My first task will be to draw a logical distinction between two sorts of adjectives, suggested by the distinction between attributive adjectives (e.g. a red book) and predicative adjectives (e.g. this book is red); I shall borrow this terminology from the grammars. I shall say that in a phrase an A B (A being an adjective and B being a noun) A is a (logically) predicative adjective if the predication is an A B splits up logically into a pair of predications is a B and is A; otherwise I shall say that A is a (logically) attributive adjective. Henceforth I shall use the terms predicative adjective and attributive adjective always in my special logical sense, unless the contrary is shown by my inserting the adverb grammatically. (¶ 1)

There are familiar examples of what I call attributive adjectives. Big and small are attributive; x is a big flea does not split up into x is a flea and x is big, nor x is a small elephant into x is an elephant and x is small; for if these analyses were legitimate, a simple argument would show that a big flea is a big animal and a small elephant is a small animal. Again, the sort of adjective that the mediaevals called alienans is attributive; x is a forged banknote does not split up into x is a banknote and x is forged, nor x is the putative father of y into x is the father of y and x is putative. On the other hand, in the phrase a red book red is a predicative adjective in my sense, although not grammatically so, for is a red book logically splits up into is a book and is red. (¶ 2)

I can now state my first thesis about good and evil: good and bad are always attributive, not predicative, adjectives. this is fairly clear about bad because bad is something like an alienans adjective; we cannot safely predicate of a bad A what we predicate of an A, any more than we can predicate of a forged banknote or a putative father what we predicate of a banknote or a father. We actually call forged money bad; and we cannot infer e.g. that because food supports life bad food supports life. For good the point is not so clear at first sight, since good is not alienans--whatever holds true of an A as such holds true of a good A. But consider the contrast in such a pair of phrases as red car and good car. I could ascertain that a distant object is a red car because I can see it is red and a keensighted but colour-blind friend can see it is a car; there is no such possibility of ascertaining that a thing is a good car by pooling independent information that it is good and that it is a car. This sort of example shows that good like bad is essentially an attributive adjective. Even when good and bad stands by itself as a predicate, and is thus grammatically predicative, some substantive has to be understood; there is no such thing as being just good or bad, there is only being a good or bad so-and-so. (If I say that something is a good or bad thing, either thing is a mere proxy for a more descriptive noun to be supplied from the context; or else I am trying to use good or bad predicatively, and its being grammatically attributive is a mere disguise. The latter attempt is, on my thesis, illegitimate.) (¶ 3)

We can indeed say simpliciter A is good or A is bad, where A is a proper name; but this is an exception that proves the rule. For Locke was certainly wrong in holding that there is no nominal essence of individuals; the continued use of a proper name A always presupposes a continued reference to an individual as being the same X, where X is some common noun; and the X expresses the nominal essence of the individual called A. Thus use of the proper name Peter Geach presupposes a continuing reference to the same man; use of the Thames a continuing reference to the same river; and so on. In modern logic books you often read that proper names have no meaning, in the sense of meaning in which common nouns are said to have meaning; or (more obscurely) that they have no connotation. But consider the difference between the understanding that a man has of a conversation overheard in a country house when he knows that Seggie stands for a man, a Highland stream, a village, or a dog. In the one case he knows what Seggie means though not whom; in the other case he does not know what Seggie means and cannot follow the drift of the conversation. Well, then if the common noun X expresses the nominal essence of the individual called A; if being the same X is a condition whose fulfilment is presupposed by our still calling an individual A; then the meaning of A is good/bad said simpliciter, will be A is a good/bad X. E.g. if Seggie stands for Seggie is a good man, though context might make it mean Seggie is a good deerstalker, or the like. (¶ 4)

The moral philosophers known as Objectivists would admit all that I have said as regards the ordinary uses of the terms good and bad; but they allege that there is an essentially different, predicative use of the terms in such utterances as pleasure is good and preferring inclination to duty is bad, and that this use alone is of philosophical importance. The ordinary uses of good and bad are for Objectivists just a complex tangle of ambiguities. I read an article once by an Objectivist exposing these ambiguities and the baneful effects they have on philosophers not forewarned of them. One philosopher who was so misled was Aristotle; Aristotle, indeed, did not talk English, but by a remarkable coincidence ἀγαθός had ambiguities quite parallel to those of good. Such coincidences are, of course, possible; puns are sometimes translatable. But it is also possible that the uses of ἀγαθός and good run parallel because they express one and the same concept; that this is a philosophically important concept, in which Aristotle did well to be interested; and that the apparent dissolution of this concept into a mass of ambiguities results from trying to assimilate it to the concepts expressed by ordinary predicative adjectives. It is mere prejudice to think that either all things called good must satisfy some one condition, or the term good is hopelessly ambiguous. A philosopher who writes off most of the uses of good as trivial facts about the English language can, of course, with some plausibility, represent the remaining uses of good as all expressing some definite condition fulfilled by good things--e.g. that they either contain, or are conducive to, pleasure; or again that they satisfy desire. Such theories of goodness are, however, open to well-known objections; they are cases of the Naturalistic Fallacy, as Objectivists say. The Objectivists' own theory is that good in the selected uses they leave to the word does not supply an ordinary, natural, description of things, but ascribes to them a simple and indefinable non-natural attribute. But nobody has ever given a coherent and understandable account of what it is for an attribute to be non-natural. I am very much afraid that the Objectivists are just playing fast and loose with the term attribute. In order to assimilate good to ordinary predicative adjectives like red and sweet they call goodness an attribute; to escape undesired consequences drawn from the assimilation, they can always protest, Oh no, not like that. Goodness isn't a natural attribute like redness and sweetness, it's a non-natural attribute. It is just as though somebody thought to escape the force of Frege's arguments that the number 7 is not a figure, by saying that it is a figure, only a non-natural figure, and that this is a possibility Frege failed to consider. (¶ 5)

Moreover, can a philosopher offer philosophical utterances like pleasure is good as an explanation of how he means good to be taken in his discussions? Forget the uses of good in ordinary language says the Objectivist; in our discussion it shall mean what I mean by it in such typical remarks as pleasure is good. You, of course, know just how I want you to take these. No, of course I cannot explain further: don't you know that good in my sense is a simple and undefinable term? But how can we be asked to take for granted at the outset that a peculiarly philosophical use of words necessarily means anything at all? Still less can we be expected at the outset to know what this use means. (¶ 6)

I conclude that Objectivism is only the pretence of a way out of the Naturalistic Fallacy: it does not really give an account of how good differs in its logic from other terms, but only darkens counsel by words without knowledge. (¶ 7)

What I have said so far would meet with general approval by contemporary ethical writers at Oxford (whom I shall henceforth call the Oxford Moralists); and I now have to consider their positive account of good. They hold that the features of the term's use which I have described derive from its function's being primarily not descriptive at all but commendatory. That is a good book means something like I recommend that book or choose that book. They hold, however, that although the primary force of good is commendation there are many cases where its force is purely descriptive--Hutton was batting on a good wicket, in a newspaper report, would not mean What a wonderful wicket Hutton was batting on. May you have such a wicket when you bat. The Oxford Moralists account for such cases by saying that here good is, so to say, in quotation marks: Hutton was batting on a good wicket, i.e. a wicket such as cricket fans would call good, i.e. would commend and choose. (¶ 8)

I totally reject this view that good has not a primarily descriptive force. Somebody who did not care two pins about cricket, but fully understood how the game worked (not an impossible supposition), could supply a purely descriptive sense for the phrase good batting wicket regardless of the tastes of the cricket fans. Again if I call a man a good burglar or a good cut-throat I am certainly not commending him myself; one can imagine circumstances in which these descriptions would serve to guide another man's choice (e.g. if a commando leader were choosing burglars and cut-throats for a special job), but such circumstances are rare and cannot give the primary sense of the descriptions. It ought to be clear that calling a thing a good A does not influence choice unless the one who is choosing happens to want an A; and this influence on action is not the logically primary force of the word good. You have ants in your pants, which obviously has a primarily descriptive force, is far closer to affecting action than many uses of the term good. And many uses of the word good have no reference to the tastes of a panel of experts or anything of the sort; if I say that a man has a good eye or a good stomach my remark has a very clear descriptive force and has no reference to any panel of eye or stomach fanciers. (¶ 9)

So far as I can gather from their writings, the Oxford Moralists would develop two lines of objection against the view that good has a primarily descriptive force. First, if we avoid the twin errors of the Naturalistic Fallacy and of Objectivism we shall see that there is no one description, natural or non-natural, to which all good things answer. The traits for which a thing is called good are different according to the kind of thing in question; a knife is called good if it is UVW, a stomach if it is XYZ, and so on. So, if good did have a properly descriptive force this would vary from case to case: good applied to knives would express the attributes UVW, good as applied to stomachs would express the attributes XYZ, and so on. If good is not to be merely ambiguous its primary force must be taken to be the unvarying commendatory force, not the indefinitely varying descriptive force. (¶ 10)

