You are here

Chapter VIII: The Proof of Moral Knowledge


The moral good has been defined as harmonious happiness, or as that organization of interests in which each enjoys the non-interference and support of the others, whether within the personal life or the life of society. This becomes the moral “first principle.” It sets the standard by which objects are deemed morally good or bad, and is the premise from which right, duty, and virtue are to be derived. It provides the most general predicate of moral judgment and the basic concept of moral knowledge. How is it to be proved? The moral philosopher is compelled not only to produce evidence, but to decide what kind of evidence may properly be demanded.

Opinion on the question of moral knowledge oscillates between the extremes of dogmatism and scepticism. There are no judgments which are held with a greater degree of commitment and sense of certainty by the layman, and with a greater acknowledgment of ignorance by theorists. Moral beliefs are in the strange position of being held with passionate conviction despite apparently unanswerable doubts. It is easy to understand why the layman, compelled to meet the deadline of action, should resort to dogmatism; but why is it that moral theory, relieved of the exigencies of action, and emancipated from dogmatism, should move to the opposite extreme of scepticism?

Moral scepticism, or the view that moral judgments cannot be proved true, usually takes one or the other of two forms: the assertion that there are, strictly speaking, no moral judgments; and the assertion that moral judgments are relative or circular.

It is held, in the first place, that so-called moral judgments, although they assume the verbal form of sentences, having subject, predicate and verb, are not really judgments, accountable to evidence and demonstrably true or false. This most radical form of moral scepticism takes two forms: the dismissal of moral judgments as failing to satisfy the requirements of verification, and the interpretation of moral judgments as merely acts of self-expression and social control. These two views are usually held jointly, the one discrediting moral judgments as “nonsense,” the other explaining the role of such nonsense in human life.

The denial of the verifiability of moral judgments is based on a certain conception of what knowledge is, derived from the more “exact” sciences, and followed at various distances by other sciences which call themselves by the name of ‘science.’ The philosophical exponent of this cult is known as the logical positivist, according to whom knowledge is of two kinds, a logico-mathematical knowledge, which is “analytical,” that is, yields no more than is put into it by definitions and postulates; and experimental knowledge, which is “synthetic,” that is, yields new discoveries, and which consists of hypotheses verified by data of sense in certain controlled, repeatable, and measurable situations.

Before accepting this strict interpretation of scientific method as a definition of knowledge in general there are certain considerations to be weighed. In the first place, since the method is derived from physics, it is not to be assumed that it should be applicable to non-physics. In the second place, the method itself creates problems which the method itself cannot solve: problems such as the precise nature of verification, or of the activity of constructing hypotheses. Indeed the central thesis of logical positivism itself — its exclusive claim to the title of ‘knowledge’ and its description of the process of knowledge — is not proved by the logico-positivistic method. The literature of logico-positivistic philosophy is not a part of the corpus of physics. And in the third place it is to be noted that no science except physics (and possibly chemistry) employs the method of physics; so that the strict application of this standard would exclude the social sciences, and even the biological sciences. Whatever be the role assigned to these sciences, whether that of autonomy, or of half-finished physics, a theory of morality has every right to a similar role.

The other part of moral scepticism, namely, its interpretation of socalled moral judgments in terms of self-expression and social control, has already been examined in its application to judgments of value in general.1 It is argued that moral judgments, like other judgments of value, have the intent not of describing an object but of conveying an attitude. The fact is, however, that this is not peculiar to moral judgments, but applies to all judgments in some degree, and to moral judgments only in a comparatively high degree. Judgments are not uttered in a vacuum, but are appropriate to some particular context, and since they are uttered in words, they are a form of communication, even if they are uttered “to oneself.” Any verbal utterance, whether or not it employs such words as

‘right’ ‘ought,’ and ‘good,’ may serve as a mere gesture, epithet, threat, command, or enticement. And it may be freely conceded that moral judgments since they closely concern action and human relations are the most likely of all judgments to be used with such an “emotive” intent. But that moral judgments do sometimes assert propositions which can be argued and more or less successfully demonstrated, is incontestable; and it is the business of moral theory to look to those moral judgments which do assert propositions, and to inquire concerning the evidence which can be cited in their support. The question is, “What and how do moral judgments know, when, and insofar as, they are acts of cognition?

The second argument for moral scepticism is the charge of relativity. Moral judgments are held to be peculiarly or fatally relative. But, as has been pointed out, relativity is vicious only when it is concealed. Moral judgments commonly apply a standard without stating it; as when a hurtful act is simply pronounced “wrong,” when it should be pronounced “wrong-by-the-standard-of-harmonious-happiness.” Generally speaking the standards of a person or group are so ingrained and habitual that they are brought to light only by a detached observer, or by a prolonged process of self-justification; with the result that the judgment affirms an absolute when it really means a relationship. This, again, is not peculiar to moral judgments. Men speak of a tree as “big,” without specifying whether they mean height or girth; or refer to a man as “great” without specifying whether they mean powerful or wise; or characterize an event as “novel,” without specifying the accustomed events with which it is contrasted.

