You are here

XI: The Nature of Goodness

THERE are two reasons which make it necessary for any student of ethics to devote attention to the nature of goodness. One is that we habitually think of certain kinds of disposition and action, and of certain persons, as morally good; and we can hardly hope to know clearly what is meant by ‘morally good’ unless we know both the nature of goodness in general, and what distinguishes moral goodness from any other form of goodness. The other is that a great part of our duty—indeed, according to a widely accepted theory, the whole of our duty—is to bring what is good into existence. Even if we reject that theory it must be admitted that where no special duty such as that of promise-keeping is involved, our duty is just to produce as much good as we can.

The first thing that strikes us in examining the meaning of the word ‘good’ is the very wide variety of things to which we apply the name. We apply it to persons, to their characters, to their actions, to their dispositions, to tools and machines, to works of art, to states of affairs; and the suspicion naturally arises that we cannot be using the word in the same sense in all this variety of applications. Not that the variety of the things to which we apply it necessarily involves a variation in the meaning in which we use it; for a word like ‘animal’, for instance, is applicable to very various things, ranging from Julius Caesar and Shakespeare to the amoeba, and yet in calling any of them an animal we are using ‘animal’ in a single sense, i.e. as standing for a certain very general type of structure and life, which admits of great variety in detail. But there are certain cases in which, on comparing two different applications of a word, we can see that the word cannot mean the same in both cases. Thus we may speak of a horse as being fast, and we may also speak of a movement as being fast, but we see that while the horse is fast in the sense that it moves a relatively great distance in a relatively short time, a movement is fast not in this sense (for a movement does not move), but in the sense that it is the moving over a great distance in a short time. Aristotle generalized this consideration into the doctrine that when any term is applied to things in more than one category, it must have different meanings in these different categories; and he considered that ‘good’ is a case in point, since it can be applied to substances, qualities, quantities, relations, actions, passivities, times, and places.1 It does not seem to me that his general doctrine is justified. It would seem, for instance, that certain times and places might both be good in the single sense that it is useful for certain human purposes that certain events should take place at them. And in any case from the mere variety of the things to which a term is applicable, even where the variety amounts to a difference of category, it cannot safely be inferred that the term itself must have different meanings. Yet the example of ‘fast’ has shown us that sometimes the application of a term to things in different categories is possible only in virtue of the term's having two meanings.

When a term is applied to things in different categories, and yet there is some connexion between its meanings in these different applications, and not a mere chance using of the same noise in entirely unconnected meanings, Aristotle suggests that the relation of the different meanings may be either that they are derived from one single central meaning, or that the things called good contribute to one end, or that there is an analogy between the different meanings.2 He does not expand these alternatives at all, but I think we may take the first alternative as particularly suggestive. In our instance of ‘fast’, the relation between the two meanings is that both are connected with the covering of a great distance in a short time, a horse being fast because it does this, and a movement fast because it is this. And it would seem likely that the relation of the meanings of the word ‘good’ when we apply it to men, to characters, and to actions or emotions, is of the same order. But it is much harder to be sure of this in the case of a term like ‘good’, which in one of its senses at least seems to be unanalysable, than in the case of a term like ‘fast’, which is always analysable in terms of an interval covered and of the time taken to cover it. We can analyse the meaning of ‘fast’ in its two applications, and see that the definitions which unfold its meaning in the two cases are different but connected in a definite way. We cannot do that with ‘good’, or at least it is not initially clear that we can. The Oxford English Dictionary very judiciously gives as its primary definition of ‘good’ ‘the most general adjective of commendation, implying the existence in a high, or at least satisfactory, degree of characteristic qualities which are either admirable in themselves or useful for some purpose’. Probably no more definite account than this will cover the whole variety of the applications of the word. Probably the only universal precondition of our using the word is the existence of a favourable attitude in ourselves towards the object. And this may give rise in some minds to the thought that what we are asserting of the object is that it is the object of such a favourable attitude—which would at once imply that ‘good’ is a relational term, signifying that there is a certain relation between the object and him who judges it to be good. To correct this, it may be enough to refer to a point made by Meinong, and thus summarized by one of his expositors:

‘Language serves a double function: it expresses our states of mind and it means or refers to the objects of those states of mind. A man who utters the words “red” or “blue” gives expression to a peculiar inner experience through which he is living, but he is not meaning or referring to this experience. He is talking about certain properties which can only be manifested in extended objects.’3

‘If I make use of the word “sun”, I am, whether I wish it or not, giving expression (Ausdruck) to the particular mental process called an idea, to the fact that, either in perception or imagination, something is being set before my mind. But at the same time, in so far as I express this idea, I also refer to a certain physical object, namely the sun, and this reference to the sun is the meaning (Bedeutung) of the word. If a person hears me use the word “sun” he can take this word as a sign of a certain idea in my mind, whose existence he can infer with high probability. But it is perfectly plain, as Meinong points out, that this idea is not what I mean when I speak of the sun; unless I am introspecting, and attempting to examine the mental states which accompany my use of words, nothing can be farther from my thoughts than my own ideas. What I am concerned with, what I am referring to, is an extended physical object millions of miles away, which does not resemble my mental processes and stands in no real relation to anything in my mind.’4

In the same way what we express when we call an object good is our attitude towards it, but what we mean is something about the object itself and not about our attitude towards it. When we call an object good we are commending it, but to commend it is not to say that we are commending it, but to say that it has a certain character, which we think it would have whether we were commending it or not.

What then is the characteristic which we ascribe to something when we call it good? I do not think that there is any one characteristic which we are ascribing wherever we call something good. What unites all our applications of the word is not a single connotation of ‘good’, but this single type of attitude, the favourable attitude, which we are always expressing, the meanings of ‘good’ itself being very various. To bring out this variety, I think it useful first to distinguish the adjective or attributive use of the word from its predicative use, i.e. the usage in which we say ‘a good so-and-so’ from that in which we say ‘so-and-so is good’. When we say ‘a good-so-and-so’, what we are ascribing to something is ‘goodness of its kind’; and this has various meanings according as we are speaking of (1) a person or of (2) a thing. I venture to quote here some sentences from a previous discussion by myself.

‘In case (1) the root idea expressed by “good” seems to be that of success or efficiency. We ascribe to some one a certain endeavour, and describe him as a good so-and-so if we think him comparatively successful in this endeavour. It might be thought that in certain cases (e.g. “a good singer”, “a good doctor”) another idea is in our minds, viz. that the person in question ministers to our pleasure, or to our health—in general to the satisfaction of some desire of ours. But our pleasure or our health comes in only incidentally in such cases; it comes in just because the endeavour we are imputing to the person in question is the endeavour to give us pleasure or to improve our health. It does not, therefore, it would appear, form part of the general connotation of “good” when thus used. We can in this same sense call a man “a good liar”, not because he contributes to the satisfaction of any of our desires, but because we think him successful in what he sets out to do.

‘In case (2) there appear to be various elements included in what we mean by “good”. We seem to mean in the first place (a) “ministering to some particular human interest”. A good knife is essentially one that can be successfully used for cutting, a good poem one that arouses aesthetic pleasure in us. But there is also here (b) the notion that the thing in question is one in which the maker of it has successfully achieved his purpose—a notion which might be called the “passive” counterpart of the notion explained under (1). As a rule both the notions (a) and (b) appear to be involved in our application of “good” to anything other than persons; but sometimes the one and sometimes the other predominates. There is, however, (c) a third element, less seriously intended, in our application of “good” to non-persons. When we speak of a good lie or of a good sunset we are half-personalizing lies and sunsets and thinking of this particular lie or sunset as succeeding in that which all lies or sunsets are trying to achieve; i.e. we are, not quite seriously, transferring to non-persons the meaning of “good” appropriate to persons.’5

I think it is clear that this adjunctive use of ‘good’ has no importance for ethics. The meanings that are important for ethics are that in which we say ‘such-and-such a man is good’, meaning ‘morally good’, and that (or those) in which we say (rightly or wrongly) ‘virtue is good’, ‘knowledge is good’, ‘pleasure is good’. It is obviously very important for ethics to discover which of such statements as these three are true, because we have agreed that one of our main duties is to produce as much that is good as possible, and we can attach no concrete meaning to this till we have discussed what things are good in this specially important sense. We may agree that what we are ascribing to things when we call them merely good of their kind is something that can be defined in a purely naturalistic way, by reference to human wishes and their fulfilment; the question remains whether ‘good’ in its predicative sense can be so defined, or indeed can be defined at all.