This argument is a mere fallacy; it is another example of assimilating good to ordinary predicative adjectives, or rather it assumes that this assimilation would have to be all right if the force of good were descriptive. It would not in fact follow, even if good were an ordinary predicative djective, that if good knife means the same as knife that is UVW, good means the same as UVW. UVW. Triangle with all its sides equal means the same as triangle with three sides equal, but you cannot cancel out triangle and say that with all its sides equal means the same as with three sides equal. In the case of good the fallacy is even grosser; it is like thinking that square of means the same as double of because the square of 2 means the same as the double of 2. This mathematical analogy may help to get our heads clear. There is no one number by which you can always multiply a number to get its square: but it does not follow either that square of is an ambiguous expression meaning sometimes double of, sometimes treble of, etc., or that you have to do something other than multiplying to find the square of a number; and, given a number, its square is determinate. Similarly, there is no one description to which all things called good so-and-so's answer; but it does not follow either that good is a very ambiguous expression or that calling a thing good is something different from describing it; and given the descriptive force of A, the descriptive force of a good A does not depend upon people's tastes. (¶ 11)

But I could know what good hygrometer meant without knowing what hygrometers were for; I could not, however, in that case be giving a definite descriptive force to good hygrometer as opposed to hygrometer; so good must have commendatory not descriptive force. The reply to this objection (imitated from actual arguments of the Oxford Moralists) is that if I do not know what hygrometers are for, I do not really know what hygrometer means, and therefore do not really know what good hygrometer means; I merely know that I could find out its meaning by finding out what hygrometers were for--just as I know how I could find out the value of the square of the number of the people in Sark if I knew the number of people, and so far may be said to understand the phrase, the square of the number of the people in Sark. (¶ 12)

The Oxford Moralists' second line of objection consists in first asking whether the connexion between calling a thing a good A and advising a man who wants an A to choose this one is analytic or empirical, and then developing a dilemma. It sounds clearly wrong to make the connexion a mere empirical fact; but if we make it analytic, then good cannot have descriptive force, for from a mere description advice cannot be logically inferred. (¶ 13)

I should indeed say that the connexion is not merely empirical; but neither is it analytic. It belongs to the ratio of want, choose, good, and bad, that, normally, and other things being equal, a man who wants an A will choose an A that he thinks good and will not choose an A that he thinks bad. This holds good whether the A's we are choosing between are knives, horses, or thieves; quidquid appetitur, appetitur sub specie boni. Since the qualifying phrase normally and other things being equal, is necessary for the truth of this statement, it is not an analytic statement. But the presence of these phrases does not reduce the statement to a mere rough empirical generalization: to think this would be to commit a crude empiricist fallacy, exposed once for all by Wittgenstein. Even if not all A's are B's, the statement that A's are normally B's may belong to the ratio of an A. Most chess moves are valid, most intentions are carried out, most statements are veracious; none of these statements is just a rough generalization, for if we tried to describe how it would be for most chess moves to be invalid, most intentions not to be carried out, most statements to be lies, we should soon find ourselves talking nonsense. We shall equally find ourselves talking nonsense if we try to describe a people whose custom was, when they wanted A's, to choose A's they thought bad and reject A's they thought good. (And this goes for all interpretations of A.) (¶ 14)

There is, I admit, much more difficulty in passing from man to good/bad/man, or from human act to good/bad/human act, if these phrases are to be taken as purely descriptive and in senses determined simply by those of man and human act. I think this difficulty could be overcome; but even so the Oxford Moralists could no deploy a powerful weapon of argument. Let us suppose that we have found a clear descriptive meaning for good human act and for bad human act, and have shown that adultery answers to the description bad human act. Why should this consideration deter an intending adulterer? By what logical step can we pass from the supposedly descriptive sentence adultery is a bad human act to the imperative you must not commit adultery? It is useless to say It is your duty to do good and avoid doing evil; either this is much the same as the unhelpful remark It is good to do good and avoid doing evil, or else It is your duty is a smuggling in of an imperative force not conveyed by the terms good and evil which are ex hypothesi purely descriptive. (¶ 15)

We must allow in the first place that the question Why should I? or Why shouldn't I? is a reasonable question, which calls for an answer, not for abusive remarks about the wickedness of sking; and I think that the only relevant answer is an appeal to something the questioner wants. Since Kant's time people have supposed that there is another sort of relevant reply--an appeal not to inclination but to the Sense of Duty. Now indeed a man may be got by training into a state of mind in which You must not is a sufficient answer to Why shouldn't I?; in which, giving this answer to himself, or hearing it given by others, strikes him with a quite peculiar awe; in which, perhaps, he even thinks he must not ask why he must not. (Cf. Lewis Carroll's juvenile poem My Fairy, with its devastating Moral: you mustn't.) Moral philosophers of the Objectivist school, like Sir David Ross, would call this apprehension of one's obligations; it does not worry them that, but for God's grace, this sort of training can make a man apprehend practically anything as his obligations. (Indeed, they admire a man who does what he thinks he must do regardless of what he actually does; is he not acting from the Sense of Duty which is the highest motive?) But even if ad hominem You mustn't is a final answer to Why shouldn't I?, it is no rational answer at all. (¶ 16)

It can, I think, be shown that an action's being a good or bad human action is of itself something that touches the agent's desires. Although calling a thing a good A or a bad A does not of itself work upon the hearer's desires, it may be expected to do so if the hearer happens to be choosing an A. Now what a man cannot fail to be choosing is his manner of acting; so to call a manner of acting good or bad cannot but serve to guide action. As Aristotle says, acting well, εὐπραξία is a man's aim simpliciter, ἁπλὼς, and qua man; other objects of choice are so only relatively, πρός τι, or are the objects of a particular man, τινός; but any man has to choose how to act, so calling an action good or bad does not depend for its effect as a suasion upon any individual peculiarities of desire. (¶ 17)

I shall not here attempt to explicate the descriptive force of good (bad) human action; but some remarks upon the logic of the phrase seem to be called for. In the first place, a tennis stroke or chess move is a human act. Are we to say, then, that the description good tennis stroke or good chess move is of itself something that must appeal to the agent's deisre? Plainly not; but this is no difficulty. Although a tennis stroke or a chess move is a human act, it does not follow that a good tennis stroke or a good chess move is a good human act, because of the peculiar logic of the term good; so calling a tennis stroke or a chess move good is not eo ipso an appeal to what an agent must be wanting. (¶ 18)

Secondly, though we can sensibly speak of a good or bad human act, we cannot sensibly speak of a good or bad event, a good or bad thing to happen. Event, like thing, is too empty a word to convey either a criterion of identity or a standard of goodness; to ask Is this a good or bad thing (to happen)? is as useless as to ask Is this the same thing that I saw yesterday? or Is the same event still going on?, unless the emptiness of thing or event is filled up by a special context of utterance. Caesar's murder was a bad thing to happen to a living organism, a good fate for a man who wanted divine worship for himself, and again a good or bad act on the part of his murderers; to ask whether it was a good or bad event would be senseless. (¶ 19)

Thirdly, I am deliberately ignoring the supposed distinction between the Right and the Good. In Aquinas there is no such distinction. He finds it sufficient to talk of good and bad human acts. When Ross would say that there is a morally good action but not a right act, Aquinas would say that a good human intention had issued in what was, in fact, a bad action; and when Ross would say that there was a right act but not a morally good action, Aquinas would say that there was a bad human act performed in circumstances in which a similar act with a different intention would have been a good one (e.g. giving money to a beggar for the praise of men rather than for the relief of his misery). (¶ 20)

Since the English word right has an idiomatic predilection for the definite article--we speak of a good chess move but of the right move--people who think that doing right is something other than doing good will regard virtuous behavior as consisting, not just in doing good and eschewing evil, but in doing on every occasion, the right act for the occasion. This speciously strict doctrine leads in fact to quite laxist consequences. A man who just keeps on doing good and eschewing evil, if he knows that adultery is an evil act, will decide that (as Aristotle says) there can be no deliberating when or how or with whom to commit adultery. But a man who believes in discerning, on each occasion, the right act for the occasion, may well decide that on this occasion, all things considered, adultery is the right action. Sir David Ross explicitly tells us that on occasion the right act may be the judicial punishment of an innocent man that the whole nation perish not; for in this case the prima facie duty of consulting the general interest has proven more obligatory than the perfectly distinct prima facie duty of respecting the rights of those who have respected the rights of others. (We must charitably hope that for him the words of Caiaphas that he quotes just had the vaguely hallowed associations of a Bible text, and that he did not remember whose judicial murder was being counselled.)1 (¶ 21)

I am well aware that much of this discussion is unsatisfying; some points on which I think I do not see clear I have not been able to develop at proper length; on many points (e.g. the relation between desire and good, and the precise ratio of evil in evil acts), I certainly do not see clear. Moreover, though I have argued that the characteristic of being a good or bad human action is of itself bound to influence the agent's desires, I have not discussed whether an action of its nature bad is always bad and on all accounts to be avoided, as Aristotle thought. But perhaps, though I have not made everything clear, I have made some things clearer. (¶ 22)