Moral judgments are also relative in the sense that the meaning of the predicate is ambiguous and variable. Acts are declared right or wrong without any clear awareness of what is meant by ‘right’ and ‘wrong,’ some men meaning one thing and some another, or nothing at all; with the result that it is impossible to determine which of two opposing moral judgments is true and which false. But this, also, holds in some degree of all judgments, even of those carefully considered judgments which are called ‘scientific’ and in which undefined terms are reduced to a minimum. Are sun spots a cause of weather? This depends on what is meant by ‘cause,’ and it is quite possible that a conflict of opinion on the subject should be insoluble because of a difference of meaning.

Moral judgments can be challenged or postponed by substituting for the question, “Is it so?” the question, “What do you mean?” But so can any judgment. And even when they mean the same, men hold different opinions relatively to their degree of capacity and access to evidence. No doubt such conflicts of opinion are peculiarly common in the area of morals, where bias is passionate, where the unknown factors are numerous, and where the trains of reasoning are long and complicated. But there are such differences of opinion in every field of inquiry.


The knowledge of morality differs from other kinds of knowledge not qua knowledge, but in its subject matter. Before this thesis can be proved certain common misunderstandings must be removed.

It is sometimes contended that the knowledge of morality differs radically from all other knowledge in that its data are feelings. But this contention is equivocal. The datum of feeling may be taken to signify the state or act of feeling; or it may be taken to signify what is felt. In the first case, it is the feeling that is observed, whether introspectively or behavioristically; in the second case, it is the object of the feeling which is observed, and the feeling is the observing. In the first case the data do not differ from those of psychology, which embraces feeling along with sensation, and other modes of the mental life, within its subject matter. In the second case the data are those qualities or other “neutral entities” which are immediately apprehended by feeling, as colors are immediately apprehended by visual sensation, and tones by auditory sensation. In either case there are empirical data by which hypotheses can be verified.

Closely allied with this misunderstanding is the contention that moral knowledge, unlike other knowledge, must move the will at the same time that it convinces the intellect.2 It is held that the judgment “This is right,” or the judgment “I ought to do this” cannot be true unless one is disposed to act accordingly. But this would exclude the possibility of being inclined to a certain performance because of seeing that it is right. It would render meaningless the judgments “This would be the right thing to do under the circumstances,” or “You ought to do this,” or “He ought to do this.” It would not provide for moral education, which, having first judged what a man's duties are, then seeks to inculcate living up to them.

The ground has now been cleared for the positive statement that moral knowledge possesses the same general characteristics, and is subject to the same discipline, as all knowledge. It is true or false according to the evidence. It must avoid contradiction. It must invent and verify hypotheses. It must be faithful to the specific purpose of knowing, despite all temptations to the contrary. It must be self-denying, and accept the verdict pronounced by the facts or necessities of its subject matter. It must define its terms. These and all other formal criteria, or maxims, which are applicable to knowledge in general, are applicable to moral knowledge in particular, and in the same sense.

Subject to these generalities which characterize all knowledge, there are two kinds of moral knowledge, derivative and basic. When an act of homicide is judged to be wrong it is ordinarily sufficient to call it ‘murder.’ That is deemed sufficient, since it is assumed that murder is wrong. This judgment may be subsumed under some other accepted generalization, such as the right to life, or the goodness of security and order. But if one follows this line from premise to premise, and if one avoids circularity, one arrives eventually at an ultimate premise or first premise which cannot be similarly deduced.

The application to the standard of harmonious happiness is evident. It is judged that things are morally right and wrong, good and bad, obligatory and forbidden, judged by the standard of harmonious happiness. There are two judgments, the judgment which adopts the standard, and the judgment which applies it. The fundamental question of moral knowledge is the question of the proof of the first or basic judgment. It is a judgment about a standard, and to the effect that a specific standard, such as harmonious happiness, occupies a peculiar place among standards, and is entitled to be designated as “the moral standard.” This is not a moral judgment in the sense of assigning such predicates as ‘good,’ ‘right,’ and ‘ought.’ Moral theory, whether it asserts that the ultimate moral standard is happiness, or that the moral right or good is indefinable, or that duty is obedience to God, or that the right is the reasonable, stands outside the whole circle of such judgments, and makes non-moral statements about them.

The first condition which such a theory must satisfy is that the proposed standard should be in fact a standard, or qualified to be a standard. If harmonious happiness can be truly affirmed to be the moral standard, it must so agree with human nature and the circumstances of human life that men can adopt it by education, persuasion, and choice; and, having adopted it, can govern their conduct in accordance with its requirements. It must be qualified to serve as a criterion by which human interests, acts, characters, and organizations can be classified and ranked. The evidence that it satisfied these requirements will be found in the fact that it is so adopted and employed.

If, however, harmonious happiness is to be proved to be the moral standard, to the exclusion of other standards for which a similar claim is made, it must possess further and unique qualifications. Otherwise it will be merely one standard among many, differing only historically. There would be no ground of persuasion by which the adherent of another standard could be converted to this standard. It could be judged in terms of this standard, but there could be no judgment between them. The standard of harmonious happiness would have no theoretical precedence.


Before formulating and defending a conclusion as to the ultimate proof of moral knowledge, it will be profitable to review and criticize certain traditional doctrines which have been widely held, both by philosophers and by common sense. The most notable of these are the intuitionist proof, the rationalistic or logical proof, the metaphysical proof, and the psychological proof. The very shortcomings of these doctrines are instructive.

The first and most familiar of these alleged proofs is the appeal to intuition, or to a moral absolute pitch. There is, it is claimed, a sheer rightness of the right, or a sheer oughtness of the ought, or a sheer goodness of the good. He who knows where to look for it and is properly attuned will find it, and that is the end of the matter. The proof that an act is right is that it is seen to be right; the proof that one ought to perform an act is that one is aware of the obligation to perform it. The proof that a state of affairs is good is that its goodness shines forth.

This view has a pre-critical form which appeals to common sense because it escapes the pangs of thought and analysis; it coincides with habit or the intellectual line of least resistance. Post-critical intuitionism gives a philosophical sanction to this non-philosophy. It accepts a now largely discredited notion of self-evidence. In modern geometry, the “axiom of parallels” is taken as a “postulate”; but the intuitionist in geometry held that it is self-evident that two parallel lines can never meet: it is plain that it is so. According to this view, syllogistic reasoning in general is a chain suspended from some first premise which is self-evident, and which transmits its manifest truth to all the conclusions which follow from it. Similarly, according to moral intuitionism, when one considers a statement in which ‘good,’ ‘right,’ or ‘ought’ appears as the predicate, one sees both what the predicate means and that the statement is true. It needs no further evidence that what it contains within itself.

In current discussion of moral matters, intuition is commonly invoked against the hateful thesis that “the end justifies the means.” It is held that the means plainly is or is not right or obligatory, independently of good or evil consequences.

The question of means and ends is never, in human life, a simple relation of a single means to a single end. The means to one end invariably affects other ends. Even so trivial a choice as means of transportation to a given destination, whether one travels by train or by plane, has anticipated consequences over and above the arrival at the desired place at the desired time — consequences for one's pocketbook, or for one's enjoyment of the landscape. In political action a means to an end, such as the peace of Europe, or the supremacy of Germany, or the triumph of socialism, at the same time crosses, cancels or confirms the means to a hundred other ends. When the means is condemned on moral grounds this does not imply that it is not justified by its end, but that the means to the end of harmonious happiness is to be given priority over all the means to all other ends.

Arthur Koestler discusses this issue in a brief, but impressive article entitled “The Dilemma of Our Times,” with the subtitle “Noble Ends and Ignoble Means.”3 He first illustrates the “dilemma” from the tragic fate of Captain Scott and his four companions who perished because they refused to abandon one of their number who fell ill. The writer then finds the same dilemma in modern political action, and points out that Chamberlain's sacrifice of Czechoslovakia and Hitler's and Stalin's purgings of their opponents represent the opposite procedure, parallel to what Captain Scott would have done if he had sacrificed Petty Officer Evans to the safety of the rest.

This analysis illuminates the difficulties of the moral life, but throws no light on the principles involved. It is evident that to disregard consequences would not have provided a solution; for then there would have been no dilemma and no tragedy. It is implied that if it had been possible to bring all of the party to the destination, though at the cost of delay and suffering, or if Evans had willingly acceded to his own sacrifice, that would have been the solution, since the deed would then have represented an agreement of all persons whose interests were at stake. In that case, apparently, the end would have justified the means.

If the end does not justify the means, what does justify it? The bare dictum that the end does not justify the means has no force whatever. At best it is a confused expression of several truths: that a means is ordinarily a means to many ends; that a given end does not justify its means regardless of other ends; that the means to a higher end overrules the means to a lower end; that a means is sometimes also an end.

The intuitionist may select one of several concepts as the moral ultimate. This role may be assigned to right, to ought, or duty, or to good, better, or best. The judgments, “England was right in declaring war on Germany,” “Hitler ought not to have invaded Poland,” “A world subjugated by Hitler would have been an evil and not a good world,” are all moral judgments. The intuitionist chooses one as the basic intuition and derives the others from it. If he chooses “good,” then the “ought not” of Hitler is argued from the comparative good of the state of affairs which would have resulted from his non-aggression, and the “right” of England is argued from the comparatively good effects of her intervention. When right or ought is selected as the basic concept, the ensuing state of affairs is argued good from the right or ought of the act which brought it about.

The most serious weakness of the intuitionist view lies in the plurality and inconsistency of these basic intuitions. What one intuitionist holds to be basic another holds to be derivative. That the same intuition should be affirmed and rejected by equally attentive and scrupulous intuitionists, cannot but create a suspicion that they are all mistaken, and that their alleged intuitions are not intuitions at all, but dogmas, assumptions, or meaningless words.

Except among doctrinaire intuitionists there has developed a justifiable scepticism of intuitions. The justification lies in the historical fact that so many intuitions have failed to stand the test of time, in the increased understanding of the way in which pseudo-intuitions are generated, and in the appeal of the intuitionist himself to supporting evidence for what, if it were intuitive, would need no support.


As moral intuitionism derives its philosophical repute from Plato, so moral rationalism owes its prestige among philosophers to the influence of Kant. It would be unprofitable to embark on a meticulous examination of the Kantian doctrine. Many scholars who have been tempted into this labyrinth have disappeared and have never been heard from since. What Kant has come to mean (whether he did mean it or not!) is most instructive when it is most freely interpreted.

Kant defined the moral first principle in terms to which no exception can here be taken. To paraphrase the several statements of his “categorical imperative,” morality prescribes that all human persons shall be treated as ends in themselves, and that life shall be organized accordingly. Each person shall subordinate the pursuit of his own end to the requirements of a harmonious society of such persons. A follower of Kant has expressed this ideal in terms even closer to those used in the present exposition:

If we cannot perfectly realize this new ideal, if absolute harmony is unattainable, one can still walk in the light of the ideal. One can say: “I will act as if all these conflicting aims were mine. I will respect them all….” This ideal … says: “The highest good would be attainable if all the conflicting wills realized fully one another.”4

But this definition of the moral ideal, acceptable as it is, does not touch the question of proof. It is the contention of Kant that this ideal of a harmony of persons is the only ideal which agrees with reason. The faculty of reason is identified with logic, and logic in turn is construed to mean two things: consistency, taken to include implication as well as non-contradiction; and universality, taken to mean the subsumption of particulars under generals. In short, it is argued that the above ideal is the only ideal which can be adhered to consistently, and under which all acts of all persons can be subsumed.

As a matter of fact, however, there is no principle of conduct which cannot be adhered to consistently — even the principles which Kant and his followers would most stoutly have rejected. It is possible, theoretically to adhere consistently — all too consistently — to the principle of personal aggrandizement or the principle of race supremacy. It is theoretically possible that these principles should be practiced by all persons; the result would, it is true, be deplorable, but it would not be illogical. Nor can it be said that a person who practiced one of these abhorrent principles could not will that others should practice the same. He might prefer a power of ascendancy gained by overcoming resistance to one in which the rival tamely submitted. To use Kant's famous example, one might adopt the maxim of mendacity and at the same time prefer a company of rival liars to a company of the credulous.

One may argue that there are certain kinds of rules which are such that their social benefits accrue only in proportion as they are generally observed. This would be the case with most of the familiar moral maxims, such as veracity, honesty, and respect for life. But then the proof of these rules lies not in their logic but in the social benefits which accrue. Or, one may argue that what people really want is the kind of society in which what is good for one is good for all. But do they? And if they did, the proof would be an appeal to what people actually want, and not to the formal principles of consistency and universality.

The fundamental flaw in the rationalistic theory is its failure to distinguish between logic and the facts of life. The latter have been smuggled in, thus creating the false impression of deriving a moral principle directly from logic. Kant confused logical universality with social universality and inconsistency with conflict, and therefore found it illogical that different persons should practice different codes. But he himself spoke of man's “unsocial sociability,”5 meaning that man's ideas of social organization spring from his actual experience of social conflict. This is a profound observation. The strength of his position lies not in his futile attempt to “deduce” his moral imperative from reason, but in his ideal of a harmonious society in which persons are moral finalities, and which not only serves all persons but commands their assent.


Authoritarianism, taken as moral theory, can be briefly dismissed. It may be combined with any moral principle, providing incentive, enforcement, or “sanction” for a right or good determined on other grounds. Sheer authoritarianism, authoritarianism pure and simple would define right, ought, and good as obedience of command, If ‘right’ be taken as the fundamental moral term, an act is right because commanded, and not commanded because it is right. And the same is true of ‘ought’ or of ‘good,’ if one of these be selected as fundamental. It is doubtful if sheer authoritarianism has ever been consistently maintained. It should demand “passive obedience,” obedience without asking the reason why. But every authority — God, priest, ruler, teacher, or parent — has been justified as worthy of obedience, because of his knowledge or power, or because he himself conforms to the moral value which he is supposed to create by his command.

Even a so-called “absolute” authority such as the Christian God, is commended to those who obey him, by certain qualifying moral attributes. He may be deemed all-wise, in which case he is taken as the supreme expert or guide in the moral practice; or he may be deemed all-powerful, and hence the giver of happiness or unhappiness to those who obey or disobey his will. He may be taken as the Creator and Providence by whom right and good are realized in nature and history, and by whom their triumph is guaranteed. Or, finally, he may be taken as himself the supreme embodiment of moral perfection, a model to be emulated or loved. In each case some prior moral standard is introduced. In each case there is a moral ground for obedience; a judgment of right, ought, or good by which the authority is accredited.


The authority of God may be justified on the ground that God is the ultimate being. This is a variety of the metaphysical proof, to which attention is now directed. Such a theory, if it is to mean anything, requires degrees of being, or a difference within what is, between that which “really” is, and that which “apparently” is. If moral value were simply equated with what is, then the difference between right and wrong, ought and ought not, and good and evil, would be obliterated; since both terms of these antitheses are. Granted some difference between appearance and reality, ‘right,’ ‘ought,’ and ‘good’ may then be defined as agreement with reality versus appearance, or with the deeper, broader, total, more fundamental, and more long-range, aspect of things, and their opposites with what is superficial, transitory, and partial.

There will be as many varieties of this theory as there are varieties of metaphysics, whether materialistic or idealistic, pluralistic or monistic; but to present the theory in its generality these varieties must be disregarded. Its adherents must be prepared to say, “Right, obligation, or goodness, whichever is taken as the moral first principle, means agreement with the fundamental nature of things, whatever it be”: whether the mechanical order of a physico-chemical world, or a Darwinian evolution by the survival of the fit, or a Marxian course of history determined by economic forces and class conflict, or the progressive unfolding of a beneficent purpose, or the realization of a perfect spiritual being, or an infinite substance, or a congeries of accidents and unrelated items.

This view, like authoritarianism, is rarely if ever held in its purity.6 Ordinarily agreement with reality is justified by ascribing some moral value to reality itself. But the theory then becomes circular. Right, ought, or good are defined as agreement with a right or good world or with a world which is what it ought to be.

If moral value is taken to mean agreement or disagreement with the deep, long-range, or total aspect of things, whatever it be, it is possible to preserve a distinction between positive and negative moral value, but the implications are strange indeed. Morality then requires that a man be “in tune with the infinite” whatever tune the infinite plays. It implies that one ought to mount the Cosmic Bandwagon whatever its route and destination. It implies that if, broadly speaking, the world is in fact a jungle, under the reign of tooth and claw, then it is right and dutiful to sharpen one's teeth and claws and participate to the best of one's ability in the general carnage.

The metaphysical theory of morality most widely held at the present time is known as the theory of “self-realization.” It is divisible into two doctrines, anthropological and cosmological. They may be combined, or the first may be held without the second.

The anthropological doctrine identifies morality with realization of human nature. That man possesses certain peculiar faculties which distinguish him from other animals, and that individual men may differ in the degree to which they manifest these faculties, is indisputable. But it is to be noted that the human faculties singled out by the self-realizationist are his moral faculties, so that the doctrine contains a disguised circularity: morality is to consist in the realization of man's moral nature. Men have many characteristics, morphological, anatomical, physiological, and psychological, which are not found in other animals. Man is sometimes defined as the “laughing animal,” but no self-realizationist has ever argued that the more one laughs the better, or that the perfect man would be he who laughed longest and loudest. The human faculty which is commonly selected for emphasis is “reason”; which, in this context, is referred to as man's “ruling faculty,” designed to preside over and regulate his passions. But when so conceived as the faculty for introducing unity among diverse interests and correcting their mutual conflict, reason is man's moral faculty. Morality is not deduced from human nature, but human nature is selectively defined in terms of morality.

Self-realizationism of the cosmological type asserts that in proportion as man develops his characteristically human parts or his whole self he coincides with an Absolute Being, which is supremely good because it is supremely real. But this view, also, is ordinarily circular, and contrary to fact as well. Thus it has been contended that there is only one will — a universal or divine will — which is the moral will. All will, insofar as it is understood, or insofar as it is “really” will, is a will which chooses the moral good. To this view there is a double objection. If there were such a will it would not be moral because it was real, but, being real, it would also be moral on other and quite different grounds. The second objection is that there is no such all-comprehensive single will, actual, virtual, or implicit. There are many wills, one of which is the moral will, or the will to harmonious happiness.


The psychological proof is the most ancient and the most persistent of the proofs of morality, and the proof that is most readily accepted by laymen as well as by philosophers. By the psychological proof is here meant the argument that human beings are so motivated that morality will serve them best — if they only know it. To whatever person one addresses the argument, one can prove morality to him as the intelligent way by which to promote the interests by which he is already governed. It is not necessary to change his heart, but only his head.

The psychological argument has taken two forms, the most venerable of which is known as “egoistic hedonism.” According to this view, which is now generally discredited, all men are governed by the motive of private pleasure. According to the second view, men are governed by a variety of motives, some of which are innate and universal, others of which are acquired and variable. Whichever view of motivation is adopted it is held that morality can be proved to any doubter by showing him that it will give him what he wants. If he wants his private pleasure, morality will give him that; if he is moved by any one of a number of interests, then, whatever its object, morality will give him that. Morality, in short, is the universal instrumental good.

It is this second theory of motivation which is usually assumed when it is argued that morality and “enlightened self-interest” coincide. Taken as a proof of the moral standard this view is ambiguous. It may mean that the standard is defined as that principle the adoption of which will promote every man's interest as it stands. But there is no such principle. Or it may mean that a certain principle otherwise defined will, in point of fact, coincide with every man's interest as it stands. As regards the principle of harmonious happiness, however, this is not true.

The psychological proof owes its vogue to the fact that if successful it would induce not merely intellectual acceptance, but the corresponding performance. If a man can be shown that morality will give him what he already wants he will not only assent to it theoretically, but will pursue it. Right, ought, or good is grafted upon the stem of his existing motivation. He is at one and the same time both convinced and incited. The term ‘persuasion’ is often used to embrace both effects, being deemed incomplete until it is translated into action.

But when the proof of morality is required to appeal to the existing inclinations of the person to whom it is addressed, the evidence must vary in each case. It is not possible to argue to a second judge in terms of the self-interest of a first. A principle proved by appeal to the self-interest of Hitler could not be proved to Churchill, since it would not appeal to his self-interest. Or would it? Is it not possible that if Hitler's self-interest and Churchill's self-interest were both sufficiently enlightened, their requirements would coincide? Might it not be that if every man's self-interest were enlightened it would agree with the self-interest of every man and so with the general good? Possible, yes, but highly improbable. There is a presumption against it, as there is always a presumption against a sameness of conclusions deduced from different premises. The coincidence of each self-interest with all self-interest would be a remarkable coincidence.

As a matter of fact this coincidence does not occur, or occurs only occasionally, partially, and under special conditions. Thus, for example, while the interests of most persons are promoted by peace, the interests of some persons are better promoted by war; if, as sometimes happens, they prefer the fish that thrive in troubled waters. In that case the more enlightened their self-interest, the more persistently and methodically will they trouble the waters. The militarist who promoted humane feelings, or the racist who promoted justice and humanity, would be a fool; he usually knows his business better. No moralist, in the ordinary sense of the term, has anything to teach him! There is an enlightenment for every self-interest, narrow and broad; and the conflict of self-interest with self-interest is not annulled, but may be sharpened, by increased enlightenment.

How far the principle of individual self-interest and the principle of harmonious happiness coincide depends on how the self in question is constituted. Given a man whose interests are already harmoniously attuned to those of others it is then a matter of indifference whether the argument for morality is presented in terms of his self-interest or in terms of the interests of all. But since unhappily there is no guarantee that self-interest is so attuned, it must be morally certified before it can be taken as the premise of the argument. Morality can be argued from self-interest only when it has already been put into self-interest.

That men should be inclined to accept the coincidence of self-interest and the general happiness is easily intelligible. It eases the strain of effort. A high officer of the American army is quoted as saying, apropos of the atomic bomb:

We must have a group study the problem and decide what is best for the people of the United States — and that means what is best for the people of the world.7

In other words, for an American to serve the people of the world, it is only necessary to serve the American people, which is easier and simpler — although not so easy or simple as is usually supposed.


The rejection of the traditional arguments may be taken to imply that there are no arguments for the moral standard or first principle here proposed, and that no alternative is left but to postulate it or abandon it altogether. But there still remain arguments to be advanced in its support — arguments which, though they may not satisfy everybody, at least have the merit of being appropriate to the thesis which is to be proved.

In the first place, the standard of harmonious happiness is capable of being agreed on — both theoretically and practically. It satisfies the requirement of cognitive universality and objectivity; that is, it is the same for all knowers who address themselves to the subject. Since the norm of harmonious happiness acknowledges all interests, its affirmation is free from the so-called “personal equation.” As the astronomer recognizes all stellar facts regardless of the accidents of the observer's history, and thus overcomes the geocentricism which has led men to affirm that the heavens move about a stationary earth, so the theory of harmonious happiness overcomes that egocentricism which has led moral observers to subordinate all interests to their own, or to those of their neighborhood, class, or nation. It embraces human perspectives within a total system of relationships. It places itself in all points of view, and fits them together. It discovers alien and remote interests, and makes allowance for the ignorance which it cannot wholly dispel. It is impartial. It says, in effect, that since it is interest as such which generates good, and a harmonious relation of interests which constitutes moral good, to him who makes the judgment his interest is just one among the rest. Since the principle of harmonious happiness deals with the nature of interest in general, and with its types of relationship, it is applicable to all interests and persons.

But while the theoretical proof of the moral principle is obliged only to satisfy the knower as knower, the principle here proposed will tend also to appeal to each knower's will. The good of harmonious happiness, since it embraces all interests, is to some extent to everybody's interest, and thereby obtains a breadth of support exceeding that of any other good. Every person, including the person to whom the argument is addressed, has some stake in it.

The extent to which the harmonious happiness of all men will reward any given man will vary widely. In the absence of propinquity, interaction, and communication, and so long as this condition prevails, it may not reward him at all. When this aloofness is diminished, its reward will depend on how his particular personal happiness is constituted. All men, no doubt, have some spark of humanity, and are affected by the happiness or misery of others — but some men more than other men, and some men scarcely at all. The same is true of the extent to which men's means and ends reinforce one another. This varies with men's vocations, all the way from the recluse who is interested in solitude to the man of business whose affairs are complexly intertwined in a network of employers, workers, buyers, sellers, producers, consumers, and bankers, which now extends around the earth.

The norm of harmonious happiness, furthermore, is the only norm which is capable of appealing to all men not only severally but jointly. It is the only norm which promises benefits to each interest together with all other interests. It does not rob Peter to pay Paul, but limits Peter in order to pay both Peter and Paul.

Hence the norm of harmonious happiness is doubly universal. It is universal in the theoretical sense: its nature and its implications are objective, and the judgments in which it is employed are equally true for all judges; and being abstracted from particular interests, it is applicable to all human situations. It is also universal in the social sense; its promised benefits accrue to all men, and to all men collectively. It is a norm on which all men can unite and agree — both theoretically and practically.


Making due allowance for the possibility of error in general, and for the degree of its probability in any particular field of inquiry, it may properly be argued for any theory that it agrees with widespread opinion. Opinion concerning the physical world is trustworthy in proportion as it can be attributed to observation. The relation between the sun and the earth, for example, reflects the observation of the alternation of day and night. This opinion has to be corrected to take account of the place of the observer, and the influence of the religious dogma which made the earth the scene of the drama of salvation. But whether the sun moves about the earth, or the earth about the sun, or the two move relatively to one another, the empirical fact of the periodic rising and setting remains undisturbed. And so with moral opinion. It has to be corrected to take account of non-evidential influences; not only such general influences as also affect physical opinion, but the peculiar pressures which arise from the fact that moral opinion is so closely connected with action as to be of special concern to society. These non-evidential influences being discounted, there remains an “experience of life” which has taught men the consequences of action and the ways to live prosperously together.

Again and again, in all spheres of life, and in all the ages of man, it has been observed that there are certain procedures by which the destructiveness of conflicting interests can be mitigated, and by which they can enjoy the benefits of peace and coöperation. Overlaid as it is by prejudices of many sorts, this lesson has been repeatedly learned, extended to new situations, and transmitted to future generations.

In spite of the marked differences of moral opinion which appear in different social groups and historical epochs there is nevertheless a notable amount of agreement. The disagreement is notable only because there was once an expectation of perfect agreement, and because of the shocked surprise with which the unfamiliar is always greeted. Language provides an analogy. The first stage is the assumption that all people speak the same language; the second stage is the discovery that there are strange, absurd, and unintelligible languages; the third stage is the discovery that all men use language, and that all languages have their common laws and meanings. In the matter of moral opinion the extreme relativists are those who have reached only the second of these stages.

If morality is taken as that organization of life by which conflict is escaped and by which coöperation is achieved, then the moral problem is universal; and it is, after all, not surprising that amidst all historic, ethnic, social, economic, and evolutionary aberrations there should emerge a broad knowledge of the points of the moral compass. This knowledge appears in generally accepted maxims, precepts, and virtues.

The theory here proposed reaffirms the standard virtues of antiquity — courage, temperance, wisdom, and justice. The good of harmonious happiness requires, like any end, a brave will that is not dismayed by obstacles, and effort sustained without complaint through long stretches of time. It requires a moderation of appetites lest in their excessive indulgence they should rob one another. It requires enlightened mediating judgments, that is, a true representation of ends and an intelligent choice of means. It requires a distribution of goods to each interest in accordance with a judgment which represents all interest. Christianity did not reject these virtues, but added faith, hope, and love; and these, also, are endorsed by the present theory. Harmonious happiness is an ideal, and if an ideal is to be pursued there must be a steadfast belief in its attainability by means that lie beyond present knowledge, and a confidence in its actual attainment in the future. The pursuit of the harmonious happiness of all requires a sympathetic concern for one's fellow man — a sensitiveness to their pains or frustrations and an impulse to help.

Other funded moral wisdom falls into line. The most generally accepted of all maxims, the Golden Rule, is justified because the harmonious happiness of all requires that each man shall put himself in the place of other men, and recognize their interests, however cold and remote, as of the same coin with those warm and intimate interests which he calls his own. Veracity signifies the need of communication as the condition of all human intercourse. Honesty is that keeping of agreements which is essential to security and to concerted plans. Selfishness is that preoccupation with the narrower interests of self, family, class, or nation which obstructs the longer and wider vistas demanded by universal happiness.

These maxims and virtues are not invariably accepted. They are sometimes defied and they are frequently ignored. It cannot, however, be said that they are peculiar to Western Europe, or to capitalistic societies, or to Christianity, or to the modern world. They cross all such divisions, and when, as today, life is organized on a wider scale, to include all nations, all dependent and backward groups, and all hitherto unprivileged persons and classes, it is to this body of moral opinion that men appeal. Equally significant is the fact that when men differ as to the specific applications of moral opinion it is to the standard of harmonious happiness that they look for common ground. And it is by this standard that men criticize and justify their major social institutions — conscience itself, polity, law, economy — and by which they define the places in human society that are to be allotted to art, science, education, and religion.


The proof of the moral standard is “empirical” in the full, rather than the limited, sense of that term. In the limited sense, the term ‘empirical’ is sometimes applied to that part of science which consists of a summary of observations, rather than to that whole in which a conceptual theory is framed in conformity with the requirements of logic and mathematics and then verified by observation. If the theory of morals is to be considered empirical in the full sense, it must be a system of concepts verified by the data of human life.

There is much talk at the present time of a “scientific ethics”: an urge to make ethics scientific, and a discussion of whether ethics can be scientific. There is no doubt that there can be a science which deals with conscience. This is now a recognized part of a scientific sociology, or social psychology, or of a scientific history. There is a certain propriety in giving this inquiry the name of ‘ethics,’ but if this nomenclature is adopted, then ethics must be distinguished from the science of morals.

The ultimate data of moral science are not men's approbations and disapprobations, but conflicts of interest, and the organizations of interests by which they are rendered non-conflicting and coöperative. The proof of any theory of morality is its adequacy and correctness as a description of these data. It does not attempt, nor can it be expected, to describe man or human life generally. The moral life of man must be human, and it must be living; and a description of it must therefore embrace these sets of facts. But the attention of moral theory is focused on certain selected aspects of man which distinguish him as a “moral” being; and on those aspects of human life which are peculiarly characteristic of his “moral” activity.

It is no disproof of the present doctrine of harmonious happiness to point out that men are not harmoniously happy, or are inharmoniously unhappy. All that needs to be proved is that there is a prolonged and widespread attempt to be harmoniously happy; that men are capable of such an attempt; that they can and do take steps in the direction of harmonious happiness; and that they can and do measure their steps by the standard of harmonious happiness. The fundamental claim for the present view is that it describes a peculiarly widespread, fundamental, and persistent human pursuit for which ‘moral’ is the most appropriate name.

  • 1.

    Cf. above, Ch. I, § 4.

  • 2.

    For a more detailed examination of this question cf. the Author's “Value and Its Moving Appeal,” Philosophical Review, 41 (1932).

  • 3.

    Commentary, 1 (1946).

  • 4.

    J. Royce, The Religious Aspect of Philosophy, Houghton Mifflin Company, 1887, pp. 144–5. Since Royce was one of the teachers who most influenced the Author during his student days it is quite possible that the present work reflects his influence. The Author is more conscious of the influence of William James, whose principle of “inclusiveness” (cf. his essay entitled “The Moral Philosopher and the Moral Life”) so strikingly resembles the thought of Royce as represented in the above passage.

  • 5.

    Eternal Peace and Other International Essays, tr. by W. Hastie, 1914, p. 9.

  • 6.

    Perhaps the nearest approach to such purity is to be found in Jonathan Edwards:

  • 7.

    Attributed to Gen. L. R. Groves, New Republic, Jan. 28, 1946. Similarly, Secretary of Defense Charles E. Wilson is quoted as saying that what is good for General Motors is good for the United States.

From the book: