of what is involved in fully taking the moral point of view.

TWO QUESTIONS

We can now deal with the question, sometimes raised, whether an action is to be judged right or wrong because of its results, because of the principle it exemplifies, or because the motive, intention, or trait of character involved is morally good or bad. The answer, implied in what was said in Chapters 2 and 3, is that an action is to be judged right or wrong by reference to a principle, or set of principles. Even if we say it is right or wrong because of its effects, this means that it is right or wrong by the principle of utility or some other ideological principle. But an act may also be said to be good or bad, praiseworthy or blameworthy, noble or despicable, and so on, and then the moral quality ascribed to it will depend on the agents motive, intention, or disposition in doing it.

Another important question here is: What is moral goodness? When is a person morally good and when are his actions, dispositions, motives, or intentions morally good? Not just when he does what is actually right, for he may do what right from bad motives, in which case he is not morally good, or he may fail to do what is right though sincerely trying to do it, in which case he is not morally bad. Whether he and his actions are morally good or not depends, not on the rightness of what he does or on its consequences, but on his character or motives; so far the statement quoted from Hume is certainly correct. But when are his motives and dispositions morally good? Some answer that a person and his actions are morally good if and only if they are motivated wholly by a sense of duty or a desire to do what is right; the Stoics and Kant sometimes seem to take this extreme view. Others hold that a man and his actions are morally good if and only if thev are motivated primarily by a sense of duty or desire to do what is right, though other motives may be.present too; still others contend, with Aristotle, that they are at any rate not morally good unless they are motivated at least in part by such a sense or desire. A more reasonable view, to my mind, is that a man and his actions are morally good if it is at least true that, whatever his actual motives in acting are, his sense of duty or desire to do the right is so strong in him that it would keep him trying to do his duty anyway.

Actually, I find it hard to believe that no dispositions or motivations are good or virtuous from the moral point of view except those that include a will to do the right as such. It is more plausible to distinguish two kinds of morally good dispositions or traits of character, first, those that are usually called moral virtues and do include a will to do the right, and second, others like purely natural.kindliness or gratefulness, which, while they are nonmoral, are still morality-supporting, since they dispose us to do such actions as morality requires and even to perform deeds, for example, in the case of motherly-love, which are well beyond the call of duty.

It has even been alleged that conscientiousness or moral goodness in the sense of a disposition to act from a sense of duty alone is not a good thing or not a virtue -- that it is more desirable to have people acting from motives like friendship, gratitude, honor, love, and the like, than from a dry or driven sense of obligation. There is something to be said for this view, though it ignores the nobility of great moral courage and of the higher reaches of moral idealism. But even if conscientiousness or gjood will is not the only thing that is unconditionally good, as Kant believed, or the greatest of intrinsically good things, as Ross thought, it is surely a good thing from the moral point of view. For an ethics of duty, at any rate, it must be desirable that people do what is right for its own sake, especially if they do it gladly, as a gymnast may gladly make the right move just because it is right.

MORAL RESPONSIBILITY

There are many other questions and topics that should be taken up here, but we can deal with only one of them -- that of moral, responsibility. In one of our many kinds of moral judgments, we attribute moral responsibility to certain agents. There are at least three kinds of cases in which we do this.

  1. We sometimes say, in recommending X, that he is responsible or is a responsible person, meaning to say something morally favorable about his character.
  2. We also say, where Y is a past action or crime, that X was and is responsible for it.
  3. Finally, we say that X is responsible for Y, where Y is something still to be done, meaning that he has the responsibility for doing it.

Here, saying that X is responsible in the first sense is roughly equivalent to saying that X can be counted on to carry out his responsibilities in the third sense; responsibility of this first sort is another of the second-order virtues we should try to cultivate. To say that X has certain responsibilities in the third sense is simply to say that he has obligations, either because of his office or because of his previous commitments to do certain things, and hence is a straight normative judgment of obligation. The most interesting new problem comes up in connection with ascriptions of responsibility of the second kind. Under what conditions is it correct or right to judge or say that X was responsible for Y?

Saying that X was responsible for Y seems, at first, to be a causal, not a moral, judgment; and one might, therefore, be inclined to say that "X was responsible for Y" simply means "X caused Y," perhaps with the qualification that he did so voluntarily, intentionally, etc. But to say that X was responsible for Y is not merelyfo make a causal statement of a special kind. Neither is it simply a statement that X was able to do Y, as the "ible" ending suggests. Suffixes like "ible" and "able" do not always indicate an ability. They may have a normative meaning. Mill's critics have often criticized his argument that the way to prove something is desirable is to show it is desired, just as the way to prove something is audible is to prove it is heard. They point out, quite rightly, that "desirable" does not mean "can be desired" as "audible" means "can be heard"; rather it means something like "good" or "ought to be desired.'' Similarly, it seems to me, to say that X was responsible for Y is to say something like "I would be right to hold X responsible for Y and to blame or otherwise punish him." Or, perhaps, saying that X was responsible for Y under certain conditions is simply one way of holding him responsible. In the former case, it is a normative judgment; in the latter, it is a kind of act, like making an assignment. In either case, it is not a causal statement even of a special sort. But in either case, we may ask under what conditions it is right to ascribe responsibility to X.

It seems clear at once that one of the conditions required is that X was able to do it and another is that he, in fact, did it (i.e., caused it voluntarily, intentionally, etc.). These are necessary conditions of his being responsive or being held responsible. Are they sufficient?

Aristotle held, in effect, that an individual is responsible for his act if and only |if (1) its cause is internal to him, i.e., he is not compelled to act by someone or something external to him, and (2) his doing it is not a result of any ignorance he has not brought about by his own previous choices. Then and only then can his action be said to be "voluntary." These two conditions are clearly among those necessary for responsibility; we may, in fact, understand them to be included in the second of the conditions just listed. Are there any others? G. E. Moore, P. H. Nowell-Smith, and others have held that a man is not responsible for an action unless he could have done otherwise if he had chosen to do otherwise or if his character and desires had been different. This view is obviously correct; in fact, it is essentially a restatement of Aristotle's position. But it is compatible with determinism, for it insists only that the causes of an action must be internal, not that the action must be uncaused. As far as this view is concerned, a man's choice may be determined by his own beliefs, character, and desires (which, in turn, may be determined by previous causes), and yet be free and responsible.

Many philosophers and theologians have thought, however, that this view is not satisfactory and that a man is not responsible for an action unless he not only could have done otherwise if he had chosen but also could have chosen otherwise. Moreover, they argue, he could have chosen otherwise only if his choice was not simply the result of previous causes such as his beliefs, character, desires, heredity, and environment. In other words, they contend that moral responsibility is incompatible with determinism as it is usually conceived, and that "freedom of a contra-causal kind" is among the conditions of moral responsibility. This was Kant's view, and it has recently been forcefully defended by C. A. Campbell, from whom the words just quoted were taken, and R. M. Chisholm.

On the other side, some determinists have maintained not only that determinism is compatible with moral responsibility, but that moral responsibility presupposes determinism, and that it is really indeterminism that is incompatible with morality.

FREE WILL AND RESPONSIBILITY

It is in this way that the problem of free will and determinism comes upon ethics. Here determinism is the view that every event, including human choices and volitions, is caused by other events and happens as an effect or result of these other events. Indeterminism denies this, and adds that some events, among them human choices and volitions, happen without any cause or explanation. Part of the problem is whether either of these views is true; however, this question belongs to metaphysics and must be left to one side. We can only briefly consider the other part of the problem, namely, whether determinism and indeterminism are incompatible with moral responsibility. The question is not whether X's being free in doing Y is a condition of its being right, wrong, or obligatory. One does not, when he is trying to decide what he ought to do, look to see whether or not he is free. He assumes he is. The question is only whether X's being responsible for Y presupposes his having been free (and if so, free in what sense) in doing Y. But then, according to what was said about the second usage of "responsibility" a little while ago, we are really asking whether it is right to hold X responsible for Y, to praise or blame him and possibly to reward or punish him, if determinism is true or if indeterminism is true. We are asking a question of normative ethics, not, as is usually thought, one of logic or meta-ethics. The question, "Is moral responsibility compatible with determinism (or indeterminism)?" asks not whether determinism (or indeterminism) is logically compatible with responsibility, blame, etc., but whether it is morally compatible with them. It asks whether we are morally justified in ascribing responsibility, in blaming, etc., if we take determinism (or indeterminism) to be true.

Now, although philosophers differ about this, ascribing responsibility, blaming, punishing, and the like, may be regarded as acts we may or may not perform. We say, for instance, "What you did was wrong, but I don't blame you for doing it." But, if they are.acts, then the answer to the question whether it is right to perform them if determinism or if indeterminism is true depends on one's general normative theory of obligation, that is, on one's answer to the question of Chapters 2 and 3. Thus, the determinists who have held that determinism is compatible with moral responsibility have, in their arguments to show this, generally presupposed a teleological theory of obligation, usually a utilitarian one. They argue that it is right to hold people :. responsible, praise them, punish them, and the like, if and only if doing so makes for the greatest balance of good over evil. In other words, like all other actions, such acts as ascribing responsibility, blaming. and punishing are justified by their results, not by anything in the past. If this view is correct, ascribing responsibility, blaming, and punishing may be justified even if determinism is true (some would add only if determinism is true), for it will not matter that the agent being blamed was not free in the contra-causal sense. All that matters is whether praising or blaming him will or will not have certain results.

Deontologists have sometimes been determinists or held moral responsibility to be consistent with determinism, for example Ross. However, most deontologists have denied that morality is compatible with determinism. They deny that ascribing responsibility, praising, blaming, and punishing are made right or wrong wholly by their results, and they insist that it matters whether the agent in question was contra-causally free or not. For, if he was not, then it is wrong to praise or blame him or even to hold him responsible, while, if he was contra-causally free, it is right not only to hold him responsible, but to praise or blame or otherwise punish or reward him. It may even be obligatory to do so. In short, as only a deontologist can do, they take a retributivist rather than a consequentialist view of the justification of such acts as praising, blaming, and punishing, and they infer that determinism is not consistent with morality and its sanctions.

How one answers the question of the bearing of determinism and indeterminism on ethics depends, then, on one's view about how such acts as praising, blaming, and holding responsible are to be justified morally; this in turn depends on one's basic principles of right and wrong. Earlier I proposed as the most adequate normative theory of obligation a form of mixed deontological theory in which the basic principles are those of beneficence and equality of treatment (distributive justice). This theory is compatible with a retributivist view about responsibility and desert, although such a view would require us to add a third principle to the effect that it is at least prima facie right or obligatory as such to apply sanctions to those who have done wrong and to praise or reward certain sorts of right-doing. But, like Socrates and many others, I find such a retributivist theory of the justification of punishment and other sanctions (or retributive justice) quite incredible. It seems to me they are to be justified, if at all, by their educative, reformatory, preventive, or encouraging effects. This view is compatible with the form of deontologism proposed earlier. For according to our theory, as well as on teleological theories, it is possible to hold that the function of holding people responsible and applying sanctions is not retribution but education, reformation, prevention, and encouragement. All we need to add to what the utilitarians say is that their function is to promote equality as well as welfare.

If we take this consequentialist position about the justification of sanctions and ascriptions of responsibility, then, like the teleologists, we can go on to maintain that they may be justified even if there is no such thing as contra-causal freedom. All that is necessary to justify them is that they actually have a capacity and a tendency to have the desired effects on people's future behavior. In short, according to our theory, we may also conclude that determinism is compatible with moral responsibility, as most recent English-speaking moral philosophers have thought.

However, if we accept such a view of the justification of the use of responsibility-ascriptions and moral sanctions, we must make two assumptions. (1) We must assume that people are normally free to do as they choose. If, by nature, they were like ants, bees, or even monkeys, if they had all been thoroughly brainwashed, if they were all neurotically or psychotically compulsive throughout, or if they were all always under a constant dire threat from a totalitarian ruler of the .worst kind, then it would be pointless to try to influence their behavior in the ways that are characteristic of morality (it must be remembered that the threat of punishment is a legal rather than a moral instrument, except in the form of blame and the like). Moral sanctions, internal or external, could not then be expected to have, the desired effects. (2) We must also assume that the choices and actions of people normally have reasons and are reasonably predictable, and are not the result of such wholly chance swervings as were attributed to the atoms by the Epicureans or as are now attributed to sub-atomic particles by some indeterminists who appeal to recent physics in support of their position. Otherwise, again, we should have to regard it as generally pointless to try to influence people by such methods as are part of the moral institution of life -- holding them responsible, blaming or praising them, inculcating a sense of duty in them, setting them examples, reasoning with them, and so on.

The second assumption is clearly compatible with determinism. The. only question is about the first. But a determinist can perfectly well allow that we are often or even normally free to do as we choose, at least if we live in a society that permits us such freedom. That is, he can consistently hold that we are or at least may be free to act and to choose in accordance with our own desires, beliefs, and character. All he is required to insist on is that our beliefs, desires, and traits of character have causes.

It is, however, often argued that the second assumption is incompatible with indeterminism, and that the moral institution of life is therefore inconceivable without determinism. This contention is not entirely convincing. (1) Even if there is some indeterminism in the human sphere, it may still dc possible that there are statistical regularities in human behavior of such a sort that our actions are to some extent predictable and influenceable by such things as moral sanctions. I must confess, however, to a feeling that indeterminism makes things rather too "chancy"; it seems to mean that there is an element of sheer chance in our decisions and this hardly seems to be compatible with our being free to do as we choose. (2) There may be a third alternative besides determinism and indeterminism. Some of those who believe that morality presupposes contra-causal freedom reject both of these opposing theories, for example, Kant, Campbell, and Chisholm. They deny both that our choices are always caused by previous events in accordance with natural laws and also that they are in any way matters of mere chance. Instead, they argue for a special kind of agency; they hold that a self or person is a unique agent capable of a kind of "self-determination" that is not a function of previous causes and yet is not a matter of chance but of. choice, intent, and purpose. Such a view could accept both of the above assumptions, yet reject determinism.

Thus, it is possible to make the two assumptions which are necessary for moral responsibility and not to be a determinist. However, for the reason indicated, I doubt that indeterminism can be regarded as wholly satisfactory. As for the self-determination theory just described, it has not yet, in my opinion, been worked out in any satisfactory way,7 and a discussion of it would involve us in metaphysical questions we cannot consider here. For this reason, it seems best to try to defend the view that determinism is compatible with moral responsibility, and I have elected to do so, although in doing so I do not mean to imply that I regard determinism as true or that self-determinism is false.

There is still another alternative. This is to argue, first, that determinism is true; and second, that it is inconsistent with moral responsibility and possibly with the whole institution of morality. This view has at least been approximated recently by Paul Edwards and John Hospers and by some interpreters of psychoanalysis. If one adopts this position, however, one must be prepared to propose either that the moral institution of life be radically reconstructed or that it be dropped altogether and replaced by something entirely different. Some such drastic proposal may turn out to be correct, but until its two premises have been more conclusively established than they have been, it seems the better part of valor to espouse the position here taken.

There was a young man who said, "Damn!
It grieves me to think that I am
Predestined to move
In a circumscribed groove,
In fact, not a bus, but a tram."
As we saw, however, a determinist may allow that we are normally free to do as we choose, to act in accordance with our own beliefs, desires, and character. Thus, although he may be a fatalist, he need not be one. He may quite consistently regard us more as buses than as trams and more as drivers than as buses. Fatalism does appear to be inconsistent with moral responsibility, but this does not show that determinism is.

It is crucial for our view to hold that in any society with enough social freedom to have a morality, normal human beings are or at least may be free to do as they choose in the sense indicated; we must also hold that our having this sort of freedom is sufficient for the purposes of morality, so that contra-causal freedom is not required. For, if either of these propositions can be shown to be false, it is vain to contend that morality in the form in which we have known it historically is consistent with a non-fatalistic determinism. The first proposition, however, we may regard as plausible enough for present purposes. The only serious doubt that might be cast on it is due to the work of the psychoanalysts and, if I understand them, even they hold that we may be free in the sense in question, at least after we have been successfully psychoanalyzed. The second proposition is harder to be sure about, and much of the debate centers around it. I do not see, however, that it has been shown that morality requires us to be free in a contra-causal sense in addition to being free in the ordinary sense of being free to do as one chooses, free to do Y if one chooses, and free not to do Y if one chooses not to. That the latter is the ordinary sense of "free" is shown by the fact that when I ask you, outside of a philosophical discussion, if you are free or did something freely, you do not look about to see if your decisions are uncaused, but only to see if anything is compelling you or if your actions are an expression of your own desires and character in the light of your own beliefs. So far as I can see, it is morally justifiable to hold people responsible, and to praise and blame them if and insofar as they act freely in this sense, provided, of course, that doing so is in accordance with the principles beneficence and justice.

On this subject, however, as on most others in our province, one must be careful not to be dogmatic. Milton says that after their fall from heaven some of the devils

        reasoned high
Of Providence, Foreknowledge, Will and Fate,
Fixt Fate, free will, foreknowledge absolute,
And found no end, in wandering mazes lost.

Their problem was complicated by certain theological questions we have not raised, but it is always possible that we are as lost as they -- possible, but not necessary, for, if earth has any advantage over hell, we can still hope to find an "end," even if we have not already found one.


Notes

1 The Science of Ethics (New York: G. P. Putnam's Sons, 1882), pp. 155, 158.

2 Treatise of Human Nature, Book III, Part II, opening of Sec. I.

3 For a fuller discussion see my "Prichard and the Ethics of Virtue" Monist (1971), 54, 1-17.

4 The Religious Aspect of Philosophy (New York: Harper & Row, Publishers, 1958), Harper Torchbook edition, pp. 156-57. See selections in Frankena and Granrose, Chap. IV.

5 On Some of Life's Ideals (New York: Holt, Rinehart and Winston, Inc., 1899), p. 20.

6 Life's Ideals, p. 51.

7 But see R. M. Chisholm, Human Freedom and the Self (Lawrence: The University Press of Kansas, 1964). Reprinted in Frankena and Granrose, Chap. IV.