I HAVE suggested that things that are good in the predicative as opposed to the adjunctive sense fall into two classes: (1) those that are good in the sense of being worthy objects of admiration, and (2) those that are good in the sense of being worthy objects of satisfaction. Both of these come, from one point of view, within the scope of ethics; for a thing's being good in either of these ways brings into being a prima facie obligation to produce that thing; we feel under an obligation not only to promote good activities, but also to promote the pleasure and diminish the pain of others. But goods of the second type are not themselves, as such,1 morally good. Nor, again, are all goods of the first type themselves as such morally good; excellent scientific or artistic activity is good but not morally good. I wish now to consider that part of class (1) which is morally good. What we are apt to think of first, when we ask ourselves what kinds of thing are morally good, is certain types of voluntary action, proceeding from certain motives, such as the wish to do one's duty or the wish to diminish the pain of others; and we might be disposed therefore to identify moral goodness with goodness of will. But this would be a mistake. For if we hold that actions are morally good when and because they proceed from certain motives, we can hardly fail to ascribe moral goodness to those same desires when they do not lead to action. They may not lead to action either because the circumstances fail to suggest any action by which we might produce what we desire, or because some other desire is stronger; but in either case the desire itself may be of the same kind and of the same intensity when it is not followed by action as when it is; and if it is what makes the action good when action follows, it is also good when action does not follow. And if we widen our conception of what is morally good to include certain desires, we cannot refuse to include also certain emotions. If desire for another's pleasure is good, so also is satisfaction at his actual pleasure; if desire to relieve another's pain is good, so also is sorrow at his actual pain. In fact satisfaction or dissatisfaction at an existing state of affairs is of exactly the same value, morally, as desire to bring such a state of affairs into existence, or to prevent it from coming to exist. And if we may group desires and satisfactions together under the heading of ‘interests’, interests, no less than actions inspired by interests, may be morally good.
But there is something further that has to be included among the things that are morally good. So far I have spoken of actual felt desires and emotions, or satisfactions. Take now the case of a man who habitually, when he attends to (for instance) the pleasures of other people, takes an interest in them, but who is not at the moment attending to them, either because he is asleep or because he is attending to something else. There is no means of knowing directly how his state differs from that of a man who is habitually indifferent to the pleasures of others; for it is only the physical effects of actions that can be perceived, and only actual desires, emotions, and the like that can be discovered by introspection. But we may feel certain by inference that the state of a man who is habitually unselfish differs somehow from the state of one who is habitually selfish, when both men are asleep or otherwise engaged; for if their state during the period of inattention were exactly alike it would be unintelligible that their behaviour afterwards should be different. It is, indeed, conceivable that the only difference between them, during the interval, should be a difference in the state of their bodies; and if that were so, there would be nothing that is morally good existing through the interval, but only morally neutral conditions which lead to morally unneutral results. But not only our whole moral life, with its accompaniments of repentance and remorse for the past, but even the ordinary facts of memory, are witness to the continued existence of the self through intervals such as those of sleep; and the soul's nature and state in sleep must be just as definite as its nature and state in waking life, though it is in some respects different; for nothing individual that exists can be in any respect indefinite. Answering to the difference that there is between the behaviour of a selfish and that of an unselfish man, it is reasonable to suppose that there is a difference between their characters when they are not behaving at all.
But, it may be said, all that exists when the two men are not behaving is potentialities of behaving, or tendencies to behave. The answer to that is that there is no such thing as potentiality that is not rooted in actuality. That which potentially has the characteristic a can have it only by actually having some other characteristic b. We have, as I have said, no means of knowing what is the actual characteristic that distinguishes the selfish from the unselfish man, when neither is behaving selfishly or unselfishly; we can only say that there must be some difference between the two characters which actually exists, and becomes the cause of their different behaviour when the occasion arises. The man who habitually behaves bravely is in some sense really brave even when he is asleep, or when no occasion for bravery is present; and his bravery has moral goodness when it is dormant no less than when it is being exercised.
We may say, then, that what is morally good is acts of will, desires, and emotions, and finally relatively permanent modifications of character even when these are not being exercised. Some might think that we are coming nearer to what is most truly good as we proceed thus ‘inwards’, from what is perceived to what can only be discovered by introspection, and then to what can only be divined by inference. But that would be quite a mistake. What is perceived by the senses has, indeed, no moral value, for all that the senses perceive is a man's body performing certain movements. But then that was not what we meant by a moral action; a moral action was the setting oneself, from a certain motive, to effect a certain change, and this is as truly inward as anything can be. What can be said, however, is that a character is a larger and grander bearer of moral goodness than any single manifestation of character—whether it be an action, a desire, or an emotion—can be.
But if a character is the grandest bearer of moral value, it is also true that we can build up our conception of an ideal character only by considering first the various elements, the various interests, that would compose it, and by adding that in the ideal character these various interests would be present with intensities proportioned to their goodness. We must begin by asking what are the various interests that are morally good.
It is generally agreed among moralists that action owes its goodness, and the measure of its goodness, to the motive from which it springs. The most noteworthy exception to this is Kant, who maintains that an action is good only when it is done not from a motive, but from a maxim. But this is due to his using the word ‘motive’ in an unusually narrow sense, a sense such that the sense of duty is not reckoned as a motive. The limitation is contrary to the natural meaning of the word, and in the natural meaning of the word Kant is at one with other moralists in saying that moral goodness depends on the motive from which the act is done. Any attempt, then, to decide the measure in which different kinds of action possess moral goodness involves as a preliminary some attempt to state the various kinds of motive from which action can spring.
It has been in the past a widely held view that all action springs from the desire for pleasure, and the first modern philosopher who seriously sought, by an account and classification of motives, to set this view aside, was Bishop Butler. Butler's account was that besides self-love there are two other general motives, benevolence and conscience, and two groups of highly particular motives, ‘terminating upon objects peculiar to themselves’; one of these groups consisting of desires for such things as food, drink, water, and shelter, each such desire tending primarily to the good of the individual, and the other group consisting of desires for such things as esteem, each such desire tending primarily to conduce to the general good; these latter desires are related to general benevolence very much as the first group of particular desires is related to general self-love.2
In its general lines Butler's attack on the description of human nature as being actuated only by selfish motives is thoroughly justified. But his account needs some revision if it is to be made to agree with the facts. Take, for instance, his view that hunger is distinct from self-love, not a desire for pleasure but a desire for food. This is clearly correct in so far as it says that a hungry man is not necessarily a deliberate hedonist, coolly and calmly seeking his own greatest pleasure, or greatest sum of pleasures. Such an account is obviously untrue both of many of the least worthy and of all the most worthy of our actions. It is untrue of the man who ‘sells his birthright for a mess of pottage’, who under the sway of some strong instinctive impulse like hunger or lust does actions which he knows are bound to destroy his prospects of a life of happiness; and it is untrue of the man who acts from a sense of duty regardless of his personal happiness. It is with the former opposition that we are here concerned. It is obviously an inadequate answer to the question what hunger is, to say that it is the desire for food. Desire is always for something not yet existent, but the food exists already. We shall at least have to say that hunger is the desire to eat or to be eating food. But then we may go on to ask what it is about the eating of food that attracts us. Can we not be more definite in our statement? Why is it, really, that we eat our breakfast and our dinner? To a large extent it is true that we eat our meals to make and keep ourselves fit for the reaching of some ulterior end, whether that be the doing of our duty or the attainment of success, or whatever it be. But if we ask ourselves what are the more immediate reasons that make the eating of food attractive, it seems to me that we are left with only two. There is, on the one hand—and this is what is dominant in any case of extreme hunger—the desire to get rid of a present gnawing discomfort; and there is, on the other hand—and this is dominant when we get our meals at their accustomed times and in sufficient plenty—the desire for the sensuous pleasures of taste. Suppose that a man were anaesthetized, so that he felt none of the discomfort of hunger, and that he were conscious of chewing and swallowing but felt no pleasure in these processes. It seems to me clear that in such a case the eating of food would not attract him at all, except for one of the ulterior causes which I have mentioned only to set them aside as irrelevant to the question: What is the intrinsic motive for eating? It would seem then that our desire for food is a desire to eat food (1) as freeing us from a certain pain, and (2) as giving us a certain pleasure, the one element or the other predominating as the hunger is more or less acute. And if so, the hedonist will be entitled to reckon the desire for food as an illustration of his general thesis. But it is an instance of self-love not in the sense of desire for pleasure in general, but of desire for a particular pleasure or for relief from a particular pain, or for a combination of the two.
In principle, this account seems to cover a great part of the life of most human beings—that it is a search, not for pleasure in the abstract, but for particular pleasures. It is often thought that hedonism can be refuted by urging that it erroneously concludes, from the fact that pleasure is felt in anticipating our action or its results, or again from the fact that pleasure normally accompanies the fulfilment of desire, that pleasure is the object of desire. And that is a true criticism of the arguments offered by some hedonists. But it should not lead us to overstate the case and say that it is normally just certain activities and not certain pleasures that are the object of desire. Would men seek riches if it were not for the pleasure they have experienced in the past, and hope to have again, from having riches at their command? Would they seek fame if they had not experienced the pleasure of hearing men speak well of them and were not looking forward to experiencing it again? Would they play games if they had not enjoyed the thrill of the successful control of their muscles and of the triumph over their adversary?
I start, then, with desires for particular pleasures, as being probably the commonest of all the types of desire. Secondly, out of these desires there arises in some people, and actuates them in some of their actions, a desire for their own pleasure on the whole. In so far as people are actuated by this, they become capable of giving up some particular pleasure towards which they are strongly attracted, because they think it will interfere with their attainment of the greatest amount of pleasure on the whole. Both this type of life and that previously described are selfish lives, but the former—to use the language I have used previously3—is a suggestible and the latter a planned selfishness. The latter is a sort of rationalization of the former.
Thirdly, there are desires for some particular good activity, or for the attainment of some particular virtue, or knowledge, or skill. These desires are closely bound up with desires for particular pleasures. For in general the exercise of any good activity, and even the possession of technical skill, is a pleasant thing, and is known to be such, and we can hardly be desiring the good activity or the skill without desiring the pleasure that accompanies them. Yet we can at least distinguish between the two desires as two distinct elements in our total mental state, and can say that in some cases the one desire and in others the other predominates. The skilled workman desires both to do his job well and to have the pleasure of doing it well, but one workman will be thinking more of the one and another more of the other. And some of our desires for pleasure are unaccompanied by any thought that the activities on which the pleasures supervene are good. This is true, for instance, of desires for such pleasures as those of eating and drinking, in contrast with those of virtuous or scientific or artistic activity.
Fourthly, in some people there arises out of these desires for particular forms of perfection or good activity a generalized wish for good activity. This is the motive which Aristotle describes as dominating the good man, and it is also the motive in what T. H. Green describes as the life of self-realization. In Green's account self-satisfaction, which is a particular form of pleasure, sometimes seems to predominate over self-realization; I think we may take it, however, that at bottom self-realization and not self-satisfaction is Green's ideal.
Fifthly, there are desires that particular people other than oneself should have particular pleasures. It is often difficult for an observer to know whether an act of apparent benevolence to another person proceeds from this motive, or from the wish to engage in the good activity of conferring pleasure on another, or again from the wish to have the pleasure of conferring it. But from time to time the difference between the other-regarding motive and these two self-regarding ones betrays itself even to an observer. For the person whose motive is the other-regarding one will sometimes rather stand aside and let a greater benefit be conferred on the object of his love by some one else than confer a lesser benefit himself; while one whose motive is either of the self-regarding ones will behave in the opposite way. And apart from such cases I think it is possible by introspection to distinguish the three motives, and after, when two of them or all three are present, to say which is the predominant one.
It is noteworthy that such desire for the pleasure of an individual may coexist with almost complete selfishness towards others. A mother who is capable of the greatest self-sacrifice to spare her child any pain may be at the same time quite callous to the pain of children not her own. And further, such restricted altruism is, I suppose, always accompanied by some egoism. A mother desires not only the happiness of her child, but also the happiness she herself will get from seeing her child happy and from her child's companionship and affection. Nevertheless, it is a mistake to describe the motive of her action as égoïsme à deux. There may be much egoism in it, but one, and often the predominant, element in it is an altruism, very restricted in its scope, but very strong and in itself quite disinterested. And the same is to be said of other restricted altruisms.
Sixthly, in some people there supervenes on this restricted altruism a generalized altruism in which the pleasure or happiness of all human beings, or even of all sentient beings, becomes an object of desire.
Seventhly, there are desires for the exercise of good activities by, or the improvement of character or intellect in, some particular person or persons other than the desirer. Clearly this is an additional component in the total attitude of most parents towards their children, and of many men towards their fellow countrymen. And eighthly, there is a generalized form of this, which is the desire for the perfection of all human beings. And each of these can obviously become a motive to action.
Ninthly, it seems to me that we must recognize as a distinct motive the desire that some one else should suffer. It might be suggested that what is at work here is the desire to have the pleasure of making him suffer, or of seeing him suffer. But that would be putting the cart before the horse; we should not anticipate pleasure from making a man suffer or from seeing him suffer, unless we first desired him to suffer.
Tenthly, there is at least possible a generalized desire that every one except oneself should suffer. But, to the credit of human nature be it said, it is far more doubtful whether such a desire ever really exists than it is that a generalized desire for the happiness of other people exists.
Eleventh, there is the wish to make another person's character worse in some respect. This is not a common motive; for in most cases, where a man seems to an observer to be setting himself to corrupt the character of another man for the sake of doing so, he is really not attracted by the thought of the other person's becoming worse, but is using the corruption of his victim's character as a means to his own pleasure or his own gain. Yet it seems to me difficult to deny that in some cases there is a real wish to corrupt the character of another, and that this is sometimes at least a component in the motive to action.
Twelfth, we can conceive of a generalized form of the last-named motive, in which the agent wishes all other men to be as bad as possible, and to be made so by his agency so far as he can make it effective. But it may be doubted if this motive has ever operated in a human heart. It would be the motive not of a man but of a devil.
So far I have spoken of motivation in which there is no thought of claims or of duties. But we must now take account of the wish to fulfil some particular claim thought of as morally binding. And finally there is the generalized form of this, in which the wish is, not to do that which is the fulfilment of a particular claim (or which is prima facie obligatory), but to do the act which is the maximum fulfilment of claims and is in the strict sense obligatory.
I have not thought it necessary to offer any proof of the existence of these several motives. I think I have found them all (with the exception of the tenth, eleventh, and twelfth) at work in myself, and I venture to think that, with these same exceptions, they all exist from time to time in many people's minds—though some of them perhaps only very occasionally. The only ground on which I think objection would be likely to be made would be the assumption of psychological Hedonism, or of the more general view of which psychological Hedonism is one form, that only states or activities of the de-sirer himself can be the object of desire; and that is a belief which rests on confusions which have often been pointed out, and will not stand the test of a scrutiny into the motives from which we actually act.
Perhaps the main ground on which it might be urged that only imagined future states of the desirer can be desired is this: If what is desired, it might be said, were a state of any one other than the desirer, then the desire should be satisfied by the mere coming into existence of the desired state; but in fact no one's desire is satisfied merely by the coming into existence of a state of some one else; the desirer must come to know or think that it has come into existence; and indeed the desire will be satisfied if the other person's state does not come into existence at all, but the desirer merely thinks that it has. Therefore, it might be urged, what is desired is not the other person's state, but the desirer's own state of confidence that the other person's state has come into existence. What a mother desires, it might be said, is not that her child should be happy, but that she should know or think it to be happy; and that is proved by the fact that she will be satisfied if she thinks it is happy, even if it is not happy in fact, and will not be satisfied if it is happy but she does not know or think that it is.
This argument rests on a confusion between the fulfilment of desire and the satisfaction of the desirer. The fulfilment of desire is simply the coming into existence of that which is desired; the satisfaction is a new mental experience in the mind of the desirer. The latter naturally does not arise unless the desirer knows or thinks that the desire has been fulfilled, whether or not in fact it has been fulfilled; it naturally arises if the desirer thinks with confidence that the desire has been fulfilled, whether or not it has. The fact that the satisfaction of the desirer depends not on the occurrence of the external event but on the desirer's opinion about it has no tendency to show that what was desired was a state of the desirer's own mind. In fact, there is no general ground on which we can rule out any imaginable state of affairs from being desired; we can only attempt to discover, by reflection on our own desires and by inference from the behaviour of other people round us, and from the facts of history, what types of imagined states of affairs in fact are desired; and this is what I have been trying to do.
It will be noticed that throughout this catalogue of motives I have distinguished a more particularized, instinctive form and a more generalized and rationalized form. In some cases one or more intermediate forms might be recognized. For instance, there may be a man in whom conscientiousness with regard to the fulfilment of all promises is strong, but conscientiousness with regard to benefiting others weak. Such a man will have a generalized sense of duty to do, out of the alternative acts which would be fulfilments of particular promises, the act which would be the maximum fulfilment of promise; but he has not reached the stage of wishing to do that which is the maximum fulfilment of all obligations, including those of beneficence as well as those of promise-fulfilment. And similarly intermediate forms might be interpolated between the particularized and the generalized motives which I have distinguished in other cases. But it would be tedious to attempt what one could certainly not complete, a minute account of all the possible intermediate forms.
It is, in general, possible to range these motives in order of excellence. We may leave out of account the generalized wish to cause pain and the generalized wish to cause moral evil, as falling below the level of human nature. Of the other motives I have mentioned, we must surely rank lowest the wish to produce moral evil in some other person. Just as we saw4 good activities to be good in a fuller sense than pleasure, being worthy objects not only of satisfaction but also of admiration, moral evil is bad in a fuller sense than pain, being a worthy object not only of dissatisfaction but also of condemnation; as is evidenced by the fact that a good man will fear it more, whether in himself or in others who are dear to him. If this is so, it is natural that the wish to produce moral evil is a worse motive than the wish to produce pain. The wish to produce pain comes, however, next to it in the scale of demerit.
The complete generalization of this wish, the wish to produce as much pain as possible for all human beings, probably does not exist as a human motive. But the partial generalization of the wish to produce pain, the wish to produce and to go on producing a maximum of pain for some individual, the wish involved in hatred, is plainly worse than the wish to produce a particular temporary pain, the wish involved in anger.
When we come to consider whether any moral value attaches to the wish to procure some particular pleasure for oneself, the question is rather complicated and difficult. I will set down what appears to me to be true about it. Pleasures themselves may be divided into three classes—those which are marks of a good nature, and themselves morally good, such as the pleasure of helping another; those which are morally indifferent, such as the sensuous pleasures; and those which are marks of a bad nature, and themselves morally bad, such as the pleasure of hurting another. The desire to get an indifferent pleasure is itself indifferent. The desire to get a morally good pleasure, as being morally good, is itself morally good; the desire to get it, as being a pleasure, is morally indifferent. The desire to get a morally bad pleasure, as being morally bad (if such a desire exists, which is doubtful), is morally bad; the desire to get it, as being a pleasure, seems morally neutral, though the accompanying indifference to the badness of the pleasure is morally bad.
The generalized wish to get a maximum of pleasure for oneself is also morally neutral (though the indifference which may accompany it as to whether the pleasures to be got are or are not the pleasures of engaging in bad activities, is morally bad). Prudence, the tendency to act on such a wish, is therefore not a virtue, but only a characteristic useful to its possessor.
The wish to promote some good activity, or some improvement of character or of intellect, in another, appears to be as certainly better than the wish to produce pleasure for another, as the wish to corrupt a character is worse than the wish to produce pain. And again, the generalized wish to promote the moral and intellectual improvement of all human beings is better than the wish to produce pleasure for them, as the wish to promote their deterioration, if such a wish existed, would be worse than the wish to make them all suffer pain.
When we compare the wish to promote perfection or good activity in another person with the wish to achieve it for oneself, I can find no ground for regarding either as better or less good than the other. Suppose I wish to bring about in myself some good activity, moral or scientific or artistic, or some moral or intellectual improvement which will lead to such activities; and abstract from any thought of the pleasure or credit or gain I can get by such activity or such improvement. What is left is an attraction towards a certain activity or change of character or of intellectual state as being good, and this is seen to be, not the same thing as, but of the same moral worth as, the wish to produce a similar activity or change in another person. I may, owing to particular circumstances, be wishing for the one when I am not wishing for the other, just as owing to particular circumstances I may be desiring the moral improvement of my children when I am not desiring the moral improvement of any one else, and may on another occasion be desiring the moral improvement of a pupil when I am not even thinking of that of my children. But in all three cases we are desiring something to come into being because it is good, and all three desires seem therefore to be of the same moral worth.
Finally, we must compare the desire to do one's duty, both in its particularized and in its generalized form, with all the other motives I have named. It seems to me clear that in either form it ranks above all other motives. For suppose that a person is attracted towards one act as being the fulfilment of a moral claim, and to another act without having this thought about it. Suppose, for instance, that he thinks of a certain use of his money as being the fulfilment of a moral claim which a creditor has on him, and is at the same time attracted towards bestowing it in charity. So long as he thinks of one act as being an act he ought to do, and of the other not as being something he ought to do, we are bound to say that he will be acting better in doing what he thinks he ought than in doing what he does not think of as something that he ought to do. It is only if he thinks of the possible object of his charity as himself having a moral claim on him, that he can be acting better in bestowing the charity than in paying the debt; and then we are no longer contrasting action from the sense of duty with action from a different motive, but action from the sense of one prima facie obligation with action from the sense of another. Or again, suppose he is attracted towards some scientific activity by the thought that it is a fine activity of the human spirit, and towards some philanthropic activity by the thought that he ought to engage in it; and suppose that he cannot do both things; we should say he was acting better in doing what he thought he ought than in doing the alternative action; we should say that only if the scientific activity also presented itself as a prima facie obligation (as it well might) could he possibly be acting morally better in preferring it to the alternative.
It might be suggested that that argument is not conclusive—that, though action from the sense of duty is better than action from any other motive when the two motives conflict, action is still better when it proceeds from the motive of love without the thought of duty occurring at all. But I do not think this can seriously be maintained. The motive of love which we now are supposing to arise unaccompanied by any thought of duty is the same in kind with the love which in the case of conflict of motives we judged to be inferior to the sense of duty; and the imagined though non-existent sense of duty which according to this suggestion is inferior to love is of the same kind as the sense of duty which we judged to be higher than love when the two conflicted; so that if the sense of duty was the better when the two conflicted, it would still be the better if it existed in the case in which it actually does not.
It is not as if the sense of duty could fairly be described as a hard impersonal devotion to an abstract principle in contrast with the warm outflow of love towards another person. In its typical manifestation, the sense of duty is a particularly keen sensitiveness to the rights and interests of other people, coupled with a determination to do what is fair as between them; and it is by no means the case that it tends to be divorced from warm personal feeling; it tends rather to be something superadded to that.
But, finally, the desire to do the act which is genuinely obligatory is better than the desire to do the act which is the fulfilment of a particular prima facie obligation. Suppose a man thinks act A to be prima facie obligatory in some respect, but act B to be actually his duty; he is obviously acting better in doing the latter act than in doing the former.
We seem to have been able to establish an order of worth among the various motives from which human action flows. But it is clear that action often flows from a combination of motives. (In such a case we may call the combination of motives the resultant motive, and the simple motives the component motives.) What are we to say of the worth of such combinations of motives? Suppose we agree that motive A is better than motive B. Is an action from motives A and B better than, or worse than, or morally equal to, one done from the better motive A? It was Kant's view, and it is probably often held, that the addition of a lower to a higher motive always involves that the action has less moral worth. Let us suppose that M does an act from the better motive A simply, and that N does it from a combination of motives A and B. There appear to be two quite different possibilities, not distinguished by Kant. The strength of motive A in N may not be great enough, without the co-operation of the other motive, to induce him to do the act, while in M it is ex hypothesi strong enough by itself to induce him to do the similar act. Then we should have to say, with Kant, that M's act is better than N's. But there is another possibility—that motive A, love of duty for instance, is equally strong in both, and that in N motive B only serves as an additional but not necessary inducement to do the act. Then we must say that the additional presence in N of motive B makes his action better than M's if motive B is itself a good one (e.g. desire to produce pleasure for another), and leaves it equally good with M's if motive B is a morally indifferent one (i.e. desire to get pleasure for himself). It will only be if the additional motive B is a positively bad one that we shall think N's action less good than M's.
There is another doctrine of Kant's, quite distinct from that just mentioned, but co-operating with it in producing his very rigoristic moral view. This is the doctrine that no motive other than sense of duty has any moral value at all, that desire to produce pleasure for another, for instance, is no better than desire to produce pain for another. This might be justified if we could regard action from any desire as simply flowing from heredity and environment, and action from sense of duty as a perfectly free undetermined action for which alone we could give the agent credit, since in it he springs quite clear of the influence of heredity and environment. But unless we can maintain this extreme libertarian position, we need not agree with Kant's denial of moral value to all desires. And plainly great violence is done to what we really think, when we are asked to believe that ordinary kindness when not dictated by the sense of duty is no better than cruelty.
Kant's picture of the ideally good man as going through life never animated by natural kindness but only by the sense of duty has always been felt by most readers to be unduly narrow and rigoristic, and if I am right, it rests on two mistakes. If we avoid these mistakes, we can think of the ideally good man as having many good motives in addition to the sense of duty, but with a sense of duty strong enough to induce him to do his duty even if the other motives were absent.
So far I have spoken of the goodness of motives. I have still to ask whether the goodness of action depends entirely on the goodness of its motives. It is plain enough that the two are connected. If a man exerted himself to bring about the very same changes which an ideally good man in the same circumstances would set himself to bring about, but in doing so was actuated only by the thought that in doing so he would be acquiring credit for himself, no one would assign any moral goodness to his action. And if he did so, actuated only by the thought that in doing so he would be hurting some one else, we should call his action morally bad. Yet to say that the goodness of actions depends solely on the goodness of their motives would be to simplify matters far too much. Suppose A does an action whereby he thinks he will produce pleasure, for instance, for B, and pain for C and D, and does it attracted only by the desire to produce pleasure for B, with comparative indifference to the pains of C and D, his motive is purely good but his action is not purely good. No bad motive has been at work, but a good motive, the wish to spare pain to C and D, has not been at work or has not been at work as strongly as it would have been in an ideal character. The way in which we judge of the goodness of an action is, I think, somewhat as follows. If A does an act which he foresees to be likely to have certain characteristics, we ask ourselves what attractions an ideally good man would have towards the act in virtue of certain of its characteristics, and what aversions he would have in virtue of others. We judge, perhaps, that an ideally good man would be more deterred from the act because it would hurt C and D than he would be attracted towards it because it would give pleasure to B; and we judge A's action bad on the whole not because of its actual motive, which is good, but because in doing it A is failing to have a strong aversion which an ideally good man would have. We judge the action by comparing the agent's set of attractions and aversions with the set of attractions and aversions which would ideally arise in face of the foreseen changes to be produced by the action.
It is easy to illustrate the point. Suppose that A out of nepotism bestows a job on B, in whom he is interested, ignoring the much stronger claims of C, D, &c, but wishing them no ill. His motive is good, so far as it goes, though it does not rank very high in the list of motives. But his action is definitely bad, because he is not being deterred as an ideally good man would be by the thought of the injustice to C, D, and the rest.
Even action done from a sense of duty may for this reason fail to have moral goodness, may perhaps even be morally bad. The nepotist may act from the thought that he ought to bestow the job on B; and so far there is an element of goodness in his action. But in so far as he is failing to be influenced by the thought that he ought to do justice to C, D, and the rest, his action is a bad one; and it is easy to imagine a case in which the prima facie obligations he is failing to be influenced by are much more weighty than those he is being influenced by; and in such a case his action will be on the whole positively bad.
An alternative way of considering the matter should, however, be considered. It may be said that such an action, in which the effective motive is a good one, and the agent merely fails to be affected by morally more weighty considerations, is not bad, but merely of a low degree of goodness. But I think this answer can be seen not to agree with what we really think. Imagine a man who is never influenced by the wish to harm any one, but who is completely selfish, never acting even from a narrow sense of duty, nor from a wish to make any one other than himself better or happier, regardless both of the rights and of the interests of every one else. We should not hesitate to call such a man a bad man. We might feel more certain of his badness than of the badness of a man who sometimes acted from the desire to hurt another, but also sometimes from the desire to help another. Yet on the view we are now considering none of his actions would be bad, since none of them would proceed from a positively bad motive, but all from the neutral motive of desire to further his own interests or his own pleasure. We should have a very bad man who never did a bad deed. But a bad character is a character from which bad acts tend to flow, and if we are clear that the character is bad, we ought to be clear that the actions which are the typical manifestations of the character are bad. We should be clear, then, that the lack of good motivation as well as the presence of bad motivation may make an action bad.
If this argument be correct, we are now in a position to see that rightness and goodness do not fall so much apart as we should think them to do if we held that goodness depends entirely on the motive present, while rightness depends not at all on motive, but on intention, or, more strictly, on the nature of that which we set ourselves to do. For an action will be completely good only if it manifests the whole range of motivation by which an ideally good man would be affected in the circumstances, a sensitiveness to every result for good or for evil that the act is foreseen as likely to have, as well as to any special prima facie obligations or disobligations that may be involved; and only if it manifests sensitiveness to all these considerations in their right proportions. But if the agent is responsive to all the morally relevant considerations in their right proportions, he will in fact do the right act. Thus no action will have the utmost moral excellence which an action in the circumstances can have, unless it is also the right action.
But if we have shown that in its limiting case a morally good action must be the right action in the circumstances, we have still left moral goodness and rightness in some very important ways independent. To begin with, a right act need not be a completely or even a partially good act. Take a case in which we should have no doubt what is the right act to do. Suppose that A owes money to another man who in addition to being his creditor is very poor and very deserving. We should agree that he ought to pay the debt, that that is the right act. But suppose that the debtor is a candidate for Parliament and the creditor one of his constituents. It might easily happen then that the debtor paid the debt, i.e. did the right act, merely to escape discredit in his constituency, and in such a case the act would be morally quite neutral. Again, he might conceivably pay the debt to encourage the creditor to some extravagance which he could not afford. Then he would be doing the right act but his action would be morally bad.
Again, an act may have a high degree of moral goodness and yet be entirely different from the right act. A man may be alive to almost all the morally significant features of alternative acts, but may (from prejudice against some individual, for instance, or from lack of imagination) fail to be attracted by just that feature of one of the alternative acts which to a person of ideal moral goodness would be the decisive feature; and in such a case he will do an act completely different from the right act.
Yet, as might be expected, goodness of character is the only condition that with even the slightest degree of probability tends to make for the doing of right acts. If a man is not morally good, it is only by the merest accident that he ever does what he ought. The act to which he is attracted by one feature of it, itself morally indifferent or bad, may be the act towards which a good man would be attracted by its whole system of morally significant features, but if it is so, the coincidence is accidental. Thus a theory which insists on the difference and mutual independence of rightness and goodness is by no means precluded from recognizing those connexions between the two which are well known to common sense.