Bertrand Russell

Russell Society Home Page

About Bertrand Russell

About the Russell Society

The BRS Library

Society Publications

Russell Texts Online

Russell Resources

JOIN the Russell Society!

Officers and Organization

Contact Us

 The Meaning of Good (1904)*

By Bertrand Russell

Review of Principia Ethica, by George Edward Moore, Cambridge University Press, 1903

WORKS on ethics suffer, as a rule, from two opposite defects. From a desire for system and simplicity, reinforced by logical confusions, they flagrantly outrage common sense in their estimate as to what things are good in themselves; and then, from a dread of consequences which are felt to be immoral, they endeavour, by flimsy and inconclusive arguments, to prove that their estimate of goods leads to the usual code of duties and sins. Both these defects are absent from Mr. Moore’s work. There is throughout a frank appeal to what appear as ethical facts, combined with an extra ordinary subtlety and care in the analysis of their implications. But, in spite of the subtlety, all the discussions are admirably clear, and are intelligible, except in one chapter, without any previous knowledge of philosophy.

It is a merit rarer, perhaps, than is supposed, to aim solely, in philosophy, at the discovery of truth. Most philosophers are interested, almost exclusively, in establishing some apparently valuable conclusions, for which they seek to find premises. The premises, when found, do not interest them on their own account; and any questioning of the premises is regarded as trivial and carping. Thus attention is concentrated on results, and a hasty, practical tone of mind is generated. In the present author, no trace of this mistake is to be found.

“It may be thought,” he says on one occasion, “that my contention is unimportant, but that is no ground for thinking that I am not in the right. What I am concerned with is knowledge only – that we should think correctly and so far arrive at some truth, however unimportant.”
In accordance with this theoretical spirit, Mr. Moore begins by dismissing the notion that ethics is solely concerned with human conduct, or with the goods attainable by human beings: ethics is the general inquiry into what is good, and into what good is. The chief contention of the first chapter is that good itself is indefinable: an ultimate, simple notion, like yellow. Not that it is impossible to define the good, i.e., the things which are good; but that what we mean when we say that a thing is good, cannot be explained in any other terms. This is established by observing that, however we may propose to define good, it is always significant to say that that which is suggested as the definition is itself good. If we say: “pleasure is good,” we say something different from: “pleasure is pleasure”; thus good cannot mean the same as pleasure. And the same process may be applied to any other suggested definition. The notion that good can be defined, is called by Mr. Moore the Naturalistic Fallacy, because usually some natural object (i.e., something which exists) is taken to be the meaning of good. He shows that, in one form or other, it has been committed by almost all ethical writers; it is involved, for example, in every attempt to infer what ought to be, from what is or will be. The remainder of the chapter is concerned, first, with distinguishing good as end from good as means – what is called good as means is merely a cause of what is good, while good as end is the same as good simpliciter – and, next, with the principle of what are called organic unities, i.e., wholes whose value is not the sum of the values of the parts. These have a very important place in estimating goods, and are frequently discussed in later chapters. An instance is the enjoyment of a beautiful object. A beautiful object which no one sees has little or no value, and a mistaken admiration also is not much prized; but, when the object admired has beauty, we get a whole which is often very good indeed.

Chapter 2, on Naturalistic Ethics, discusses theories which hold that the only good things are certain natural objects, in so far as these theories are advocated as derivable from the very meaning of good. It is shown that such theories always confuse good, in its correct and indefinable sense, with the sense which they assign to it by definition. For example, Evolutionist Ethics are apt to argue that good means more evolved, and on this to base practical recommendations. Yet, if their contention were correct, no practical consequences could follow. We ask: Why should I prefer this to that? And they reply: Because the more evolved is the better. But if they were right in the reason they give for thinking so, they have only said that the more evolved is the more evolved; and this barren tautology can be no basis for action. The meaning of two phrases cannot be the same, if it makes any difference whether we use one of them or the other; and, applying this test, it is easy to see that more evolved does not mean the same as better.

The doctrine that pleasure is the sole good is next discussed; and its refutation appears as complete as any refutation can be. Sidgwick who, alone among Utilitarians, has recognised clearly that “good” is indefinable, argues that, though other things than pleasure are valued, they appear not to be valued apart from the pleasure which accompanies them. But this only proves that pleasure is a constituent of valuable wholes, which we may admit to be usually true. In virtue of the principle of organic unities, it may happen that a whole would be worthless without a certain constituent, and yet that its value does not lie wholly in that constituent. And this appears to be often the case with pleasure. For example, pleasure is essential to the goodness of enjoyment of works of art; yet such enjoyment is commonly held to be much better than other pleasures which are quite as keen. Thus, though many enjoyments are good, they are not good in proportion to the pleasure they contain; and yet it may be that without the pleasure they would cease to be good.

Metaphysical Ethics, which are discussed in Chapter 4, agree with Naturalistic Ethics in thinking that the question: “What is real?” has a bearing on the question: “What is good?” They thus come within the scope of the Naturalistic Fallacy. But they are distinguished by the belief, that the reality which is relevant is supersensible and timeless. All the discussions on this subject are excellent, but they are impossible to reproduce within the limits of a review.

Chapter 5, on Ethics in Relation to Conduct, though it abounds in important distinctions, appears to me the least satisfactory in the book. The question discussed in this chapter is: “What ought we to do?” It is held that what we ought to do is that action, among all that are possible, which will produce the best results on the whole; and this is regarded as constituting a definition of ought. I hold that this is not a definition, but a significant proposition, and in fact a false one. It might be proved, in the course of moral exhortation, that such and such an action would have the best results; and yet the person exhorted might enquire why he should perform the action. The exhorter would have to reply: “Because you ought to do what will have the best results.” And this reply distinctly adds something. The same arguments by which good was shown to be indefinable can be repeated here, mutatis mutandis, to show the indefinability of ought. And, at a later stage, Mr. Moore becomes untrue to his own definition. In regard to moral principles, such as: “Thou shalt do no murder,” which are generally useful and generally obeyed, he holds that there must be instances where better results would follow from breaking them; yet, since we can never know when such an instance is before us, he holds that we ought always to obey such rules. This implies that we ought to do what we have reason to think will have the best results, rather than what really will have the best results. It is certain that some people, whom I refrain from naming, might with advantage to the world have been strangled in infancy; but we cannot blame the good women who brought them up for having omitted this precaution. Mr. Moore’s objection to this view is, that he thinks it a contradiction in terms to say that it was a pity a man did his duty. It must be admitted that this sounds paradoxical; yet paradox of some kind is apparently unavoidable. Mr. Moore, in consequence of his definition, is led to infer, that we can never be sure what we ought to do, since we cannot calculate all the consequences of our actions; also that no moral law can be self-evident, as the Intuitionist school suppose. If ought is indefinable, these consequences do not follow. They may, nevertheless, be true in the main; but there must be at least one self-evident proposition as to what ought to be done. This will be some such rule as, that we ought to do what, so far as we can judge, will have the best consequences; though it is doubtful if this particular rule is itself quite true.

A virtue is defined as an habitual disposition to perform acts which usually have the best results; and the notion that virtue is the sole good is discussed and rejected. It is pointed out, that those who have professed this view have yet thought it possible that virtue should be rewarded in heaven by happiness, thereby showing that they regarded happiness as also good; for if virtue were the sole good, it would be logically compelled to be its own reward. But it is admitted that such virtues as are not mere habits are good in themselves, as well as being useful as means. Their value as ends is, however, scarcely enough emphasised; and it seems a pity that so little is said about them, and about right actions informed by virtues, in the description of the things which have a high degree of intrinsic goodness.

The last chapter, on The Ideal, is the best in the book. It consists chiefly of an enumeration, analysis, and comparison, of those among the things we know which are very good. Although the results differ widely from those usual in works on ethics, they are almost all unhesitatingly affirmed by common sense. It is quite extraordinary and surprising with what certainty the author is able to appeal to our intuitive perception of values, not only in regard to fairly simple matters, but even in very complex and elaborate organic unities. There is a keen pleasure, as we read, in the sure assent with which we follow his estimates, and in the discovery that our power of judging is at once more subtle and more certain than we had supposed.

To judge of the intrinsic value of anything, Mr. Moore says, we must consider what we should think of that thing existing in isolation. We thus avoid two errors: first, the ascription of value to mere means, and, secondly, the supposition that, when one part of a good whole has no value, all the value must lie in the other parts. Applying this test, it appears obvious, that by far the best things we know are the enjoyment of beautiful objects and the pleasures of human intercourse. These two are separately considered.

Æsthetic enjoyments require for their goodness, not merely perception of the object, but also an appropriate emotion. But the emotion apart from the object has little or no value, and, if directed to an ugly object, the whole so formed may be very bad; it is necessary both to see the beautiful qualities of the object, and to see that they are beautiful. We must further distinguish according as there is or is not a belief that the object exists; in the case of imagination and the representative arts, such a belief is absent. The presence of this belief, if true (i.e., if the object does exist), makes the whole much better; if false, it makes the whole worse. Mr. Moore does not decide whether, in this last case, the whole is good or bad; yet this is a question of some practical importance. The love of God, plainly, is in itself a thing of great value; and it is almost always much weakened, if not destroyed, by unbelief. Ought an unbeliever, under these circumstances, to seek to destroy belief in others? Though the effects of belief are here relevant, it is also important to know whether the love of a good object falsely believed to exist is, on the whole, in itself good or bad; and this is one of those rare questions upon which our intuition gives no certain answer.

In regard to personal affection, all the elements present in the previous case exist, together with the fact that, when not misdirected, the affection has an object which is not merely beautiful, but good in a high degree. But, from the point of view of analysis, there is not much to add to the previous discussion of beauty.

A discussion of evils follows. Three evils are recognised as preeminent: namely, the admiration of what is evil or ugly, the hatred of what is good, and consciousness of great pain. Admiration of evil is made worse, both by judging it to be evil and by judging it to be good; but is unaffected by the existence of its object. Hatred of good is better if the good is thought to be evil, and worse if acknowledged to be good. It also becomes worse if the object is perceived to exist. With regard to pain, Mr. Moore is, so far as I know, the first to point out its lack of parallelism to pleasure, though many must have felt this. Pain, he says, is in itself a grave evil, whereas pleasure is not a great good; but, conversely, pleasure often much improves a whole, whereas pain does not make a whole much worse, and may even make it better, as in the case of sympathy for suffering. In this and other such cases, a whole formed of two evils may, as a whole, be good; hence, when one evil exists, it is sometimes good to create another. But there seem to be no cases where the whole and the parts together are good on the balance, and where yet one of the parts exists and is evil, and no good part exists. Thus we cannot maintain that the existence of evil is essential to the ideal. But its mere apprehension is essential, as may be seen by the excellence of Tragedy.

In conclusion, Mr. Moore points out, that the lack of symmetry and system in his results is not an objection to them, since there is no reason to suppose the truth symmetrical. In this we must agree most entirely: philosophy will never advance, until the notion is dispelled, that sweeping general principles can excuse the patient attention to detail which, here as elsewhere, can alone lead to the discovery of truth.

B. Russell

*  Bertrand Russell, “The Meaning of Good,” The Independent Review 2 (Mar 1904), 328-33 Review of G.E. Moore, Principia Ethica