Of the predicative applications of‘good’, it seems to me that we can distinguish three main types. (1) There is first the sense in which a hedonist might say that virtue is good. He does not think that virtue is good in itself, for he thinks that pleasure alone is this. He means that virtue is useful as a means to something that is good in itself. Here we have already come to something that, on the face of it, cannot be defined in a purely naturalistic way. One element of our definition, that in which we use the phrase ‘a means’ or some equivalent, is naturalistic, since it simply states that there is a causal relation between one thing and another thing which is or may be desired. But in saying that virtue is a means to something good in itself, we are including a non-naturalistic element in our definition.

I do not wish to call the usage of ‘good’ as equivalent to ‘a means to good’ improper. It is a perfectly sound idiomatic use of the word. But it is clearly to be distinguished from the sense of ‘good’ as ‘good’ in itself or ‘intrinsically good’ or ‘good apart from its results’, and it will be better, in speaking or writing philosophy, not to say ‘good’ when we mean useful as a means to what is good in itself, but to use this phrase or an equivalent.

But what is good in itself may be so in either of two senses. (2) We may call something which has both good elements, and bad or indifferent elements, good in itself, when we think the good elements outweigh the bad ones. Or (3) we may call something good, meaning that it is good through and through. Professor Moore calls things of the first kind intrinsically good, and those of the second both intrinsically and ultimately good.6 Phrases which would more clearly indicate the difference are the phrases ‘good on the whole’ and ‘good through and through’. Only things that are good through and through will be good in the strictest sense of the word, and the questions I want to address myself to are (1) what is the nature of that which we are ascribing when we say of something that it is good in this sense, and (2) what are the things that are good in this sense. These questions have to be to some extent considered together. Under the first question I want to consider in particular what category goodness of this sort comes under. Is it a quality? Is it a relation? Is it a relational property? Or does it form a separate category from any of these? Or does it with any other characteristic or characteristics form a category other than those named?

In my book The Right and the Good7 I have discussed at length the view that goodness is a relation or a relational property, and in particular Professor Perry's view that what we say of a thing when we say that it is good is that it is an object of interest to some one or other. I believe that in its essence the argument I have offered is right. In particular, it seems to me quite clear that there are many things which we know to be objects of interest to many people but yet unhesitatingly describe as bad. And further it seems to me clear that when, for instance, we describe a conscientious or a benevolent action as good we are ascribing to it a characteristic that we think it has in itself, apart from the reaction of any one to it.

Suppose that we deny that certain moral qualities, such as conscientiousness or benevolence, have a characteristic of goodness which is independent of any one's reaction to them, what then are we really affirming about goodness? One or other of three things: (1) that goodness is properly defined as the being the object of a certain kind of reaction, (2) that things that are good have goodness in consequence of being the objects of a certain kind of reaction, or (3) that there is no such thing as goodness at all, the only relevant fact being just the fact that certain things are the objects of a favourable reaction.

On the first view, it is asserted that what we mean when we call certain states of mind good is just that they are the objects of a favourable reaction. Now I think it is clear that we have not this meaning in mind when we normally use the word ‘good’. When we call a state of mind good we are thinking of the state of mind itself, and not of our or of any one's reaction to it. Yet it must be admitted that we often use terms not quite unintelligently, and yet without realizing precisely what we are thinking of. This admission seems to me to be required by the fact that we sometimes search for the definition of a term, and accept a certain definition as correct. We should not be searching for the definition if we already knew precisely the meaning of the term; but the fact that we accept a certain definition as correct shows that we think the definition expresses more clearly the very thing that we had in mind when we used the term without knowing its definition. Thus the fact that we have not our own reaction, or any one's reaction, distinctly in mind, when we use the word ‘good’, is not sufficient evidence that that is not its true definition. The correctness of a definition may be tested by two methods: (i) by asking whether the denotation of the term and that of the proposed definition are the same, whether the definition applies to all things to which the term applies, and to no others. But that is not enough. ‘Equilateral triangle’ and ‘triangle having all its angles equal’ have exactly the same denotation, but the one is not a correct definition of the other, since what we mean when we call a triangle equilateral is not that its angles are equal but that its sides are equal. We must therefore ask a second question, (ii) ‘does the definition express explicitly what we had implicitly in mind when we used the term?’ We may apply this double test to any proposed relational definition of goodness. Most of the relational views fail to survive either test. For we may divide them according as they identify goodness with being the object of a favourable reaction (a) by some one or other, no matter by whom, (b) by the person who judges something to be good, (c) by a majority of some class of mankind, (d) by the whole of some class of mankind, (e) by a majority of men, or (f) by all men.8 Now as regards (a), it is clear that we sometimes deny something (say hatred) to be good, or doubt whether it is ever good, when we do not doubt that some one or other has had a favourable reaction to it; so that the definition fails to satisfy the first test, and must therefore fail to satisfy the second as well. As regards (b), it may be admitted that this definition satisfies the first test. No one judges anything to be good unless he has a favourable reaction to it, and it might be possible to specify a particular favourable reaction—say approval—which we never have without judging the object to be good. But this definition fails to satisfy the second test. For while the favourable reaction is what the judgement ‘this is good’ expresses (to use Meinong's language), it is not what it asserts.9 That this is so can be most easily seen from the fact that if this were what the judgement asserts, then two people of whom one says ‘this is good’ and the other says ‘this is bad’ would not be contradicting each other, since it might be true both that A approves of the object and that B disapproves of it; whereas it is clear that A and B do mean to contradict each other. As regards definitions (c) to (f), they fail to satisfy either test. It is clear that we often judge an object to be good when we do not think that a majority, or the whole, of any class of mankind or of mankind itself is feeling an emotion of approval towards it; and it is further clear that even if we do sometimes think that a majority or the whole of some set of men is approving of object O, that is not what we mean to assert when we say that O is good.

It would be possible to try to avoid this objection by modifying the relational definition of ‘good’. We might say that ‘good’ means not ‘arousing an emotion of approval in so-and-so’, but ‘such as to arouse such an emotion when attended to’. This would get over the time-difficulty that attaches to the original suggestion, viz. that we constantly describe something as good when we have no reason to suppose that the whole or a majority of any set of men is even attending to the object in question, let alone approving of it. But the new suggestion remains open to this objection, that we often call something good when quite certainly no thought even of what the whole or a majority of any class of beings would feel if it attended to the object is even implicitly in our minds.

Thus it is not in the least plausible to identify goodness either with being the object of a favourable emotion or with the power to awake a favourable emotion.10 That is most certainly not what we are thinking of either explicitly or even implicitly when we call a moral action, for instance, good. Now no one is likely to suggest11 that the existence of this relation between an object and some mind or minds gives rise to a quality in the object (distinct from this relation) the name for which is goodness. The most plausible form in which the relational view could be expressed would be to say12 that nothing possesses the kind of intrinsic characteristic which we ascribe to things when we call them good; that some things are, however, the actual or possible objects of a favourable emotion, and that on the strength of this we mistakenly ascribe to them goodness in themselves. To say that would be more plausible than trying to persuade us that by ‘good’ we mean something which we plainly do not mean.

The fact is, however, that it is impossible to approve of anything without thinking it worthy of approval—without thinking that it has a goodness of its own which makes it fit to be approved. The view that the whole fact is that certain things are approved is one which makes nonsense of approval itself. If things were only approved, without anything being worthy of approval, the act of approval would simply be nonsensical. Approval may be misplaced in detail; the fact that a particular person approves a particular thing does not imply that that thing is actually worthy of approval; but the fact that we approve at all, rightly or wrongly, is the clearest possible evidence of a universal conviction that there are some things that are worthy of being approved. And disagreement about what things are good is just as clear evidence of this conviction as agreement about it would be.

The fact of being approved, then, which the theory we are examining seeks to identify with, or to substitute for, goodness, is a fact which could not exist apart from the thought that the object is worthy of approval, in other words is good in itself. And I believe that attention to our state of mind when we express approval of conscientiousness, say, or of benevolence shows that what we really think about them is that they are good in themselves. No one can prove that they are, but then nothing could be proved unless there were truths which are apprehended without proof; and we apprehend that conscientiousness or benevolence is good with as complete certainty, directness, and self-evidence as we ever apprehend anything.

But there are other things besides moral dispositions and actions that we habitually think good; notably the exercise of intelligence, and the feeling of pleasure; and the question arises whether these are good in the same perfectly objective and indefinable way in which good moral dispositions and actions are good. I wish to take account in particular of an argument by Professor C. A. Campbell13 to the effect that only moral virtue is good in this perfectly objective sense of ‘good’. The gist of his argument is to suggest that of the goodness of anything other than moral virtue a relational account can be given. He grants, I think, the truth of my criticism of the relational theories which identify goodness with being the object of a favourable emotion to the person who judges something to be good, or to the whole or a majority of any set of beings. But he thinks that a more complicated form of relational account can be given of the goodness of intellectual or aesthetic activity. His general view is that ‘all value judgements other than those referring to moral virtue involve an essential reference to human liking’.14 He takes as being at least the most important things, other than virtue, which we judge to be valuable, knowledge and aesthetic experience—these words being the more exact way of referring to the truth and beauty that are named in the familiar trinity of ‘goodness, truth, and beauty’. He starts by considering ‘liking’; and ‘liking’, as it is described by him, seems to stand for the two facts of desiring a thing when it is absent, and finding satisfaction or pleasure in it when it is present; or more strictly for the having of a relatively permanent disposition which leads us to desire something when we think of it as a thing we have not got, and to enjoy it when we have it. But he admits that not everything that is liked is seriously thought of as good for oneself, still less as good for man or simply as good. He therefore introduces certain distinctions between objects of our liking. The first important distinction that he draws is between things liked for themselves and things liked as means to things liked for themselves. It is obviously only the former that ‘have a direct claim to the title “value”,’ and we therefore tend ‘to identify value-for-self not with object of any liking of the self, but with object of an independent liking of the self’.15

The next distinction is one drawn between different ‘end-values’, viz. that between end-values which have also instrumental value and those which have instrumental disvalue; health and knowledge being instances of the former, idleness and gluttony of the latter. Now since the self, when it has certain likings, is conscious that then or at other times it has other likings, objects of liking which interfere with other objects of liking, more numerous than themselves or the objects of more intense liking, come to be objects of dislike on the whole.’ “Good-for-self” will now mean object not merely of an independent liking, but of an independent and integral liking of the self—an “integral” liking being definable as one which is substantially consistent with the likings of the self as a whole.’16

Professor Campbell next points out that a further modification of the meaning of value-for-self arises at the same level of self-consciousness as the modification last considered—viz. that we restrict value-for-self to the objects of likings which are fairly permanent as well as independent and integral. But at the same time the self becomes aware of the possibility of its coming to have likings which it can foresee to be in a high degree integral and relatively permanent as well as independent; such as a liking for scientific pursuits or for music. The objects of such likings, the things we should like to like, are naturally therefore also recognized as things good for self.

So far, Professor Campbell has offered a very persuasive account of how an individual may naturally organize into an order of importance the objects of his various likings, on the ground of their independence, integralness, and relative permanence; and it is only surprising that he does not assign more weight to a characteristic of likings which he sets aside as of but slight importance, viz. their intensity.17 One would have thought that that should count for as much, in the establishing of the hierarchy, as the characteristics to which he has attached weight.

He next points out that we are aware that other selves have their likings, as well as ourselves, and suggests that by inter-subjective intercourse we discover what things are independent, integral, and relatively permanent objects of liking to other men as well as to ourselves, and come to think of them as not merely good for self, but good for man. And, finally, he holds that, having arrived at the conclusion that certain things are good for man, we drop the qualification and describe them as simply good, and come to think of them as if their goodness were intrinsic, i.e. flowed from or were consequent upon their own nature, independently of any relation to human likings.

I have already pointed out the difficulty that arises if we identify goodness with the being an object of liking to the whole or a majority of mankind or of any class of mankind. The difficulty is that we often call particular things—particular activities of the intelligence or of the imagination—good, when there is not present, even obscurely in the background of our mind, the thought that these particular activities have ever been contemplated by, still less been liked by, the whole or a majority of mankind or of any class of mankind. Professor Campbell tries to get over the difficulty by saying that they are thought of as objects of liking to human nature. But this way of putting the matter cannot be accepted. Human nature is a name for a certain set of powers and dispositions which we think of as common to all men. And no such set of powers and dispositions likes anything. What has likings is a particular man or particular men, and to say that something is an object of liking to human nature is only a loose way of saying that it is an object of liking to all or most men in virtue of their common human nature, or else that it is an object of liking to men in so far as they share in a nature which is regarded as normal or ideal human nature. Now if the first alternative is adopted, we are still faced with the difficulty that many particular things are judged good when we have not the slightest reason to believe that all or most men are even aware of their existence, still less that they like them. And if the second alternative is adopted, we are really falling back on the thought that there are certain things which, whether they are or are not liked by men, are worthy of being liked by them and would be liked by them if they had the ideal human nature in perfection. But as soon as we fall back on the notion that certain things are worthy of being liked, we are deserting the purely naturalistic account of goodness (other than moral goodness) for which Professor Campbell is arguing.

It appears to me that Professor Campbell might have put his case more plausibly if he had adopted a different line from that involved in the use of the phrase ‘object of liking to human nature’. The most obvious objection to saying that the statement ‘so-and-so is good’ means, when made explicit, that so-and-so is liked by all or most men, is that we constantly say of some particular activity of knowledge or of aesthetic imagination that it is good, when we do not think the particular activity in question is even being attended to or ever has been attended to by all or most men. But it might be suggested that in judging it to be good we are really saying that it is an instance of a kind of thing which we know or think to be an object of liking to all or most men. We know that knowledge, or the successful use of the intelligence, is liked by all or most men, and we therefore express admiration of a particular activity of knowledge or intelligence, though we do not suppose that all or most men have this particular activity before their minds at all. ‘Good’ in such an application would then mean ‘instance of a kind of thing which all or most men like with an independent, integral, and relatively permanent liking’.

The question must be asked, however, whether it is a true account of what we mean when we say that knowledge is good, to say that we mean that it is the object of such a liking to human nature. The alternatives must be pressed: to say that such-and-such a thing is an object of liking to human nature is either a merely historical, statistical statement based on a comparison of the actual likings of particular men, or there is involved in it an appeal to an ideal human nature. Take the first alternative. It must be first remarked that we know nothing either of the likings of man in the earliest stages of his evolution, or of what his likings will be in stages still to come. All that we can say is that there is considerable evidence that all or most men, during the period of human history of which we know something, have liked, for instance, knowledge. And all that we should be justified in saying on the basis of this is that within these limits of time knowledge has been good. We should have to admit that there may have been a time at which people disliked or were indifferent to knowledge, and that if so, knowledge was then bad and ignorance or error was good, or else all three were indifferent—and that such a time may come again in the future.

One might admit for the sake of argument that many of our admirations for particular types of intellectual activity—for the spinning of particular types of theory, for instance—rest upon no better basis than this. The individual finds that certain theories give him pleasure, and he discovers that most of his contemporaries who attend to them also get pleasure from them, and on that basis he judges them to be good. The fashion may change. One generation likes absolutist theories, another likes relativistic theories. One likes monistic theories, another likes dualistic or pluralistic ones. But as to the intrinsic preferability of knowledge to ignorance and error, of that sort of use of the intelligence which notices differences where they exist and identities where they exist, which draws from premisses only the conclusions that they warrant, have we not an a priori certainty that, whether or not all or most men always have liked and always will like these things—which we cannot possibly know—they are intrinsically better than their opposites, better worth having, more worthy of admiration, whether they receive it or not? And it must surely be admitted that, even if we often call one theory or way of thinking better on the ground of the actual preference of ourselves or of our generation, that which makes it really better, if it is so, is not in the least our preference, but its possession of such characteristics as I have suggested, the recognition of differences and of identities where they exist, the drawing of conclusions that are warranted by the premisses; in other words, its being of the nature of knowledge and not of mere opinion whether true or false.

If, on the other hand, we rest our judgement that certain things are good for man not on a historical and statistical study of the actual likings of individual men, but on the notion of a normal or ideal human nature, we are really saying of these things not that they are liked but that they are worthy of being liked, and are worthy of being liked because they are in themselves good.

Professor Campbell makes two claims for his account: (1) that it yields a list of goods for man which in fact agrees with the list of intrinsic goods that is suggested by those who adopt a non-relational view of goodness, and (2) that it accounts better than such a view does for the varieties of opinion that are held in different periods and in different communities as to what things are good.

The first of these claims is, I think, justified, and that it should be so need not surprise any one who holds the nonrelational view. For if there are things intrinsically good, and if the human mind has the power of apprehending their goodness, just as it has the power of apprehending other aspects of reality, it will naturally be satisfied by them when present, and attracted by the thought of them when absent, i.e. will ‘like’ them. And our liking for them will have the characteristics of the liking whose objects Professor Campbell's view identifies with things good for man. The liking for them will of course be an independent liking of them, a liking of them for their own sakes and not as a means to something else. Further, since it depends on an intellectual apprehension of their goodness, it will be a more integral liking than our likings for the pleasures of the senses. And for the same reason it will depend less on accidents of circumstance and will be a relatively permanent liking—not present always with equal strength, nor even present at all when we are absorbed in the pleasures or pains of the moment, but present as a permanent undercurrent of our interests. And further, since it is a liking for certain activities not as being enjoyed by us but as being what they are in themselves, its objects will naturally coincide with the things which Professor Campbell describes as being goods for man and not merely for self. Thus the coincidence between the lists of goods recognized on the relational and on the non-relational view is only what might be expected. In any case it cannot possibly furnish an argument for either view against the other.

The second claim is that the relational account explains better than the non-relational the varieties in the valuation of goods from age to age and from community to community. This claim I must resist. The holder of a non-relational view is not bound to hold that all the intrinsic goods he believes in must always be valued by all men, still less that they must always be placed in their true order of value. Here, as elsewhere, varieties of opinion are no indication that there is not an objective truth that is there to be apprehended. In the realm of natural science, for example, all sensible people agree that there is a completely objective truth to be apprehended, but we have no difficulty in reconciling this with the fact that different ages and different communities hold very different opinions. Different ages and different communities differ in their degree of mental maturity; each age and each community is liable to have prejudices and erroneous presuppositions of its own. To one age it seems self-evident that nature abhors a vacuum, and that natural species are fixed; to another neither of these presuppositions gives any satisfaction. It cannot really, I think, be contended that there is more variation between the opinions of different ages or communities about what things are good, than there is between their opinions about matters of natural science, where the laws of nature are admittedly objective and are unchanging.

In particular, Professor Campbell is on very dangerous ground when he thinks that virtue is intrinsically good and that knowledge is not. For surely the variations in the opinions of different ages about the ranking of the different virtues is more striking than the variations about the ranking of intellectual activities; and if variation were an argument against intrinsic goodness in the latter case it would be at least equally so in the former. The truth is that it is not an argument against the objective or non-relational view in either case, and that this view can give as good an account of varieties of opinion as the relational view can; I will not claim that it gives a better.

It seems to me, then, that knowledge, or perhaps we should rather say the activity of the mind which leads to knowledge, is good, not in the sense that human nature likes having it (although in fact most men do like having it), but in the sense that it is an admirable activity of the human spirit; that this activity owes its excellence not to our liking it, but to its being conducted according to its own proper principles, i.e. according to the principles discovered by logic; and that different instances of this activity are good in proportion as they are conducted according to these principles.

The main other good which Professor Campbell deals with is aesthetic experience. I think we should here distinguish—not that there is not some affinity between the two—between aesthetic enjoyment and artistic creation. The first is fundamentally a certain kind of pleasure (though it of course presupposes certain intellectual activities). The second is primarily a certain kind of mental activity (though no doubt the artist feels pleasure in his own activity). I will therefore reserve anything I have to say about the former till I come to discuss pleasure, and will consider now the creative activity of the artist. This, like knowledge, appears to me to be good not in the sense that we like it, but in the sense that it is an admirable activity of the human spirit; and it owes its goodness to its own intrinsic character. The characteristics that are the base or foundation of artistic excellence have not been worked out, and probably cannot be worked out, with anything like the precision with which the conditions of scientific excellence have been worked out by logic. Yet in a vague way we have some knowledge of the intrinsic features of good artistic work—vividness and breadth of imagination, vigour of execution, economy in the use of means, simplicity of plan. We think there is something admirable in these things, and it is for this reason that we honour the great artist. We think there is displayed in great art an activity of the human spirit which is admirable for its own sake, just as virtuous actions or the triumphs of the scientific mind are.

When we turn to consider whether, and if so in what sense, pleasure is good, we come to what is for me one of the most puzzling problems in the whole of ethics. The first point to which I would draw attention is that, while for the word ‘good’ when applied to moral dispositions and actions and to intellectual and artistic activities we can fairly substitute ‘admirable’ or at a lower level ‘commendable’, we cannot do this in the case of pleasant experiences, taken generally. There is nothing admirable or commendable in the mere feeling of pleasure. Another way in which the difference between good activities and pleasure is revealed is that, while we can call a man good, or at least admirable (for ‘good’ as applied to men tends to be limited to moral goodness) in respect of his moral actions and dispositions and in respect of his intellectual or artistic activities, any goodness that pleasure may be supposed to have is not in this way reflected on to its enjoyer. A man is not good in respect of the mere fact of feeling pleasure.

These facts suggest that one of two things must be true—either that pleasure as such is not good, or that it is good in some quite different way from that in which good activities are good.

And there is a further consideration which at least seems to point to the first alternative as being the true one. It is often assumed that if anything is good, there is an obligation to set ourselves to produce it, unless by an alternative act we can produce something better; and indeed it is a widely accepted view that productivity of good is the only duty. I have given reasons for holding that this view is not true—that there are other principles of duty, viz. that of fulfilling promises, that of making reparation for injuries done, and that of making a return for goods received. But I accept the principle that if something is good there is a prima facie obligation to produce it, and an actual obligation unless some more stringent prima facie obligation intervenes. Now there are two types of case in which it seems clear that we are under no prima facie obligation to produce pleasure. There are (1) pleasures that are themselves the manifestation of a bad moral nature, such as those of cruelty or of lust. It is clear that we not merely feel no prima facie obligation to produce them either for ourselves or for others. We feel a positive obligation to improve our own character, and so far as we can that of others, so as to prevent ourselves and them from having such enjoyments.

Now if this were all, it might be possible to modify the statement that pleasure is good, by saying ‘pleasures that are not manifestations of a bad moral nature are good’. But against this suggestion a fresh difficulty arises. There are (2) certain pleasures which, even when they are not the manifestations of a bad moral nature, we feel ourselves under no obligation to produce; we feel ourselves under no obligation to produce pleasures of any kind for ourselves. We feel ourselves, of course, under no obligation not to produce them, except when they are manifestations of a bad nature. But we feel ourselves under no obligation to produce even innocent pleasures for ourselves. That seems to me one of the clearest facts about our moral consciousness, though it is constantly overlooked by those who maintain both that pleasure as such is good and that there is an obligation to produce what is good.

I should perhaps say something here to substantiate two of the statements I have made or implied: (1) that we are conscious of an obligation to produce pleasure for others, and (2) that we are not conscious of an obligation to get pleasure for ourselves. (1) It is clear that the thought underlying a great many conscientious actions is the thought that by these actions pleasure will be produced for some one other than the agent. Some one might suggest that all that we feel bound to do is to refrain from producing pain for others, or to minimize their pain; and it is true that we feel these duties more acutely than the duty of positively promoting the pleasure of others. But it is surely plain, on reflection, that our sense of duty actually goes beyond this, and that we feel bound in the same sort of way, though not in the same degree, to maximize pleasure, as we feel bound to minimize pain, for others.

It is true again that much of the conscientious action which aims at producing pleasure is not actuated solely by the thought that pleasure will be produced for some one else, but also by the thought (a) that it will be produced for some one for whose well-being one has assumed a special responsibility (e.g. for one's children), or (b) that it will be produced for some one who has at present less than his due share of pleasure (e.g. for badly paid workers, or for sufferers from disease). In such cases there is involved (a) the thought of a duty to fulfil a promise, or an implicit promise, or (b) the thought of a duty to establish a just distribution of pleasure; and then the sense of a duty to produce the pleasure or remove the pain in question is greatly intensified. But I think it would on reflection be agreed that over and above these special obligations we have the sense of a duty to produce pleasure for others, just because it will be pleasure for them, and that if we had fulfilled all our promises, and if a just distribution of pleasures had already been established, there would still be a duty of going on to increase the amount of pleasure to be distributed.

(2) That we are conscious of no duty to maximize pleasure for ourselves seems to be so clear as not to need argument; and perhaps what is needed is rather some explanation of why the fact has been so much overlooked in ethical theory; it certainly is not overlooked in our natural thinking. The explanation is, I think, to be found in the history of the origin of Utilitarianism. Utilitarianism arose by critical reflection on Egoism. Bentham's sense of justice revolted against the monstrously privileged position which Egoism enjoined each individual to assign to his own pleasure, in his choice of action; but in revolting against a current system he made the mistake which has been repeated over and over again in the history of philosophy, the mistake of not questioning drastically enough the tenets of the current system. While asserting that the individual should aim at the pleasure of others, it seems not to have occurred to him to doubt the genuineness of the one obligation that Egoism had allowed. And the very method which he adopted for converting egoists to Utilitarianism, and which no doubt seemed the most hopeful at a time when Egoism was in the ascendant—that of arguing that it was reasonable for them to seek to achieve for others what they already sought to achieve for themselves—forced him to treat pleasure as something intrinsically good, and to ignore the very different ethical aspects which his own pleasure and any other man's pleasure present to any man. Perhaps only a generation for which the view that we should seek only our own pleasure is already out of date can see clearly that we are under no obligation to pursue our own pleasure at all.

Of the two types of case in which we are under no obligation to produce pleasure, an attempt might be made to explain the first without giving up the view that pleasure as such is good. With regard to immoral pleasures, it might be said that they are good qua pleasures but bad qua immoral, and that their badness qua immoral outweighs their goodness qua pleasant, and that that is the reason why we are never bound to produce them or aid in their production. To this, one who wishes to deny the goodness of pleasure as such might object, ‘Yes, but supposing such a pleasure were intensely pleasant but only slightly immoral, might it not then be our duty, on your showing, to produce it, since its goodness qua pleasant might well outweigh its badness qua immoral?’ But his opponent would have a sound reply. He could say, ‘Such a pleasure can be intensely pleasant only when it is intensely immoral; a man can enjoy cruelty intensely only if he intensely wishes to hurt another.’ If, however, both parties are agreed (as I think they ought to be) that in fact we are never under an obligation to produce immoral pleasures, they must be agreed that the goodness which springs from pleasantness is never so great as to outweigh the badness that springs from (or consists in) immorality; and this fact, as we shall see presently,18 constitutes a difficulty for any one who thinks that pleasures are good in the same sense in which moral dispositions and actions may be good.

What the last paragraph has shown is that we seem quite incapable of equating, in respect of goodness, any amount of pleasure with any amount of morally good action. I suggested in The Right and the Good19 that while both virtue and pleasure have places on the same scale of goodness, virtue begins at a higher point than that at which pleasure leaves off, so that any, even the smallest, amount of virtue is better, and more worth bringing into existence, than any, even the greatest, amount of pleasure. But I now see this (and I should have seen it earlier) to be impossible. If virtue were really on the same scale of goodness as pleasure, then pleasure of a certain intensity, if enjoyed by a sufficiently large number of persons or for a sufficient time, would counterbalance virtue possessed or manifested only by a small number or only for a short time. But I find myself quite unable to think this to be the case; and if I am right in this, it follows that pleasures, if ever good, must be good in a different sense from that in which good activities are so.

Now, however, we must turn to the facts which point to pleasure being under certain limitations a good thing. We do consider the state of pleasure, when the pleasure is not a morally bad pleasure, to be in some sense a better state of affairs than the state of pain; and we feel ourselves under a certain obligation to produce it for other people, when it is not a morally bad pleasure, and still more to prevent or minimize pain, when it is not a morally good pain (such as pain at the misfortune of another). And this is not merely because every one likes pleasure and dislikes pain; for vicious people like vicious pleasures, yet we feel ourselves under no obligation to help them to get these. Besides being liked by the persons who have them, pleasures that are not vicious have the further characteristic of being worthy objects of satisfaction for an observer, and perhaps that is the sense in which we should say that they are good, as we attempted to specify the sense in which good actions are good by saying that they are worthy objects of admiration.

The point is not that they are actually objects of satisfaction; for vicious activities may easily be objects of satisfaction to those who engage in them, but we do not for that reason call them good. The point is that we think of our satisfaction in seeing people innocently happy as a justified satisfaction; as we should most certainly think of dissatisfaction at seeing people innocently happy as an unjustified dissatisfaction.

It is worth while to point out that the satisfaction of which the pleasures of other people may be objects, and of which they are worthy to be objects, is quite different in its nature from the satisfaction which a man may feel in his own pleasant experiences. It is a sympathetic satisfaction, and sympathy by its nature must be of one man with another, and cannot be felt by a man for himself. Not only is sympathetic satisfaction different in its object from the other kind of satisfaction; it is different in its whole ‘feel’.

If we give as the reason which makes it a duty for a man to produce pleasures for other people, and not a duty to produce pleasures for himself, the fact that the former and not the latter are proper objects of satisfaction to him, we must be careful to avoid two misunderstandings which might arise, (1) We might be thought to mean that he ought to produce pleasures for other people in order to get sympathetic satisfactions for himself. That would of course be a complete misstatement of what we really think about our duty to produce pleasure, or to minimize pain, for other people. It is for their sake that we feel bound to act so, not for our own. That is why I have described the pleasures of other people as objects of satisfaction, not as sources of satisfaction; if we described them in the latter way, we should be treating them as means to the satisfaction, which is just not how we regard them when we feel ourselves bound to produce them.

But (2) the sympathetic satisfactions which we get from increasing other people's pleasure or diminishing their pain are not only satisfactions; they are manifestations of a morally good nature; and it might be suggested that it is our duty to increase the pleasure of others or to diminish their pain, because in or by doing so we bring into being these manifestations of a good nature in ourselves. This is plainly wrong for two reasons, (i) One is that which I have used to refute the former misunderstanding, viz. that it is plainly for the sake of those whose pleasure we increase or whose pain we diminish, and not with a view to bringing about any change in our own state, that we feel bound to act so. (ii) The other is that a morally good nature is just as much manifested in dissatisfaction with the pain, or lack of pleasure, of other people as in the satisfaction which we get from increasing their pleasure or diminishing their pain. When we remove the pain of another, we produce in ourselves merely the substitution of a morally good satisfaction—with his pleasure—for a dissatisfaction with his pain which is of exactly the same moral worth, so that from that point of view nothing is gained by the exchange.

We ought to consider at this point a view which might be put forward with regard to the fact that we are never conscious of a duty to get pleasure or avoid pain for ourselves, as we are conscious of a duty to give pleasure to or prevent pain for others. It might be said that both types of action are right, but that only the latter is obligatory, because in the former case there is no possibility of a moral conflict, since in it our natural desire inevitably prompts us to do that which it is right for us to do—to seek our own pleasure. I do not think that this suggestion can be accepted; for (1) the act of seeking pleasure for oneself is not merely not obligatory, but has not even the specific kind of rightness or fitness which is moral fitness. It seems morally entirely colourless. It is not blameworthy, except when it involves the omission of some duty, and it is never morally praiseworthy. But (2) even if it were, the explanation offered of its not being felt to be obligatory does not seem to meet the case. For it often happens that there is a perfectly natural tendency to seek to give pleasure to some other person, which is just as strong as is in most people the tendency to seek pleasure for themselves. This is noticeably so in maternal love. Yet no one would say that because a mother naturally seeks the happiness of her children she has no duty to seek it. She will very likely be led directly by natural affection to seek their happiness, without stopping to ask whether it is her duty. But any disinterested spectator would say that it is her duty, and she herself would agree if she stopped to ask the question. She would not say, ‘it is my pleasure and therefore not my duty’, but rather ‘it is both my duty and my pleasure’.

What light do these considerations throw on the question whether the goodness of the main things that are commonly called good—let us say virtuous action, intelligent thinking, and pleasure—is a quality intrinsic to them, or a relational characteristic, consisting in their standing in a certain relation to something else? When some entity is commonly referred to by an adjective, there is a certain presumption that it is a quality, just as, when it is commonly referred to by a prepositional phrase, there is a presumption that it is a relation or a relational property. An entity commonly referred to by an adjective may reasonably be supposed to be a quality, unless the adjective can be seen, as many adjectives can be seen, to be replaceable by a prepositional phrase. Now in describing some of the things commonly called good as fit objects of admiration, and others as fit objects of satisfaction, I have used prepositional phrases; and it is proper to inquire whether that amounts to saying that goodness is at bottom a relational property. The phrase ‘worthy of admiration’, it appears to me, does not justify the conclusion that the goodness which is so described is a relational property. For admiration is not a mere emotion; it is an emotion accompanied by the thought that that which is admired is good. And if we ask on what ground a thing is worthy of being thought to be good, only one answer is possible, namely that it is good. It would be absurd to say that a thing is good only in the sense that it is worthy of being thought to be good, for our definition of ‘good’ would then include the very word ‘good’ which we were seeking to define. I have tried to call attention to the difference between certain things commonly called good and certain others commonly called good, by calling attention to the fact that admiration is appropriate to the one and not to the other; I have not been trying to define the sense in which the one class are good, but to call attention to a fact which implies that their goodness is an intrinsic quality of them.

The same is not true of the phrase ‘fit object of satisfaction’. While admiration includes or involves the thought that the thing admired is good independently of our admiring it, satisfaction does not include or involve the thought that that in which we take satisfaction is good independently of our satisfaction. We often take satisfaction in things that we do not think good, but only pleasant. And while it is self-evident that the only ground on which a thing is worthy of admiration is that it is good in itself, it is not self-evident that the only ground on which a thing is worthy of our interest or liking is that it is good in itself.

We may now try to put in a clearer form the fact which has so far been expressed by saying that the innocent pleasures of one man are for any other man a worthy object of satisfaction. This is plainly only another way of saying that satisfaction taken by one man in the innocent pleasure of another is morally suitable, or right; and this is a preferable way of putting the matter because, instead of introducing the new and not altogether clear notion of worthiness, it defines the goodness of innocent pleasures by using a notion which has already been recognized as fundamental in ethics, the notion of rightness.

I suggest, therefore, that the sense in which from the point of view of any man the innocent pleasures of another are good is that it is right for him to feel satisfaction in them.

The account I have given of this sense of ‘good’, though it has been suggested to me not by Brentano's doctrine but by direct reflection on the facts, clearly has a close affinity with Brentano's doctrine that ‘good’ always means ‘object of a love that is right’; and it is proper that some comment should be offered on Brentano's doctrine. The Brentano school holds that ‘good’ belongs to a class of merely apparent predicates. The nature of the theory can perhaps best be seen by noting the analogy which they hold to exist between the terms ‘good’ and ‘possible’.20 That a thing, e.g. a spherical body, is possible is, they maintain, a consequence of its constitutive characteristics, and is not one of them. We call a thing possible when we think it not in itself impossible. And we call a thing impossible when and only when we reject it apodeictically (i.e. when we say ‘there cannot be such a thing’); therefore we call a thing possible when and only when we reject apodeictically an apodeictic rejection of it (i.e. when we say ‘we cannot say that there cannot be such a thing’); and this is the meaning of ‘possible’. The rejection of the rejection of a spherical body is based simply on the consideration of the conception of a sphere; and thus ‘possible’, while not a real predicate of a spherical body, is a direct consequence of its real predicates. In the same way we see, by attending to the conception of pleasure, that an emotion directed towards it and itself characterized as right cannot be other than love; and to see this is to see that pleasure is good. Thus goodness, while it is an ‘irreal determination’, is consequent on the real characteristics of that which is good.

With one of the main theses of this theory, viz. with its assertion that goodness is not a constitutive characteristic but is grounded on the real characteristics of that which is good, I am in complete agreement, and I may be allowed to refer to a passage of The Right and the Good21 in which I have argued for this view. But there are other features of the Brentano theory which do not appear to me to be correct.

In the first place, it seems to be a mistake to suppose that ‘good’, ‘bad’, ‘possible’, ‘impossible’, ‘existent’, ‘non-existent’, form a class of Scheinqualitäte consisting in relations to certain mental activities ‘characterized as right’. Consider the notion of ‘possible’ (as applied, for instance, to a square) and of ‘impossible’ (as applied, for instance, to a round square). It is of course true that the apodeictic rejection of a square (i.e. the statement ‘there cannot be a square’) is apodeictically rejected by a right act of thought (i.e. if we think rightly we see that we cannot rightly say that there cannot be a square). But the square is not possible because a right act of thought rejects the rejection of it; a right act of thought rejects the rejection of it because the square is possible. Our thought that a square is possible can be right only if and because there is a real relation of compatibility between the attributes of equal-sidedness and equal-angledness in a quadrilateral. Our thought that a round square is impossible can be right only if and because there is a real relation of incompatibility between roundness and squareness. The Brentano school is no doubt justified in regarding the judgements ‘A exists’, ‘A does not exist’, ‘A is possible’, ‘A is impossible’ as being logically very different from judgements in which some ordinary attribute like ‘red’ or ‘loud’ is predicated, but their introduction of acceptance or rejection by right thought as being what the judgements mean seems to me mistaken; and modern logic has found a much more satisfactory account of the meaning of such judgements, when, for instance, it points out that a judgement of possibility is really a judgement of compatibility, and a judgement of impossibility really a judgement of incompatibility.

The theory about the nature of goodness is therefore deprived of any support which it might be supposed to derive from being able to class predications of goodness with other judgements which, while they seem to be about objects, are really about activities of mind directed towards these objects. But it might still be a true theory about goodness, though the corresponding theories about existence and possibility are false. The first criticism I would offer of the theory of goodness is that in defining goodness as ‘being the object of a love which is right’, it fails to distinguish between the two attitudes which I have called admiration and satisfaction. One has, it seems to me, only to reflect for a very little on one's attitude towards a brave act or a fine intellectual effort, and towards a sensuous pleasure, to see how very different the two attitudes are; or rather, since there is satisfaction in both cases, how completely the element of admiration is lacking in the latter case. And further, while satisfaction at another's pleasure is simply a feeling, not involving the thought that the other's pleasure is good in itself, but only the thought that he is being pleased, admiration involves the thought that that which is admired is good in itself. If I am right in giving this account, nothing can be a worthy object of admiration—it cannot be right to admire it—unless it is also good in itself; while the pleasures of others are good, from the point of view of any man, simply in the sense that it is right for him to take satisfaction in them. Thus Brentano's theory seems to be true of ‘good’ in one of its senses, though not true of it in the other; and in so far as it is true, it is very important.

Finally, however, the theory seems to be wrong in saying that a man's own pleasures are, from the point of view of any man, good in the same sense in which the pleasures of others are. For while we can see the rightness, the moral suitability, of his taking satisfaction in the latter, we can see no moral suitability in his taking satisfaction in the former. Or again, to be glad at the pain of another is wrong; to be glad at one's own pain is either impossible, or if possible merely silly.

If our contentions are right, ‘good’ in its first sense is a nonrelational attribute; ‘good’ in its second sense is a relational attribute, but while our account of it is a relational one, it is not a naturalistic one, since it defines good in its second sense by reference to ‘right’.

To sum up the results we have arrived at: Certain moral dispositions and actions, and certain activities of the intellect and of the creative imagination, appear to be good in a way which depends entirely on their intrinsic nature, on the first being conscientious or benevolent, for instance, or on the second being logical or having the characters, harder to specify, that make artistic activity good. These things are good in a sense which is indefinable, but which may be paraphrased by saying that they are fine or admirable activities of the human spirit, and by adding that they are good in such a way that any one who has them or does them is to that extent being good himself. Pleasure is never good in this, which I should call the most proper sense of ‘good’. But the pleasures of others (except those which are immoral) are good in a secondary sense, viz. that they are morally worthy or suitable objects of satisfaction. Things that are good in the first and most proper sense we have, by a self-evident necessity, a prima facie duty to produce, to the best of our ability, irrespective of whether it is ourselves or others that are going to have or do them. Things that are good in the secondary sense, i.e. the pleasures of others, are also things that we have a duty to produce. It should be added that things which are good in the first sense are also good in the second. Activities that are good in themselves are necessarily worthy objects of satisfaction, and are thus doubly good.

If these are really two different senses of ‘good’, things that are good in the different senses do not fall on the same scale of goodness and are not comparable in respect of goodness. If they fell on the same scale, and if the duty to produce one rather than the other depended on which was the better, the prima facie duty of producing some good activity in another person would always be outweighed by the prima facie duty of producing pleasure, if the quantity of pleasure were to be sufficiently great (e.g. if it were to be enjoyed by a sufficient number of people). The natural moral consciousness finds it very hard to believe that any amount of pleasure can thus outweigh a given good activity in goodness;22 and the recognition of two senses of goodness has vindicated the natural moral consciousness. We are still free to believe that the prima facie duty of producing what is intrinsically good always takes precedence over the prima facie duty of producing pleasure for others.

At the same time, things that are good in a single sense will be comparable in respect of goodness. It will be a legitimate question whether in any given situation it is rather our duty to promote some good moral activity, or some good intellectual activity, in ourselves or others; and in deciding which we ought to do we have to rely on our very fallible apprehension of the degrees of goodness belonging to each. And if it seems paradoxical to say that a good moral activity is comparable with a good intellectual activity in respect of goodness, it is at least a paradox not peculiar to the view I have put forward; the theory of ideal Utilitarianism also contains it, and adds the greater paradox of regarding pleasure also as falling on the same scale of goodness.

For any man, his own actual pleasures are not good in either of these senses, and his imagined future pleasures are not imagined to be good in either of them; therefore the duty of producing good involves no duty of producing pleasures for himself. Yet it is natural enough, and it has been habitual in most ethical theories, to call them good. It is, however, improper to call them so. For in the proper use of any word (to recur to Meinong's distinction)23 it is used to signify something about that to which it is applied, besides expressing a mental attitude towards that thing; while in calling our own pleasures good we are, it would seem, only expressing our enjoyment of them when we have them and our attraction towards them when we have not got them.

The two proper kinds of good also have, for any one who recognizes their goodness, this attractive character; and this attractive character, or (as it has sometimes been expressed) the fact that we have a pro-attitude towards them, seems to be all that is common to these three kinds of thing that are habitually called good. Is this, then, the original usage of the word, was it originally a mere interjection expressive of attraction, and has it come to have its two significances (as opposed to its expressiveness) as men have by reflection come to see, in some of the things to which they were attracted, that they were more than attractive, that they were worthy objects of admiration, or worthy objects of satisfaction? The suggestion is plausible, but it is opposed by the grammatical form of the word—by its being an adjective. So far as I know, there is no evidence of the origin of the word ‘good’ from some primitive interjectional form, such as might have been a mere expression of attraction. Unless an inquiry by comparative philologists should discover an interjectional origin of the word, which does not seem at all likely, it seems not improbable that the word started by expressing admiration (which includes the thought that the person or thing admired is good in itself) and that it was by a sort of degeneration that it has come to have its other types of application. This leads naturally on to a further inquiry. One of the great puzzles of ethical theory lies in the sense we have of obligations to do certain things which do not seem likely to bring into being the greatest possible amount of any of the generally recognized personal goods, either in the way of good moral or intellectual activities or in the way of pleasure. We feel an obligation to do a promised service to another, far greater than the obligation we feel to do him an unpromised service, and that even when we cannot foresee any more distant personal goods which will be brought into being by our action. Similarly, we feel an obligation to make reparation for wrongs we have done, and return for benefits we have received, even when we do not think we shall be bringing more good into being for the person we have wronged or the person whose services we have accepted, than we could bring into being for some other person by an alternative act. And we feel an obligation to do justice as between different people, even when we do not think the sum of goods either moral or intellectual or hedonistic will be increased thereby. The force of this last consideration can be most easily seen by noting the facts that even the most convinced utilitarians have recognized the duty of dividing pleasure justly between man and man, even when the sum of pleasures to be produced is not increased thereby, and that some of them have recognized the duty of doing so even when the sum of pleasures to be produced in this way is less than that which would be produced by an unjust distribution.

The question that faces us is whether we may not be able to account for these facts consistently with Utilitarianism by supposing that in all these cases there is some different kind of good that is created by our action, and that that is why we ought to do the action. These other goods might in general be called situational goods.24 They would not be activities or enjoyments resident in individuals, but would involve relations between individuals. Their nature will be seen more clearly by pointing to the several instances. The suggestion would be that I ought to fulfil promises because the receipt of a service by a person to whom it has been promised is a situational good which I can bring into being by fulfilling my promise and shall fail to bring into being if I do not fulfil it; that I ought to make reparation for injuries I have done because the receiving of reparation by one who has been wronged is a similar situational good; that I ought to make a return for services I have received because the enjoyment of services in return for services is again a situational good; that I ought to do justice as between man and man because the enjoyment of happiness in proportion to merit is a situational good, over and above the good which consists in the meritorious character or its activities, and that which consists in the happiness. All of these situational goods would be goods not in the sense of being worthy objects of admiration, but in the sense of being worthy objects of satisfaction, just as for any man the pleasures of other people are.

It is to be observed that initially quite a different account of the matter might be given. It might be said that the suggestion just made in every case puts the cart before the horse—that it is not true that we ought to produce pleasure for other people because the pleasure of other people is a worthy object of satisfaction, but rather that it is a worthy object of satisfaction because we ought to produce it; that it is not true that promises should be kept because the reception of promised services is a worthy object of satisfaction, but that the reception of promised services is a worthy object of satisfaction because promises ought to be kept; and so on in the other cases. To decide between the two views, we must consider each of these branches of duty on its own merits. The question seems to me a difficult one, but I will answer it to the best of my ability.

Let us start with the duty of promoting the pleasure of other people, and the still more obvious duties of not causing pain to other people, and of diminishing their pain; for brevity I will use the phrase ‘the duty of promoting the pleasure of others’ as covering all these duties. It seems to me clear that the pleasure of other people is a worthy object of satisfaction to any man. And it is to be observed that a good man takes satisfaction in the pleasure of others quite independently of any judgement that any one has done his duty in causing that pleasure, and is dissatisfied at the pain of other people quite independently of any judgement that any one has done wrong in causing this pain. He feels a satisfaction at the mere existence of the pleasure, a dissatisfaction at the mere existence of the pain, however it has been caused, apart from any satisfaction or dissatisfaction he may feel at the way it has been caused. And again, a good man's satisfaction at the pleasure of others, or dissatisfaction at their pain, is independent of the thought that he ought to increase the pleasure or diminish the pain of others; he may feel the satisfaction or dissatisfaction before he becomes conscious of the duty, and if there is no obvious means by which he could increase their pleasure or diminish their pain, he may feel the satisfaction or dissatisfaction without coming to be aware of the duty at all. Thus it may, I think, certainly be said that a good man's taking of satisfaction in the pleasure of others is independent of any thought of duty. Two questions, however, remain: (1) Is the fact that the pleasure of others is a worthy object of satisfaction the objective basis of our duty to bring it into being? and (2) Is the thought that the pleasure of others is a worthy object of satisfaction the subjective ground of our thinking we have a duty to bring it into being? The answer to the first question seems to me to be Yes. And if so, the duty of trying to produce pleasure for others will fall under the same general principle as the duty of trying to promote good activities; it will be grounded on the goodness of the result to be produced—though the two results are good in different senses.25 The answer to the second question must be more qualified. I do not think that a good man formulates explicitly the dictum ‘the pleasure of others is a worthy object of satisfaction’ before he feels the duty to bring it into being. The position rather is that he in fact feels satisfaction at their pleasure or dissatisfaction at their pain, and feels, rather obscurely, that his interest in the pleasure of others is something that he can with moral safety follow—a feeling which he never has about his interest in his own pleasures. If this be so, it is an implicit awareness that the pleasure of others is good, in the sense of being a worthy object of interest, that becomes the ground of the sense of a duty to produce it.

Let me turn now to the duty of distributing pleasures among others in proportion to their goodness. Here, again, it seems that a good man takes satisfaction in finding goodness rewarded, independently of the thought that it was any one's duty to produce this situation, and independently of the thought that he ought to do what he can to effect in other cases the rewarding of goodness. He will take satisfaction in the happiness of the virtuous, and dissatisfaction in their unhappiness, even when he thinks this has been produced as the result of natural laws and not of moral action. And as with his interest in the pleasure of others in general, so with his special interest in the happiness of the virtuous, he feels obscurely that this interest is one that can be trusted. His sense of a duty to act justly seems, I think, to be properly said to rest on an obscure sense that the happiness of the virtuous is a good in the sense of being a morally worthy object of interest.

Turn next to the duties of making reparation for wrongs we have done, and of making a return for benefits we have received. Here again, a morally good spectator will find satisfaction in seeing these things take place; but in this case the satisfaction seems to me to depend on the previous thought that it was A's duty to make such compensation to B. It is not simply B's acquiring of a certain advantage or pleasure that a morally good spectator feels to be a worthy object of satisfaction; if this were so, we might say that A's duty to make compensation arises from the fact that what he thereby produces is good, in the sense of being a worthy object of satisfaction. What a morally good spectator thinks to be a worthy object of satisfaction is B's getting the advantage or pleasure by A's action, by A's giving it to him; and that thought rests upon the prior thought that B has a right to get it from A, or in other words that A has a duty to give it to him. The spectator's primary thought is that A by doing an injury to B or by accepting a benefit from him has by his own act put himself under a moral obligation to B, and any satisfaction the spectator feels at A's fulfilling the obligation presupposes the thought that there is an obligation, and is not presupposed by it.

And similarly it seems clear to me that, while a good man will feel satisfaction at a second man's fulfilling his promise to a third, that satisfaction presupposes the thought that the promiser has, by making the promise, put himself under an obligation to the promisee. In this case also, therefore, it appears that the rightness of the act does not depend on the goodness of the result produced, even if we admit that the result produced is good, in the sense of being a worthy object of satisfaction. The rightness of the act will, as in the cases of reparation for wrongs and return for benefits, depend on the nature of the result to be produced, but not on its goodness, since it is good only because there is a duty to produce it.

  • 1.

    Eth. Nic. 1096 a 17–29.

  • 2.

    Ibid, b 25–9.

  • 3.

    J. N. Findlay, Meinong's Theory of Objects, 28; cf. Meinong, Über Annahmen, ed. 2, 24 f.

  • 4.

    Findlay, op. cit. 61.

  • 5.

    The Right and the Good, 65–6.

  • 6.

    Ethics, 73–6.

  • 7.

    pp. 75–104.

  • 8.

    These alternatives, and to a large extent my discussion of them, are borrowed from Professor Moore's discussion of subjectivist views of the meaning of ‘right’ (Ethics, 87–132).

  • 9.

    Cf. pp. 254–5, above.

  • 10.

    View (1), p. 258 fin.

  • 11.

    View (2), p. 259.

  • 12.

    View (3), p. 259.

  • 13.

    In Mind, xliv (1935), 273–99.

  • 14.

    Ibid. 279.

  • 15.

    Ibid. 283.

  • 16.


  • 17.

    Ibid. 283.

  • 18.

    p. 275.

  • 19.

    p. 150.

  • 20.

    G. Katkov, Werttheorie und Theodizee, 147 f.

  • 21.

    pp. 121–3.

  • 22.

    Cf. p. 275.

  • 23.

    Cf. pp. 254–5.

  • 24.

    I take the phrase from N. Hartmann, Ethik2, 236 (Eng. tr. ii. 31). The German is Sachverhaltswerten (state-of-affairs values).

  • 25.

    Viz. those pointed out on pp. 271–6, 278–9.

From the book: