You are here

Chapter X. Three Theories of Goodness

1. We have held that the fundamental moral judgment was the judgment of good, not of right. In discussing the emotivists and their linguistic successors, we found reason to think that this judgment was irreducible, in the sense that it could not be resolved away into the expression of an emotion, a command, or any other non-cognitive attitude.

But if it is a judgment, what does it assert? Presumably a character of some kind belonging to the subject that is called good. What sort of character? Some moralists of eminence have thought it a character like yellow, which could belong to its subject even if there was nothing else in the world; others, have held that it is a relational property like loved-by-Smith, which could belong to something only as this was related to something else. In this chapter we shall examine the most influential view of the first type, and then the two most plausible views—themselves widely different—of the second type.

There is no doubt that the most influential advocate in recent times of the first position is G. E. Moore.1 ‘Good’ was for him the name of a simple ‘non-natural’ quality present in everything that is good intrinsically. Of course what he is considering is not instrumental goodness, the value of something as a means, but the intrinsic goodness of that which is good in itself or good for its own sake. There are many kinds of things that we regard as thus intrinsically good. There is the good man, the good picture, the good holiday, the good dessert, the good walk, the good scientific theory. Do we mean the same thing by ‘good’ in all these cases? The assumption that we do merely on the ground that we are using the same word is not quite safe. Sometimes when we use the same word there is no common element at all, as when we speak of the pile of a carpet and a pile of stones; sometimes the common element could be ferreted out only by the help of analogy, as when we speak of a black night and a black crime, or a weighty pendulum and a weighty argument. Nevertheless Moore did think that whenever we used ‘good’ of intrinsic goods we were using it in exactly the same sense, that is, were referring to exactly the same quality in all these various subjects. The goodness of a good man and of a good dessert are based, it is true, on very different characters, but the goodness itself is identical in each.

How are we to define this identical quality? That, says Moore, is impossible. The reason is not that it is so bafflingly complex, but on the contrary, that it is so bafflingly simple. If we are to define anything in logical fashion, it must be at least complex enough to enable us to single out within it a genus and species, as when we say a square is a plane figure that has four equal sides and angles. But try that process on goodness. You can start by saying ‘goodness is that quality which—’ but how are you to go on? You can only say ‘goodness is that quality which is—goodness’; there is no way of describing it except by itself; it is unique. Not that this sort of uniqueness is anything unusual or queer; it is possessed by innumerable other qualities such as sour and cold and painful and yellow. These are all perfectly familiar; yet they are indefinable because they are unanalysable, and unanalysable because they are simple.

2. The great danger here, Moore thinks, is that of confusing goodness itself with the qualities in virtue of which we call something good. It is a significant fact that goodness is always based on these further qualities as they themselves are not based on anything further. If someone were to remark that the sky was blue, we should think it absurd to ask, ‘In virtue of what do you call it blue?’ for one sees it to be blue directly. But if he were to remark that a man or a holiday was good we might quite naturally ask what it was in the man or the holiday that made him think so, and he might as naturally answer, ‘Because the man has such an unfailing sense of duty’ or ‘Because the holiday was so pleasant’. Good, then, is a peculiar kind of quality. It is not grasped directly like blue; it is always arrived at indirectly by means of other qualities. Hence it does not belong to its subject as these other qualities do. They are part of its nature in a sense in which it is not. A psychologist describing a man's characteristics would note in his description a strong devotion to duty, but he would never include as an item alongside it the goodness of this devotion, any more than a photographer of the Parthenon would record its architectural goodness as a separate term along with the details of light and shade. For this reason Moore called goodness a non-natural quality. But since it is always based in this way on natural qualities, nothing is easier than to mistake them for it. Dutifulness is good; Kant says it is the only thing that is morally good; and shortly we find him arguing that in morals being dutiful is what goodness means. Mill thought that pleasure was good, and indeed the only thing that was good; and, sure enough, we soon find him arguing that ‘good’ means pleasant. This, Moore says, is gross confusion and the tag he has attached to it of ‘the naturalistic fallacy’ has become part of the stock in trade of ethics. Once we have seen the fallacy, once we have clearly marked the distinction between goodness and what Broad has conveniently called ‘good-making characteristics’, we shall see that all the attempts to identify goodness with any of these characteristics—with pleasure, or self-realization, or length and breadth of life, or any other—may be dismissed without discussion. Goodness may inhere in these characters, but it never is any of them. It is never anything but itself, a simple, highly abstract, non-spatial quality, identical in nature as one passes from the goodness of a man to that of a sunset, a dessert, or a poem.

3. I cannot accept this theory, and will content myself with mentioning two difficulties only. (1) Those who consider it carefully keep reporting that they have great trouble in catching or verifying this quality of goodness, that it is as elusive as a will-o'-the-wisp. Conspicuous among these are of course the logical empiricists, though their testimony has to be discounted by reason of their curiously a-priorist empiricism, committing them to recognize no quality that cannot be presented in sense. But other writers of discernment who are bound by no such commitment have reported the same difficulty, for example, Joseph, Taylor, Paton, and Ewing.2 Indeed Moore himself has felt it, and felt it keenly. At one stage in his later thought he confessed himself to be drawn as sharply to the emotive theory as to his own original view of good as a quality, which could only mean that he was himself uncertain whether there was any such quality.3 The difficulty has nothing to do with indefinability, for there are plenty of indefinable qualities about whose existence no one can doubt, like yellowness and pain, nor can it be merely a matter of nonsensibleness, for there are some wholly non-sensible characters, such as number, that belong to things indubitably. Nor is the difficulty that in thinking of good one seems to be thinking of nothing; all the writers just mentioned, for example, are clear that in using the term they are referring to something belonging to the object. The difficulty is that they cannot find the simple character required. And when a large proportion of the acutest minds that consider the matter find themselves either uncertain that they have even detected this quality or else certain that they have not, one's confidence that there is such a thing is bound to be affected.

4. (2) The theory makes goodness too abstract. It draws too sharp a line between goodness and good-making characteristics. In insisting that nothing that makes the good man good, or the good dessert or person or sensation good, shall enter into their goodness, that this quality is something sharply distinct, whose nature, when displayed in these characteristics, can be singled out and set over against them, this account is introducing a division that exists only in theory, not in the facts.

Consider a case or two. Pleasure or happiness is commonly taken as a clear case of the intrinsically good. But can one draw a line within a given experience of pleasure between the pleasure and its goodness? Moore apparently thinks that one can, that the pleasure and the goodness are sharply distinct, though connected synthetically by a relation of entailment, as colour and extension are in ‘what is coloured is extended’. I cannot think that this has caught the true relation between them. That relation is more intimate. The goodness of being happy is not some isolable quality supervening upon the happiness; the happiness is itself a kind of goodness; we call it a good, as we call love an emotion, meaning, I think, that it is one of the forms in which goodness presents itself. Or take pain. By scrutinizing a pain, can one distinguish in it two different components, the pain on the one side and the badness on the other, as one can the redness and the extension of a given object? That does not seem very plausible. To be sure, badness and painfulness are not the same thing, for not all bad things are painful, but the badness of the experience lies in the pain in a more intimate sense than that in which the extension lies in the colour. For an experience to be intensely painful is a way of being bad; colour is not a kind of extension, it only entails this. Or take still another type of experience. We say that an educated mind is good, or (to avoid problems about ‘dispositions’) that our final grasp, on a certain evening, of just how the Michelson-Morley experiment led up to the theory of relativity was an experience very much worth while. Is it really possible to isolate within this experience a hard little qualitative pellet called goodness, quite distinct from the understanding itself, but absolutely identical with the goodness that we have previously found in a chocolate eclair and The Moonlight Sonata? Surely goodness is not present in this external way. The goodness of the insight is not something outside it, to which one travels on a bridge of inference. Of such an illumination of mind it would be as true to say that it was an experience of good as that it was a good experience; it does not carry goodness on its back as a rider; such goodness as it has belongs to it through and through. It is itself a species or manifestation of goodness.

5. The reply may come that even if happiness and understanding are manifestations of goodness rather than experiences to which this attaches, it does not follow that the common goodness cannot be distinguished and marked off from that in which it is embodied. The genus does not lose itself in its species. Even if not separate, it remains different and distinct. It is the element of identity that preserves itself through diversities, and in each of them it keeps a hard imperviousness to its environment, like a diamond that appears first in rock, then in mud, and then in a silver setting. This is the doctrine of the abstract universal, often associated with the name of Aristotle, though Aristotle himself was unable to hold to it. That it is not wholly unplausible is shown by its stubborn longevity. If one divides the genus billiard-ball into the species white billiard-balls and red, one may say with some credibility that the generic qualities of size and roundness are a distinct and unvarying group in both. But if one takes the genus colour, and tries in the same way to put one's finger on a common colour or colouredness, which is neither red nor blue nor yellow, but a changeless component in all of them, one cannot do it; the species red and the genus colour so interpenetrate that no wedge can be driven between them. Similarly, if one tries to conceive of life or mind as an identical something in the insect and in Goethe, one will once more be defeated. So of goodness.

If we may pursue this technical point for a moment, it was this very instance of goodness that Aristotle took to illustrate the deficiencies of the mechanical kind of logic when applied to ethics. Goodness, he said, may be of many kinds. It is true that ‘the conception of whiteness appears the same in snow and in white lead. But the conceptions of honour, wisdom and pleasure, are distinct and different in respect of goodness. “Good” then is not a common term falling under one idea.’ Indeed ‘there are as many ways of predicating good as of predicating existence’.4 This parallel between goodness and being is illuminating. Aristotle taught that there were ten categories, or ultimately different kinds of being, in the world. In a sense, therefore, they were the species of which being was the genus. But one has only to ask what exactly it is that characterizes this genus to see that if it really is a genus, it is one of a very queer kind. Can one conceive of mere being as such, which is not this or that kind of being, which is not anything rather than anything else? Such being, as was pointed out later by Hegel, is virtually indistinguishable from non-being. On consideration, Aristotle declined to call this shadow a genus at all. It became clearly conceivable only in its differentiations, and in these it was itself so profoundly differentiated that what distinguished quantity from quality, for example, was far more significant than anything they had in common. Now goodness is like being in this respect. Aristotle remarks in the Ethics that goodness is so deeply differentiated that each category has its own special forms of it. The goodness of a fine waterfall which is a substance, is not that of wisdom which is a quality, or of utility which is a relation; they are almost as different as the kinds of being in which they are embodied. And if one refuses to admit into the goodness of wisdom anything that is not equally present in the waterfall and in utility, one will fail to catch the specific goodness of wisdom. That goodness one can know only through entering into wisdom. To be wise is to realize goodness in one of its forms. If this is true, and it does seem to be so, then the goodness of wisdom does not lie in an abstract quality identical in all good things. If it did, I could learn what the goodness of wisdom was by tasting a chocolate eclair, since when I call this experience good, I have in mind the same simple quality that makes anything good. Such a consequence is absurd.5

The conclusion suggested is that goodness is objective without being an objective quality. That the goodness of Emerson's wisdom should depend on the contingency of my thinking of it and approving it, as the emotivists in effect are telling us, is absurd. The opposite view, that this goodness consists of an extremely abstract non-natural quality attached to the wisdom is, if not absurd, at least unconvincing. The goodness is ‘more deeply interfused’ with the experience, belongs to it in a more integral and constitutive fashion. To see this is an advance. But we are still far short of what we want. Granted that goodness belongs in this way to the objects that have it, we want to find if we can what goodness is in itself. What is it more exactly that distinguishes what is good from what is not?

6. Moore can help us at this point. In another part of Principia Ethica than the one we have been considering, he discusses whether an object could be good apart from anyone's consciousness of it, and he uses an illustration that will suit us all the better because it is so well known.6 Let us imagine, he says, that by a turn of our hand we could create one or other of two worlds. The first is a world that is utterly beautiful, in which all the colours and shapes and sounds are in harmony with each other, and everything in gross and detail is such as to arouse delight. The other world is the antithesis of this; it is one vast heap of putrescence and filth in which everything is of the sort to disgust and repel. Before we make our choice, there is one condition that must be laid down. This is that in neither world is the element called consciousness to be admitted. The colours, shapes, and harmonies are to be there in the beautiful world, but nobody is ever to see them or hear them. The foulness and ugliness of the other are also to be there, but since no one will become aware of them, no one will ever be offended by them. Both worlds alike are to exist outside the range of human or any other experience. Which would you choose to create?

Moore in his Principia has said he would choose the more beautiful world. It would be better that it should exist, he thought, even if there were no mind to respond to it in any way. But in the years that followed he continued to reflect on the matter and, with his remarkable freedom from prejudice even in favour of his past self, he reversed his judgment, declaring that there would be no ground for choosing either. Colours unseen, sounds forever unheard, harmonies that no one perceived or rejoiced in, if such things could be conceived, as he thought they could, would at any rate be totally valueless. What would be bad about filth if it never did or could in any degree offend? Moore's final conclusion was, then, that a universe without consciousness would be a universe without value. With this I think we must agree. There is no way of proving it except the way of ideal experiment which he adopted—the way of asking oneself whether, if all modes of consciousness were banished, anything we should call good or bad would remain? Of course there are persons who would answer this differently. Thomists, for example, continue to hold the obscure view of Aristotle that each material thing is a more or less complete actualization of what it is potentially, a more or less perfect embodiment of its special form, and is, so far, good. The disappearance of a boulder on some unheard of dark star would therefore be a loss of value to the world. Of course one may conceive of value in this way if one wishes, but it carries one quite out of touch with ordinary meaning; most men would say it made no difference whatever to the amount of good in the world whether such a boulder existed or not. Take any example of what we ordinarily regard as good or bad, imagine consciousness away, and the values vanish with it.

7. Whatever is good, then, must stand in relation to consciousness. But in what relation? Several have been suggested. It has been held that if anything is to be good, it must be a state of consciousness, an object of consciousness, an object of a particular kind of consciousness, and a fitting object of a particular kind of consciousness. To the first two of these we shall devote a word or two only. The last two deserve fuller attention.

As to the first, is it true that only states of consciousness are good? Certainly our first impulse is to say No. We call sunsets and operas good, for example, and it would be odd to hear them described as conscious states. Still, we should probably agree on reflection that it is only when in fact they are such states that they have any value. The sunset we prize is of course not that of a physicist, which consists of a mass of light waves of varying frequencies, and which no one would think of describing as good or bad; it is rather an array of colours which presumably have no existence except as seen, and in any case no value except as seen and enjoyed. And it is not the opera as a set of marks on paper that has value, but the opera as heard and delighted in.

What of the second suggestion, that whatever is good must be an object of consciousness? This too is exceptionable. A man who is wise and kindly without ever thinking of his own virtues and without being recognized by his neighbours for what he is, is not less good because his virtues are unnoticed.

This has been questioned by those who hold the third type of theory as to the relation between value and consciousness, namely that to be good or bad is to be the object of some special kind of consciousness. There are many possible forms of the theory. What is good may be identified with whatever is the object of one's desire, as by Hobbes; or of one's approval, as by Westermarck; or of the general approval, as by Hume; or of the interest of anyone at all, as by Perry. Most forms of this theory wear their refutation on their face. If anyone says, with Hobbes, that ‘good’ means ‘object of my desire’, it is enough to point out that I can meaningfully judge the serenity of Buddha to have been good in spite of the fact that, since it is in the past, it cannot now be the object of any sane desire. If anyone says, as Westermarck does, that ‘good’ means ‘approved by me’, one need only point to my unquestioning belief that there are, and have been, all sorts of good things in the world that I have never heard of, and therefore cannot think of with approval. Hume's view that ‘good’ means ‘generally approved’ is refuted by the single fact that there have been reformers who were convinced that their causes were good, even though they knew them to be disapproved of by almost every living soul.

8. To my mind, the most persuasive of these theories is that of Professor Perry whose General Theory of Value defends the view that a thing is made good or bad by being the object of anyone's interest. And ‘interest’, as Perry uses it, is a very wide term, covering liking, approving, pursuing, willing, admiring, and a host of cognate attitudes, together with their opposites. This view avoids the emotivist extravagance of saying that in judgments of good we are asserting nothing at all, for it holds that ‘good’ is not an interjection, but the name of a relational property; ‘in my view’, Perry writes ‘“good” means the “feeling toward” or more precisely, “the being felt toward”’7 and the existence of a feeling is as much a fact as that of Gibraltar; assertions about it are obviously true or false. The theory though relativist, likewise avoids the subjectivism of views like Westermarck's, for the interest that is creative of value need not be one's own; if anyone anywhere loves or hates something, wants it or admires it or detests it, that is enough to clothe it with value. Perry writes: ‘in order to create values where they did not exist before it seems to be sufficient to introduce an interest. The silence of the desert is without value, until some wanderer finds it lonely and terrifying; the cataract, until some human sensibility finds it sublime.…’8 To the question, what is good?, then, he answers, ‘any object of any interest’.

9. Is this the insight we are in search of? Unhappily, I cannot think so. While Perry's books are a mine of helpful reflections about value, and his own position seems to me not far from the truth, it will not serve as it stands. It is clearly at odds with our common meaning. What we commonly mean by goodness may be present, and believed to be present, while interest is absent, and interest may be present when value, or at least any corresponding value, is absent.

(1) Do we not often say that things have been good or bad at times when we have every reason to believe that no one felt anything about them? We have already illustrated this in commenting on the emotivists. Consider the vast weltering mass of prehistoric animal pain, when life, as Dean Inge says, was ‘a conjugation of the verb “to eat” in the active and the passive’. Was all this wretchedness free from anything bad till badness was posthumously conferred upon it, at a distance of a hundred milleniums, by a sympathetic palaeontologist? Or are we to say that the badness lay in the sufferers’ ‘negative interest’ in their own pain? But we are surely much more certain that their suffering was bad than we are about their ability to perform the process that Perry regards as essential to their badness, namely cognizing the pain with a ‘negative interest’. Or if pain is too treacherous a topic, consider the case just mentioned of the good man whose goodness is not thought about, or therefore felt about, by anyone, even himself. We should certainly not regard his goodness as something that sprang into being when first appreciated. His case is typical of a very large class which Perry has perhaps not considered sufficiently. All cases of value, he says, can be symbolized by S-R-O, in which S is an interested subject, O an object, and R the relation of interest that links them. The connoisseur in the gallery looks at a Titian with delight, and the Titian therefore has value. But what of the connoisseur's experience of the Titian as opposed to the picture itself? Does not S-R-O as a whole have value, so that anyone would say it was well that it should have occurred? Presumably Professor Perry could admit this only if there were another subject S1, who could contemplate this experience of S with satisfaction, and whose relation to it could be symbolized by S1-R-(S-R-O). But it is clear that for the great majority of our experiences no such observer exists, and therefore that most of the things of whose value we are certain namely experiences, as distinct from the objects of those experiences, must, on the theory, be valueless. This is hard to accept.

10. (2) And just as objects may have value in the absence of any interest in them, so interest may be bent upon them without producing the appropriate values. On Professor Perry's theory, everything we admire or desire is good. What leaps to mind when we hear this is that people often admire and desire what is trivial or even bad; a boy may admire with all his heart the swaggering bravado of a movie hero which more mature persons regard as thoroughly cheap. Perry admits that this is ‘the most popular objection’ to his view, and much of his discussion may be read as a reply to it. If I have understood this reply, it is in substance as follows: To talk about the true or real value of such bravado is meaningless, for there is no such thing; the only value it has it takes from people's attitudes toward it. If the boy admires it, it is good for him; if his parents dislike it, it is bad for them; its ‘true’ or ‘objective’ value is a myth. When the parents say that the boy is admiring something bad, they must therefore be taking someone else's dislike, presumably their own, as authoritative, and using this as a basis from which to condemn what he admires. Have they any justification for this? Certainly not in the sense that the object as now enjoyed by the boy is not really good, for his enjoyment so far constitutes its goodness and makes it an unalterable fact. But they could legitimately say this, that if the boy could look at his hero with the knowledge and interests possessed by themselves, or by most mature and thoughtful persons, he would dislike it. Such persons prize considerateness and intelligence, with which swaggering bravado goes ill; they therefore find it a bore and a nuisance. It is wrong to say that the boy likes what is bad; strictly speaking, that is a contradiction in terms. It is correct to say that he is liking what mature and thoughtful persons dislike, and what he himself will presumably dislike eventually. But this statement is perfectly compatible with the theory. Does this reply meet the point?

We must agree, I think, that the statement ‘such bravado is bad’ does imply that a mature person, assumed to be able to see things clearly, would regard it as bad. But is that what we mean by the statement? Clearly not. For (a) we should certainly say that if the mature man so regards an object, he does so because it is bad; not that it is bad because he does so. (b) The suggested reading would turn the remark into a curiously unlikely conditional statement. What we should ordinarily mean by it is surely something like this, that an attitude of swaggering bravado is in normal circumstances a bad thing. To this the suggested reading would tack on ‘provided it is in fact disliked by mature persons generally’. But we should surely grow restive if anyone tried to saddle us with such a condition. We were saying that this attitude was bad; period. There is no such ‘if’ about it. (c) We may significantly call the attitude bad while knowing perfectly well that mature persons generally regard it as good. We sometimes believe that the estimate even of the mature persons of our time of the goodness or badness of war, or of intellectual achievement, or of the way to treat animals, is a mistaken one. And this would be impossible if what we meant by goodness was the fact of their favour.

11. I have tried to give the most plausible answer possible on Professor Perry's theory to the objection that since one may like what is bad, the liking cannot make a thing good; and this answer seems insufficient. It would be even more insufficient if he had taken certain other tempting lines. He might have said that in judging that the boy was liking something bad, we were merely reporting that we ourselves disliked it, but he sees that to interpret value judgments in this individual way would invite anarchy by putting all such judgments on a level.

But has he escaped this objection after all? He is clear that there must be some means of revising and improving our value judgments, of weighing one liking against another, if we are to avoid ranking every adolescent enthusiasm on a plane with the insights of maturity. But the only revision open to us, on his theory, is that which comes not from ‘insight’ into what is really better, but from a change in our likings, a change not justifiable by anything but further liking. Any object is better than another if it has won a more intense, a preferential, or a more inclusive interest. ‘An object, wine, is better than an object, water: (1) if the interest in the wine is more intense than the interest in the water; (2) if the wine is preferred to the water; and (3) if the interest in the wine is more inclusive than the interest in the water,’9 that is, if it sums up a larger number of interests. An object that awakes any part or the whole of this triple liking in fuller measure than another is, so far, better than it. It is almost as if Professor Perry had said that the experience of an intense preferential, and inclusive interest were better than another, though he could not say this consistently. What he does say is that goodness depends on de facto favouring or liking. Whereas Moore and Ross hold that goodness is a ‘consequential’ quality, following necessarily from the nature of the object, Perry holds that it is connected with the object only contingently, through the chance of its being liked; a thing is good not because of what it is, but because someone happens to like it.

12. Now it is certain that goodness, in our ordinary thought of it, does not sit as loose to things as this. Do we think, to use Bentham's old case, that pushpin becomes as good as poetry if only people like it as much? If two men organize their lives, the one around the game of pushpin and the other around poetry, so that their interests in their respective objects are equally intense, preferential, and inclusive, is it therefore meaningless to question whether the objects they are engrossed in are equally good? I do not think most people would take this as obvious; I suspect they would take it as untrue. Or suppose that two men devoted themselves, with interests equal in these ways, the one to astronomy or metaphysics, the other to counting and measuring the rocks on a New Hampshire farm; would the knowledge gained, because equal in interest, be regarded as equal in value? Again I do not think so. We seem to have a stubborn conviction that whatever may be the facts of men's interest, some things are worth their devotion and others not, and therefore that our interests should be adjusted to, and appraised by, the goodness, not the other way about. We are not prepared to admit even that if the race generally should come to favour something, that would settle the question of its goodness. Suppose that in the next world war our ‘proud and angry dust’ were to raise such effective dust-storms as to force the gibbering remnant of us that remained into a troglodyte existence, wherescience and art had lost their interest, moral sensitiveness was sneered at as effeminacy, and love derided in any but its animal forms. Would the fact that these things, formerly prized so highly, were now regarded with indifference or aversion, settle anything about their value?

It might be replied, to be sure, that if we were to compare the troglodyte existence with the old life before the deluge, we should see that the latter gave fuller scope for the arousing and fulfilment of interests, and was therefore better. But what would ‘better’ mean here? If it meant that a life whose capacities were more fully realized was really or objectively better, the theory is being deserted. If it meant that people actually liked such a life better, a troglodyte could disprove it by pointing to the fact that his contemporaries did not like it better; and if it were retorted that he ought to prefer a life that fulfilled his likings more inclusively, he could rejoin, and I fear unanswerably, that he happened not to like such a life, and that in that case there was no ought about it. Between such ultimate differences in liking, the theory before us would not admit even the theoretical possibility of saying which was right, or that either was right, or that one was more nearly right than the other. The fact that one is off with an old love and on with a new does not show that the new object is better, unless one means merely that it is liked more; and what is liked more may become bad overnight if one's feelings about it shift again. Valuations may be revised, but only by further feelings, which are themselves judged only by still further feelings, and so on without limit; and the revisions may with equal propriety (because in no case with any propriety) intensify what is already felt or reverse it altogether. To seek the good is rather like shooting at a target blindfold and hearing that one is getting nearer and nearer the mark, only to find, as the blindfold drops for a moment, that there is no target there at all and that every shot may be regarded, according as one is disposed, either as a bull's-eye or as a miss.

Now it may be that there is no one meaning of ‘good’ that would fit all the cases in which we take things as intrinsically good. But to break completely with common sense is to throw overboard the most useful compass we have and we must keep to it as long as we can. And we have seen that common sense would reject the suggestion that things are made good merely by our own feeling about them.

13. Still, we must not go too far. Common sense would not reject feeling as unimportant; indeed I think it would regard this as indispensable. Leave feeling out entirely, and, as we saw in the experience of the Stoics and the younger Mill, the goodness of life, as most of us conceive it, would go too. The pursuit of knowledge, the cultivation of art, the attempt to better oneself or one's community, is simply impossible when the wind of interest dies out of one's sails. There is no savour left in the salt when desire fails; the pitcher, as the Old Testament Preacher said, stands broken at the fountain and the wheel broken at the cistern. Why turn one's hand or lift one's finger if nothing can possibly come of it that one could welcome? As James once said, all the art on the walls of the Vatican would have no value for a stray dog lost in the halls, for it is unable to notice, or if it notices, to take the slightest interest in such things. Perry and the emotivists are right in stressing that interest is a condition of goodness. Where they have gone astray, I venture to think, is in identifying interest with goodness. In the end one gags at the paradox that one can make hatred and ignorance good if one is depraved enough to like them. Most men are convinced that hatred and ignorance have a character of their own which appoints some attitudes towards them as fitting and others as unfitting. They would say that if we like them, that does not transmute them into something good, but merely reveals how blind we are.

14. This brings us to the other type of theory that would make of goodness a relational property. It has been stated persuasively by Dr Ewing in The Definition of Good. With much of what we have said about the simple-quality theory and the interest theory I think Dr Ewing might agree. But he finds important truth in both. Moore is clearly right, he thinks, that typical judgments of good mean to say something true of an object, something that would still be true, however one felt about it. On the other hand, Perry is clearly right that feelings and desires are bound up in a peculiarly intimate way both with the making of these judgments and with what they assert. How are we to reconcile the objectivity of the judgment with this curious involvement with feeling? The strength of Dr Ewing's suggestion lies in its resolute attempt at justice to both of these all but irreconcilable demands. The suggestion is that when I call anything good, what I mean is not that I favour it, or that anyone else does actually favour it, but simply that it is fitting and proper to favour it, that the thing called good is a ‘fitting object of a pro attitude’. Pro’, of course, means favouring. ‘Attitude’ is a blanket term which ‘covers, for instance, choice, desire, liking, pursuit, approval, admiration.’ ‘When something is intrinsically good, it is (other things being equal) something that on its own account we ought to welcome, rejoice in if it exists, seek to produce if it does not exist. We ought to approve its attainment, count its loss deprivation, hope for and not dread its coming if this is likely, avoid what hinders its production, etc.’10 To say that an action is good is to say that it is a fitting object of moral admiration or approval. To say that an experience is good is to say that it is worthy of being sought and prized. To say that pain is evil is to say that it is a suitable object of aversion.

15. Dr Ewing is convinced, and with reason, that this theory has great advantages. For one thing, it thins the region of controversy between the naturalist and the non-naturalist in ethics to the narrowest possible area. The naturalist is the person who thinks that ethical characters such as good and right reduce without remainder to the sort of observable characters dealt with in physical or psychological science, for example that ‘good’ means actually desired by oneself, or by people generally. The non-naturalist is the person who thinks that this reduction cannot be carried through. Dr Ewing is a non-naturalist.11 But he is inclined to think that there is only one ethical term that holds out against such reduction. This is what he calls ‘fittingness’. When we interpret ‘x is good’ by ‘x is a fitting object of favour’, the fitness of attitude to object is a relation, but it is not like before, or larger than, or part of, which are commonly thought to belong to empirical science. The fitness or fittingness of a favouring attitude to a pleasure, for example, is not an empirical relation at all. It is a relation of synthetic necessity apprehended by intuition, which is itself a form of reason. One sees immediately that, pleasure being what it is, a favouring attitude toward it is appropriate, and a disfavouring attitude is not.

But though this relation must be set down as non-natural, it is the only term of this kind that Dr Ewing finds it necessary to introduce, for all other ethical terms can be defined by means of it. Thus the controversy between Moore and Ross over whether ought depends on good is cleared up with singular ease. Dr Ewing agrees with Moore that it is always our duty so to act as to bring the most good into the world. But he then turns round and defines good in terms of ought; what is good is what ought to be favoured, in the sense that it would be fitting to favour it, that its nature calls for such favour. There is thus no conflict between good on the one hand, and ought or right on the other; you may say that either is primary as you wish; and the parties to this ancient controversy may both go away happy. The only ethical term that might make a show of resistance against reduction to fittingness is ‘ought’ in its specifically moral sense, the ‘ought’ that means ‘it is my duty to do this’; but even that he thinks may be brought into line by a double dose of fittingness: the judgment would now mean, first, ‘it would be fitting for me to do this’ and second, ‘if I do not do it, I shall be a fitting object of disapproval’.12 Thus Dr Ewing is not asking the naturalist to admit into his universe a pantheon of dubious divinities—good, right, wrong, ought, sin, guilt. Let him admit this one relation of fittingness, and he will be troubled by no more importunities from the non-natural realm.

16. The theory has a second great advantage: it gives empirical and ordered meanings to a vast range of value terms. Besides the generic ‘good’ which is now to mean fittingly favoured, there are multitudes of sub-terms for special forms of good—admirable, beautiful, lovable, charming, sublime, noble, honourable, and so on—and there is an even more extensive armoury for the forms of evil. If all these terms were taken as meaning characters in the object, each with a quality quite distinct from that of our feeling about it, the task of distinguishing these qualities would become impossible. Try to say what lovable means without reference to the quality of love, and you are lost. On the other hand, define it as ‘such that love may be appropriately felt for it’ and the definition, however vague, is in terms of a perfectly familiar relational property. So of the other value qualities. Taken as out there in the object, independent of all attitudes toward it, they return to the mysterious status of Moore's indefinable good; they are virtually or wholly characterless; and trying to relate them to each other is like trying to round up and classify the clouds on a breezy day. But everyone knows what it feels like to love or admire or reverence or desire; everyone knows many degrees of such feelings; everyone knows the difference, however hard it is to define, between approving something morally and appreciating it aesthetically. Once the value qualities are allowed to borrow content from the attitudes specially appropriate to them, the chaos of these qualities falls into an order determined by the degrees, affiliations, and mixtures of these attitudes themselves.

17. Finally, as Dr Ewing points out, the theory accords most happily with the way moral beliefs arise. We do not come to think uncleanliness, lewdness, idleness, and ignorance bad through achieving an intellectual insight into the necessitation by these things of an objective quality called ‘evil’, but rather through catching the way people feel about them. In childhood, if our mother abhors a certain person and shows it, that is enough for us; he is bad, and we are ready to do battle against him whether we know anything about him or not, just as a faithful dog acts on the principle ‘whoever my master is against is a limb of Satan and fair game for me’. It is true that as we grow up we amend these attitudes in the light of further knowledge. But this does not mean that growing up morally is a process of ascent out of the fogs and clouds of attitude to the passionless sage, sitting atop a snowy and rational peak. A graduation from feelings and desires would be a graduation right out of the moral life. What actually happens is that, beginning with a mass of likes and aversions picked up by the contagion of sympathy, we little by little correct them, partly through contact with people who feel otherwise, partly through learning more about the objects and finding that our feelings hardly suit them. Moral judgment is at first almost pure prejudice, in the sense that it is an expression chiefly of ungrounded, or ill grounded, feeling. In the educated mind it never remains that wholly. In conversation with Charles Lamb, a friend happened to name another man: ‘I h-hate that man’, Lamb stammered out. ‘Why I didn't suppose you even knew him’, remarked the other, in surprise at such an outburst from the friendliest of men. ‘I d-don't’, said Lamb. ‘I c-can't h-hate a man I know.’ Changing knowledge changes our attitudes, clearly. But it does not cause them simply to deliquesce and vanish. They remain present, and they remain essential, from first to last. Dr Ewing is right, I think, in insisting that the judgment of good, even in its most sophisticated form, does imply that its object merits the favour, or would satisfy the feelings, of a reasonable man. We may concede to the theory these important points of strength. May we then accept it without further ado? Unfortunately, there are difficulties through which I cannot quite see my way.

18. (1) First, the conviction persists that if it is fitting to favour an object, this is because the object is good already. And if it is thus good, its goodness cannot be exhausted in the fittingness of possible favour. The account of good here given seems to presuppose another good in the object, of which that account could not be repeated. To put it more concretely: a man by discipline of mind and character achieves what we call wisdom, and we say of this wisdom that it is good. That means on the view before us, that his state of mind may fitly be favoured. Is it an intelligible question to ask why it should be favoured? I cannot help thinking that it is. On the theory before us, it is not. For according to that theory, the only answer we could give is ‘because wisdom is wisdom’; we could not say ‘because wisdom is good’, because all ‘good’ means is that it should be favoured, and we should therefore be explaining why wisdom should be favoured merely by saying that it should be favoured. The question why it should be favoured is thus unanswerable because really meaningless. One can say that a state of mind should be favoured in virtue of its wisdom; one cannot say that the wisdom should be favoured in virtue of anything at all; to see that wisdom is good is to see that it should be favoured. One either sees that it should be, or one does not; and if one does not, no reason can be offered why it should. I must admit that I am not content. It still seems to me truer to say that we favour the object because it is good than to say the object is good because we may fittingly favour it. Are there any crucial considerations that may help us decide this issue? I suggest the following.

Consider what the object is like independently of any attitudes actual or possible. What is being thought of, let us say, is the wisdom of Emerson. Was it good or not? Obviously and by definition, not. On the theory, all objects, considered apart from a possible favouring or disfavouring, are neutral as respects value, for it is only in relation to such favouring that they get their value. So considered, then, nothing is better or worse than anything else. Emerson's wisdom is a matter of fact; so is the dullness of Peter Bell; and as mere natural facts, they are on a level. We now ask whether it is fitting that one of these states of mind should be favoured over the other, and the answer given is Yes. This answer is even taken as expressing a necessary insight; wisdom requires favouring in virtue of what it is. And yet on the terms of the theory itself, it is hard to regard this answer as other than arbitrary. If the goodness of wisdom lay in its own nature, as Joseph, following Plato, contended, and I, following both at a due distance, should also contend, then the favouring of it would have the rational character that Dr Ewing wishes to preserve in the moral judgment. But if the wisdom has no goodness apart from the reference to favouring it, just why should it be favoured? Dr Ewing is of course alive to the difficulty, and he gives a straightforward answer to it.

‘It will be objected against me that it is only fitting to approve, or have a pro attitude towards, what is good because we first know or believe it to be good, and that if we did not believe it to be good, there would be no ground for such an attitude, so that the attitude would not be fitting. The answer is that the ground lies not in some other ethical concept, goodness, but in the concrete factual characteristics of what we pronounce good. Certain characteristics are such that the fitting response to what possesses them is a pro attitude, and that is all there is to it.’13

My difficulty is to see how a pro attitude could be fitting merely to ‘concrete factual characteristics’, and Dr Ewing seems at times to feel this himself. Thus he writes: ‘A principal objection made by me against naturalist definitions of “good” was just that, if “good” were defined naturalistically, it would be no more rational, right, fitting to pursue the good than the bad and that good would carry with it no moral obligation to pursue the good.’14 Now unless I have misunderstood, the naturalistic qualities here said to offer no ground for favouring or pursuit are the same qualities which, in another context, we are told require a favouring response. And it seems clear that they cannot be both. If these factual characteristics do deserve a response of moral favouring or disfavouring, an important part of Dr Ewing's case against naturalism, which seems to me very effective, will have to be abandoned. If they do not, can he say, as he does, that they are ‘such that the fitting response to what possesses them is the pro-attitude’? The phrase ‘such that’ is puzzling. What does it mean? At one point Dr Ewing suggests that ‘I might well admire something just because I thought it worthy to be admired’,15 with no idea of what it was in the object that was admirable; but he adds immediately that if this thought is to be justified, the qualities that are admirable must be pointed out; a blank cheque could hardly be honoured at the bank where our admirations are stored. If Dr Ewing were to explore the implications of ‘such that’, he might find them carrying him to a fuller objectivism than that which he now holds. The moral judgment emerges from his account a remarkably ingenious bridge, with the agent's favour at one end, a non-natural relation of fittingness or congruity forming the imposing span, and the other end resting on nothing visible to the ethical eye. To some relations of fittingness natural qualities could indeed give support. The roundness of the hole is fitting to the roundness of the peg; the shape of the shore fits exactly the shape of the ocean; between two notes in tune there is congruence in frequency of wave. But in these fittings there is no goodness. Goodness comes only when in the object there is something that answers to favour, something ‘such as’ to require and deserve it. And you cannot fill that blank called ‘such as’ with neutral natural qualities, with round holes or coastal inlets or physical frequencies; you cannot even fill it with St Francises or with the giving of all one has to feed the poor, if these are considered solely as natural events. As such, they are mere facts, none of them better than any other. If saintliness and generosity are such as to merit favouring, it must be because there is something in them that goes beyond their ‘factual characteristics’ and equally goes beyond a mere blank cheque on our favour. What is this? I think we must answer, a goodness that they have already. Unhappily the nature of this goodness eludes us still.

19. (2) This is the fundamental objection to a theory that seems to me very nearly right. But it may be well to mention one or two others that are perhaps less objections than difficulties. For one thing, considering how important it is, the relation of fittingness is left unfittingly obscure. Dr Ewing thinks that to say ‘wisdom merits favour’ is a synthetic necessary proposition, thereby attracting around his ears a Pandora's box of buzzing cavils from those who hold that there are no necessities outside logic and only analytic necessities inside it. His refusal to flee before this cloud of gnats I find admirable, meaning more, I think, than merely that to admire it would be fitting. But to say that a relation of congruity between an object and an emotion may carry necessity in it is in these days a courageous thing to say, and even though true, needs explanation. To be more specific, Dr Ewing objects, as I do, to the idea of an abstract identical goodness in St Francis, the Moonlight Sonata, and a chocolate eclair. But for all one knows, he holds that the fittingness of our favour for these three is itself identical, that the fittingness that marks our favour of a generous deed is the same as that which marks a ready responsiveness to humour, a due appreciation of music or painting, religious reverence, and the hatred of cruelty. Furthermore, I gather he would say, with Professor Broad and Sir David Ross, that the act of a man who performs a rescue at sea ‘fits the situation’. I think he would say that the relation is an internal one, in the sense that the agent has it to the situation in virtue of the nature of that situation. If so, does it vary as its terms vary? And how is its necessity related to that of logic and mathematics? I speak as one who is convinced, as Dr Ewing is, that the current exclusion of rational necessity from all domains except formal logic is an unfortunate passing fashion. But just for this reason, I should like to see the case against this, which is surely a powerful one, more fully developed.

20. (3) Thirdly, if goodness were the same as fittingness of favour, they could never fall apart, but it seems to me that occasionally they do. Favour and disfavour include, as we have seen, many kinds and intensities of attitude. I think Dr Ewing would say that if one of our children were starving, our judgment that this was bad would mean that the strongest aversion toward it, acute grief about it, and energetic efforts to remove it, would be fitting. Now there happens to be another child starving in central China. To feel nothing about this would certainly be unfitting if we knew its plight. On the other hand, perhaps no one would say that the suffering had the same claim on our feelings as that of our own child. There are many feelings, felt with great intensity, that are obviously suitable in the case of our own child, which we could hardly be expected to show about one that was remote and all but unknown. Now if to call anything bad is to say that it is the fitting object of anti-attitudes, then when we call the remote child's starvation bad, we must mean that it is a great deal less bad than our own child's starvation, since the attitudes fitting in this case are less various and less intense. But if it were put to us expressly that this was what we meant, I think we should say ‘No, I didn't mean that at all; what the Chinese child is going through may be as bad as anything my own child is suffering, or even worse; I cannot feel as keenly about remote evils as I do about those nearer home, and have no sense that I ought to, but I can still recognize quite clearly that those evils are as great as those near by.’ In short, the judged badness may be the same while the fitting attitudes differ; the one, therefore, is not the same as the other.

Dr Ewing is aware of this difficulty. His reply would be, I take it, that the respect in which the attitudes may fittingly differ is not that in which their fitness determines goodness. ‘To speak of the good of two persons as equal means that, other things being equal, neither had more claim to pursuit than the other; but if I have a special relation involving obligation to one person which I have not to the other, other things are not equal.’ In the case of my own child I have a duty arising out of its nearness, its being mine, and my being able to help; in the other case, I have admittedly no such duty. But this difference in duty arises from taking into account much besides the respective sufferings of the children. If we confine ourselves to these sufferings, we find that the attitude of aversion appropriate to the one is the same as that appropriate to the other, and thus the fitting attitudes tally with the degree of judged badness.

This is surely an effective reply when the difference of attitude is a difference of duty. I am not clear that it is equally so where duty is not involved. We read in the morning paper about a train wreck three miles off, in which many lost their lives. We note that no friends of ours were involved, and there is nothing we can do about it, yet we are much stirred. We then go into our study and read about certain ‘old unhappy far-off things and battles long ago’ which involved, we know, much more misery than the wreck of yesterday. About these gigantic far-off ills we feel only a gentle melancholy, and we are not, in our own eyes, guilty of impropriety in so feeling. Here the amounts of evil assigned to the events are the reverse of what they ought to be if the degrees of aversion accepted as fitting in our attitudes were the measure of badness. Again, suppose that two persons, oneself and another, perform actions of equal courage or moral goodness. It is certainly fitting that one should admire the action of the other man. Dr Ewing agrees that it would be unfitting to indulge in the same admiration for oneself.16 Fittingness and goodness seem once more to fall apart. Dr Ewing, I am sure, has considered this difficulty, but I am not clear what his answer would be.

21. If the criticism we have offered is well founded, the goodness of anything is not defined or exhausted by the fitness of favouring it. Favouring it will be fitting if it is good, but that is not what its goodness means. To say that something deserves to be favoured is to say that there is already in the object something that lifts it out of neutrality and justifies taking a positive rather than a negative attitude toward it. That in the object which makes it fitting to favour it can hardly be the fittingness itself. We are therefore thrown back again upon our apparently endless search.

But we have made some progress. We have seen that the fundamental question of ethics is what is meant by ‘good’, and in this chapter we have considered three theories of what it means, offered by three very able moralists. So far as we could make out, they were all partly wrong and partly right. Moore was wrong in holding that goodness is a simple unanalysable quality, but right in holding that when we talk about the goodness of someone's pleasure or the badness of someone's pain, we are talking about something objective, in the sense of belonging in the object, and not merely in our attitude toward it. Perry was wrong in holding that value consists in interest, but he was right in holding that for the person who has no feeling about an object, that object has no value, and hence that there is no value in the world except as relative to consciousness. Ewing was wrong in reducing goodness to the appropriateness of favour, though he was right in holding that goodness does not vary with attitudes actually taken. We have now a conundrum set for our solution. The goodness of an experience is an objective character of it, not reducible to feeling and not necessarily variable with feeling, and yet in some sense dependent on it. What sort of character is this? We shall try to suggest an answer in the next chapter.

From the book: