1. For the achievement of a good life, indeed for the achievement of anything good, thought and feeling are both necessary. The exclusion of either makes the other sterile. So much we have seen. But what does each contribute?
Unfortunately this is too large a question for us. Thought and feeling contribute so subtly and variously to the achievement of every form of good that no full account of their interplay is practicable. If we are to deal with the issue at all, we must narrow it sharply. Happily we may do so while keeping before us that form of the issue which is of principal interest to the moralist of cur time. Though the functions of reason and feeling in the good life generally are beyond our compass, we may quite well discuss with profit their functions in moral judgment. This question, though smaller than the other, is still of crucial importance. Upon its outcome depends the solution of many of the issues most eagerly canvassed in our day—whether, and in what sense, there are objective moral standards, whether right is relative or not to one's own taste or opinion, or to that of a particular culture, whether moral disagreements are matters of belief or of feeling, whether moral convictions can be made out by rational argument. Indeed, a more fruitful ethical question could scarcely be raised.
Let us get it clearly before us without delay. When we decide that a certain action is right, or our duty, what is it that does the deciding? Our reason? Our feeling? Or both? Or perhaps something different from either?
The answer of common sense would probably be ‘Both’. Judgments on moral questions seem to stand about midway between purely rational judgments on the one hand and pure expressions of taste or feeling on the other. When we say that Richard's killing of the princes in the Tower was wrong, we are expressing something of a different kind from what we are in ‘a circle has a larger area than any polygon inscribed in it’, and this difference is not simply one of subject-matter. In a judgment that murder has been committed, our feelings are engaged; we are not performing an ethical classification merely; we are expressing revulsion toward the act; while in the judgment about the circle, nothing of the kind need be involved. On the other hand, it seems equally clear that when we call the killing wrong we are expressing more than taste or feeling. Our remark is not like the sigh of boredom with which one might contemplate a gallery of non-objective paintings, or a swearing at someone in anger. We are not merely ejaculating or exclaiming; we mean to say something true, something that many would think as self-evidently true as what we said about the circle. On the surface, at least, a moral judgment is the expression both of an insight into truth and of an attitude or feeling.
But which is it primarily? When we say an action is right or wrong, do we mean to assert something as true, and then find on later inspection that our assertion was accompanied by feeling, or do we mean to express attraction or repulsion, though in a form of words deceptively like a statement? Among contemporary philosophers each of these views is firmly held. Professor Paton writes: ‘It is just as certain that deliberate cruelty is wrong, as it is that grass is green or that two and two make four. Cruelty cannot be consistently willed by men who are trying to live coherent lives in the service of a coherent society.’1 Here a moral judgment is regarded as an assertion of which we may demand truth and coherence. Rudolf Carnap takes an opposite view. Of the statement, ‘killing is evil’, he holds that ‘it is merely the expression of a certain wish’; ‘this statement is not verifiable and has no theoretical sense, and the same thing is true of all other value statements’; ‘a value statement is nothing else than a command in misleading grammatical form’.2 It is obvious that both these views cannot be true.
2. The issue, though so acute at present, is one to which much thought has been devoted by acute minds in the past. We shall do well at first to follow their lead. In considering the larger conflict between reason and feeling in western morals, we have seen that history has provided some very illuminating experiments. In considering more narrowly the place of these two functions in ethical judgment, we shall again find it helpful to give some attention to history. We shall find, indeed, that in modern thought there has been a more or less steady advance, in which the suggestions of one school have been corrected and supplemented by the suggestions of another, so that the issue as it now comes to us has been greatly clarified and sharpened. Here again, for simplicity's sake, we shall narrow the field. The problem before us has been of particular and persisting interest to the succession of British moralists. Not only have all of them been concerned with it in one way or another; they have generally argued with their immediate predecessors in mind, so that their thought displays an instructive dialectic which brings the problem in the end to our own doorstep. I shall try to follow the main steps of this dialectic. But my concern is logical mainly, not historical, so I shall not feel bound at all points to cling to the historical order; what I shall try to show is how a series of able thinkers have successively closed blind alleys for us, and narrowed the problem to its present form.
3. If we by-pass the large figure of Hobbes, whose thought, however impressive, is not much to our present point, the first British thinkers to give themselves to our problem were thoroughgoing intellectualists. Some of them, notably Cudworth and More, took their cues from Plato. One of them, Samuel Clarke, a careful student of Newton, approached ethics from the side of mathematics and offered a case for the rational character of moral judgment that is so simple, clear, and modern in its sound, as to impress even a critical reader of the present day. He stated this case in the Boyle Lectures for 1705, which he delivered in St Paul's Cathedral. We may remark in passing that after the lectures had appeared in print, Clarke received a series of criticisms from an anonymous correspondent in Gloucester, criticisms so cogently argued that in a later edition Clarke published them together with his replies. It turned out that the weighty critic was a twenty-one-year-old student in an academy at Tewkesbury, who had induced a confederate to put Clarke off the scent by posting his letters in Gloucester. The later history of these young rogues is a lesson on the tragic fate of deceivers. The writer of the letters grew up to be Bishop Butler, his confederate to be Archbishop Secker of Canterbury. It was characteristic of Clarke's chivalrous spirit that, not many years later, he got his young critic appointed to the Rolls Chapel where Butler preached his classic Sermons on Human Nature.
Clarke's contention is that there is the closest kind of analogy between the fitness we find in right conduct and the necessity we find in mathematics. We see in a particular case, by the help of pins or apples, that two and two make four, or that a whole is not equal to its part. By implicitly comparing cases, we come to see that the special natures of apples and pins are irrelevant, that the relation is the same whatever the material, and that the whole as such is greater than the part. Our mind works in the same way in morals. A boy ‘borrows’ without permission a companion's ball; someone, to his indignation, serves him in the same way and then he begins to see that principles are involved, that the act, as such, of taking other people's property without their leave is wrong, and that treating other people as one would object to being treated oneself is in the nature of the case unreasonable—or wrong; they mean the same thing.
‘He that refuses to deal with all men equitably, and with every man as he desires they should deal with him, is guilty of the same unreasonableness and contradiction in one case as he that in another case should affirm one number or quantity to be equal to another, and yet that other at the same time not to be equal to the first.’3
According to Clarke, then, when we call a particular act right or wrong, we are implicitly asserting a rule which, when abstracted and put nakedly before us, is self-evident. If I owe another man a duty this is not in virtue of everything in my nature, or everything in his, or of all the relations between us; that is why it is so easy to go off on a false scent in judging conduct, and why the reflective moralist has an advantage over the plain man. If I have the choice between telling someone the truth and telling him a falsehood, and, as far as consequences are concerned, cannot see that either course is better than the other, I should still tell him the truth, by reason of what he is and what I am, and the relations that hold between us. I am a being capable of speech who knows something he does not; he is a being who is interested in knowing what I know; an intelligent man would see as plainly that when two such natures were brought into touch, one should supply truth to the other, as that when a pentagon is inscribed in a circle it must have the smaller area. In both cases the relation is a necessary one, prescribed by the natures of the terms. This necessity in the moral case Clarke describes as ‘fitness’. His use of this term reminds us strongly of its use by the deontologists of our own day:
‘that there is a fitness or suitableness of certain circumstances to certain persons, and an unsuitableness of others, founded in the nature of things and the qualifications of persons, antecedent to all positive appointment whatsoever; also that from the different relations of different persons one to another, there necessarily arises a fitness or unfitness of certain manners of behaviour of some persons toward others, is as manifest as that the proportions which flow from the essences of different mathematical figures have different congruities and incongruities between themselves.…’4
To the question, then, whether the judgment of right, in its main intention, is an expression of reason, of feeling, or of both, Clarke's answer is unequivocal: it is purely an expression of reason. To see that an act is right is to see that in view of the natures of the persons involved and the character of the circumstances, the act is fitting. Such insight is not only cognition, it is rational cognition, because it involves a grasp of the logically self-evident; it is as intellectual as anything in Euclid.
4. Now that there is much in our moral judgments which answers more or less closely to this description seems to me true. (1) In our judgment that a gratuitous falsehood is wrong there is surely something very like self-evidence. Such a judgment is not like ‘this rose is yellow’, where, for all we can see, the flower might just as well have some other colour, for the wrongness of lying seems to arise out of the character of lying as such. It does not always arise out of ill effects that we can foresee, since even if there is nothing to choose between the foreseeable consequences of lying and telling the truth, we should still find something morally unfitting in wanton deception. Nor does the judgment seem like a deduction from anything else. If someone were to challenge us and say that in these circumstances lying was a perfectly right and proper course, we should suspect that he was being cleverly perverse, and if he proved serious and persisted, we should probably be puzzled how to proceed. The case, as Clarke would remind us, is like that of two and two making four; if someone challenges that, you feel little hope of finding anything clearer or more certain that you can offer as evidence for it.
(2) We feel called upon to make our moral judgments consistent with each other, just as we do our geometrical judgments. When David was exulting over his success in stealing Bathsheba from Uriah, Nathan told him the story of the rich man who stole a poor man's ewe lamb; and when David swore that the man should die for it, Nathan pointed his prophetic finger and said ‘Thou art the man.’ Though David's moral sense was not of the acutest, even he could see that if what the rich man had done to the poor man was wrong, what he himself had done to Uriah could hardly be right, for to say so would be in principle to contradict himself. In short, judgments of lightness would seem to be expressions of reason not only when taken individually but also when taken jointly, in the sense that in making them we bow to the obligation that they should fit consistently together.
(3) They seem to be like mathematical judgments, again, in claiming to be impersonal and objective. We should think it absurd if a man who differed from another as to the sum of a column of figures suggested that on these matters tastes might differ, that we must not be intolerant, and that each of us had a right to his own opinion. Regarding a picture by Matisse, we may resign ourselves to such diversity, though even here we should do so rather reluctantly. But this is certainly not our attitude on moral issues, in spite of all that some uncritical anthropologists have done to encourage us in it. When a popular leader appears who argues that the ‘big lie’ is justifiable if he can get it widely enough accepted, or that breach of treaties is right if there is profit in it for the breaker, we do not regard this as an interesting and legitimate novelty in moral taste; we hold that a man has no right to such a taste; and we think that the man who allows his ambitions and desires to determine his position on such issues shows the same lack of integrity as the scientist who cooks his results to his own advantage. One can no more make such things right by feeling or calling or thinking them right than one can make the column of figures come out according to desire by becoming emotional about it. Against writers like Hobbes, who had held that right and duty were conventions merely, which did not exist in ‘the state of nature’, that conduct first began to be right or wrong only when men made an agreement to obey the law, Clarke's case was crushing. He had only to point out that unless obligation were already recognized before the agreement was made, the very agreement that was supposed to bring it into being would have had no force. The claims that morality make upon us, like those of logic and mathematics, may help to explain our habits of thinking, but they are plainly more than habits of thinking themselves. A habit we can change. We can no more make the gratuitous infliction of pain upon another right than we can square the circle.
5. Such considerations as these make the rationalist reading of moral judgment highly plausible. Yet Clarke's argument proved so unpersuasive that within his own lifetime a reaction set in toward the extreme opposite of his view. The idea that a judgment of right is really the same sort of thing as a judgment in mathematics failed to convince the majority of moralists, and has continued to fail. Why?
(1) For one thing, because the concepts used in moral judgment have nothing like the definiteness to be found in mathematics. Mathematical concepts—notions of two and four and equality, of straight line, circle, and triangle—are as sharp and clear as human notions can possibly be, and as they are exchanged from mind to mind, we can be sure, for all practical purposes, that they remain absolutely the same. But who could say that in judging ‘I ought not to deceive another gratuitously’, the concepts of the two natures, in virtue of which the duty follows, has anything like this precision? Clarke tries to secure this precision for his concepts by making them abstractions and saying that just as the geometer uses, not the concrete figure before him, but the abstract and perfect circle which it stands for, so when we think about moral cases, we are not thinking about Jones and Brown, but about the rational-human-being-as-such displayed in each. But let anyone try to conceive precisely what the humanity is in Brown in virtue of which Jones should tell him the truth, and then compare the result with his concept of a circle, and the contrast will be painfully clear. And where the terms related are vague, the relations that unite them will, as a rule, be correspondingly vague. The precise necessities of geometry would vanish if the lines were not perfectly straight and the circles only moderately circular. Yet this is the sort of unsatisfactory material with which our moral judgments must commonly deal. We are not dealing with abstractions, but with persons, whose differences count. A man would fail grotesquely in his duty if he treated his wife as if she were the Platonic idea of a wife, his friends as if they were types out of La Bruyère, and his community as if it were the eternal essence of all communities, laid up in heaven. That is just not the way in which a morally sensitive person thinks. Hence the rationalist of this type is confronted with a dilemma—So far as he achieves definiteness by skimming the pure cream of abstraction from his terms, the grounds for his ascription of rightness disappear and his thought becomes ethically empty. On the other hand, so far as he preserves those individual differences which are all-important in actual judgment, the mathematical analogy melts away.
6. (2) Again, mathematics is timeless; it considers consequents, but never consequences. Most moral judgments, on the contrary, are passed on temporal acts that have trains of temporal consequences. Such consequences or the pre-vision of them, totally irrelevant in any mathematical context, are of the first importance for judging rightness and wrongness. At first glance it is plausible, as we have seen, to say that if lying is wrong, and self-evidently wrong, it is purely in virtue of men being rational beings. But see where this leads. If the connection is a necessary one, then between such beings lying must be wrong everywhere and always. And we cannot really accept that. It implies, for example, that the theatre manager was doing wrong when, knowing that the theatre, with many children present, was on fire, he announced to the audience in regretful tones that the leading lady had been taken ill and the performance would have to be adjourned, thereby averting panic and perhaps saving many lives. One cannot disregard consequences without breaking disastrously with our actual moral thinking.
Clarke never succeeded in reconciling their importance with his quasi-mathematical ethics. He did point out that many of our duties would remain duties, even though there was nothing to choose between the foreseeable consequences of doing them and not doing them. This was indeed significant, but it was not enough. Putting it on a merely individual basis, suppose a man sees that by doing his duty, he will involve himself in certain ruin or death; does that leave the duty unaffected? Clarke's answer to this showed the strain he was under; indeed it abandons his main position. He admitted that ‘men never will generally, and indeed it is not very reasonably to be expected that they should, part with all the comforts of life, and even life itself, without expectation of any future recompense’.5 But this is to set up two different kinds of reasonableness, one of which lies in doing the duty prescribed by abstract reason, and the other in ordering one's action with reference to the consequences. Plainly these two might conflict. And Clarke had nothing left, no convenient super-reasonableness, to arbitrate between them.
The best he could do was to suggest that if a man did his duty at great cost to himself, he would not, after all, be the loser in the end, since God would see to it in the next life that his sacrifice was duly rewarded. Unfortunately even this retreat was not logically open. For when asked on what ground he believed that God was thus just and benevolent, he gave an answer that made the appeal to this later reward untenable. The ground, he said, was that since the principles of right conduct, like those of mathematics, were self-evident and necessary, they must be God's principles also. But this was to argue that only because it was an absolute duty to disregard consequences did we have any ground for believing that we might reasonably take them into account; reason was still giving two incompatible verdicts. Besides, if God himself was committed to the first kind of reasonableness, why should he regard with favour, and generously reward, conduct that could only be justified by the other kind of reasonableness? Clarke's position thus collapsed. He saw that the appeal to consequences must be allowed; but could find no way of allowing it that did not involve him in incoherence.
7. (3) But perhaps the most obvious objection to this sort of rationalism, and the one that in fact prevailed against it, is that it gives no place to feeling in moral approval. Men remain obstinately convinced that in such approval feeling does have a part, though precisely what, they are not clear. They would be ready to agree that in mathematical judgments it plays no part at all. There is no reason to suppose that a man bereft of the power of feeling would have trouble in understanding what another man was saying as he ran through the multiplication table, since nothing but intellect is involved here. But suppose he heard what people commonly say as they read their newspapers. They read about a captain who sticks to his ship when it is in danger of foundering and does his best to save it, and they use such words as ‘courageous’, ‘noble’, ‘devoted’, in describing his conduct. Or they read of the use by certain governments of drugs, sleeplessness, and torture, to extract from prisoners the sort of confession they desire, and describe such proceedings as ‘base’, ‘wicked’, and ‘dastardly’. Would a man incapable of emotion catch what they meant? If the conduct of the captain or the torturers consisted merely of their bodily movements, he could no doubt follow such conduct with his mind's eye as clearly as anyone else. Bat what would he make of their nobility or their dastardliness? Are these emotionally neutral, like three-sidedness and circularity? Surely not. Feeling seems to have penetrated their very essence. And a man incapable of any glow of admiration or sympathy, or any warmth of moral indignation, who viewed the human scene as if it were nothing but a set of elaborate intersections of circles and triangles, would remain blind to what moral judgments are about.
8. In the second stage of the dialectic, the intellectualism of Clarke, Cudworth, and More was abandoned for an opposite extreme. Shaftesbury and Hutcheson thought the true analogue of moral judgment was to be found not in mathematics but in aesthetics, not in the judgment of necessity but in that of beauty. Both were men of keen artistic sensitiveness. Both were students of Aristotle, and for Aristotle goodness and beauty of character were indistinguishable; he used the compound word καλοκαγαθος to suggest their convergence. The two philosophers, again, were inclined to restrict their thinking about ethics to a single one of the provinces, that of virtue and vice, that is, the state of one's dispositions or habitual affections, and this is the part of ethics—as opposed, for example, to the theory of merit, punishment, or obligation—to which the analogy of beauty most readily applies. Finally, they fixed their attention on that part of the moral act where the analogy is most plausible. No one would say that there need be anything beautiful in the outward embodiment of goodness, in Sidney's act, for instance, of passing the cup to the wounded soldier; the beauty lay within, just as the moral ugliness of Iago lay within, and as the beauty of character as a whole lies in the harmony and proportion of the affections.
What is it, then, that we are attributing to an act when we call it right, or to a character when we call it good? Is it simply beauty? Shaftesbury and Hutcheson often write as if it were. But that is not what they really mean. Goodness is not beauty, though it is the same sort of attribute and arouses in us similar feelings. Can it be defined? Hutcheson, the more exact and careful of the two, says that ‘an action is formally good when it flowed from good affection in a just proportion’. This is not very helpful, since we are moved to ask at once what is then meant by ‘good’ when thus used of an affection, and what is meant by a ‘just’ proportion. It is useless to look to these writers for sharp answers to such questions, and probably if they had been pressed, they would have replied, like many writers of the present day, that ultimate moral qualities are beyond analysis or definition, though not for that reason either unreal or unclear. One cannot define ‘relation’ or ‘time’ or ‘space’ or ‘red’ or ‘sweet’, but most people do not find these notions obscure. Similarly we know perfectly well, these writers would say, what we mean by calling conduct good, even though the notion resists analysis into anything simpler. We are referring to a quality belonging not to outward behaviour, and not to the consequences of that behaviour, but to the attitude of will and feeling that it springs from; ‘we have a distinct perception of beauty of excellence in the kind affections of rational agents’.6
9. What faculty or process is it by which this quality is grasped? Plainly not reason or intelligence. Intellectual gifts are useful in scheming for the future, but to reduce the perception of goodness or badness, as Clarke very nearly did, to the grasp of logical necessity seemed to these philosophers absurd. Men of fine moral perceptions have often been helpless infants in logic and mathematics. What is needed is something more akin to sympathy, though it is not precisely that either, but rather something rendered possible by sympathy. Here again our best key lies in the perception of beauty. What are we doing when we appreciate music? We are hearing sounds, of course, but the beauty of the music does not consist of the sounds, nor even the arrangements of the sounds; it is a quality that belongs to this arrangement, incapable itself of being heard, and yet as indubitably there as the sounds themselves. Since our experience of it is not active, as in thinking, but passive as in seeing or hearing, we may well call it a process of sense; but since it is not a sensing of any material thing or any strictly sensuous content, we may well distinguish it from our other senses as ‘the aesthetic sense’.
Now the perception of goodness is closely parallel to this. The goodness of a man's conduct is not a characteristic like movement that can be actually seen in his behaviour; it is not his affection for others, nor even, if we speak exactly, the harmony or balance of his affections, though the two writers sometimes say such things. It is a non-sensuous quality attaching to his affections. And what better name can be found for our response to it than ‘the moral sense’? Like the sense of beauty, this moral sense is so imbued with feeling that our authors find no harm in describing it as itself a kind of feeling. To perceive the goodness of an action is to approve it; to approve it is to feel approbation for it or to appreciate it; and appreciation is emotion. The pendulum thus swung in a generation from the extreme rationalism of Clarke to its opposite pole in Shaftesbury and Hutcheson. What was approved in conduct was no longer a fitness apprehended by reason, but a quality belonging to the agent's feelings, a quality whose presence we could discern only with the aid of sympathy, and whose apprehension was itself essentially a matter of feeling. The seat of moral judgment had abruptly shifted from head to heart.
10. The next step in the dialectic was taken by a philosopher of greater stature than any of these men, David Hume. Hume is sometimes thought of as an arch-destroyer. And for those whose allegiance is to reason in ethics and metaphysics, the description is no doubt justified. Certainly in ethics he is much nearer to Hutcheson than to the rationalist Clarke. Still, he departed widely from both, while partly accepting both; if our dialectic were Hegelian, Hume would represent a synthesis of which Clarke and Hutcheson were thesis and antithesis. And since his position is a highly sophisticated one which is still thought by many philosophers to be in all essentials sound, it will be well to look rather closely at the roles that he assigned to reason and feeling in moral approval.
Hume held as firmly as Shaftesbury and Hutcheson did that the rightness of conduct could not be perceived or established by reason. In writing on ethics, both in the Treatise and in the Inquiry, he makes it his first concern to show this, and he offers a variety of arguments for it. None of these, singly or taken together, seem to me conclusive. But Hume's authority has carried so much weight that we shall do well to examine at least those of them that he regarded as most decisive.
11. (1) First, to say that the rightness of action is perceived by reason, he argues, is equivalent to saying that an insight of reason can move us to act, but since it is clear that no such insight can move us to act, the perception of rightness cannot be a rational one. This must be expanded a little. Suppose I see that I ought to do a certain act, say keep a promise, and suppose this ‘seeing’ is a rational process. Then, says Hume, we are asked to believe that this purely rational perception of duty can serve as a motive for doing my duty. If this were not true, if we could never say that we did something because we saw we ought to do it, then rational conduct, in the sense of conduct ordered by reason, becomes impossible; and it is for the rational ordering of life that rationalist writers on morals are pleading. Very well, is it the case that knowing something to be our duty does at times make us do it? No, Hume answers, never. By thinking about keeping our promise, we may see that keeping it is bound up with much that we want—the confidence and goodwill, for example, of the man to whom we have made it, but that which moves us to action is never anything that thinking reveals to us; it is the chance existence in us of desires for what thinking may disclose. If we felt no desire for the confidence of others, or anything else that is bound up with keeping promises, knowledge, even if omniscient, would not produce the faintest flicker of action; ‘reason of itself is utterly impotent in this particular’.7 Since, then, the rationalist view of how duty is perceived entails that this perception should be able to move the will, and this implication is false, that which entails it must also be false, and the rationalist view must be rejected.
In spite of the stress Hume placed on it, this argument is quite unconvincing. (a) The entailment on which it turns is no necessary part of the rationalist case, and could hardly have been thought so except by confusing different meanings of ‘approval’. If, in considering whether to keep a promise, I approve the proposal, that may mean, and often did mean indiscriminately among earlier writers on ethics, either (i) the recognition that the act would be right, (ii) the feeling of an ‘approving’ emotion toward it, (iii) the desire to do it, or (iv) the actual election of it in practice. Now there is no reason why a rationalist should not hold that (i), the recognition of the act as right, is quite distinct from the others, and indeed may be present while any or all of the others are absent. It is perfectly possible to say that my recognition of an act as the right one is a purely intellectual affair, but that I shall not go on to do it unless moved by something else, such as a feeling, desire, or resolution. What the rationalist is maintaining is that the discernment or recognition itself is the work of reason, and one does not show this false by first making it mean much more than it need mean, and then refuting this gratuitous addition.
(b) Such plausibility as Hume's argument has is largely gained by his choice of examples. When the rationalist is told that, on his own theory, the perception that X is his duty is necessarily connected with an impulse to do it, he may be tempted to say, ‘Yes, this does seem to be true and is surely harmless; why shouldn't I believe it?’, and only when he is securely entangled does he realize that he has strayed into a corner where Hume's web has been spun for him. But he need not have strayed there at all. If, instead of talking about the judgment of duty he had talked about the judgment of right, as applied to other people or to his own past, Hume's criticism would have had small plausibility. The judgment that one's past keeping of a promise was right, or that some tennis player at Wimbledon did right when, as reported in the morning paper, he threw away a point, is plainly not connected with any election on my own part to do the act approved. Here the insight and performance are sharply disconnected. If Hume's discussion had focused upon judgments of this kind instead of those of obligation, the rationalist would have perceived more easily the illicit burden of implication that had been foisted upon him.
(c) Hume is assuming in his argument that reason in fact never does move the will, and he assumes this on the strength of evidence given in his book on the passions. When one turns to this book and studies the evidence,8 one again finds it indecisive. To review that evidence would take more time than we can afford. But it is worth pointing out that unless this often questioned and somewhat questionable premise is true, the argument obviously fails.
12. (2) Hume maintains, in the second place, that if the perception of rightness is a matter of reason, we shall have to say that rightness is the same as truth, and this is absurd. What reason deals with is propositions which are capable of being true or false, of being relevant to each other and contradicting each other. A proposition about rightness may be true or false; but if rightness or fittingness is, as rationalists claim, something belonging to the act itself, then its inherence in the act is no more true or false than the inherence of juiciness in an apple or of hardness in a stone; it is merely a fact which, as neither true nor false, is outside the province of reason. The status given to rightness by the rationalists is thus inconsistent with the way in which they allege we know it. If it were merely a relation among ideas, they might well hold that it is known by reason. But they cannot maintain at once that it is a fact or relation in the nature of things and also known by reason.
This is a preposterous argument. If, when we know that A is B, we are knowing only a proposition and not a fact, then we can never know that our proposition itself is true. For in order to know that, we must, according to Hume, be able to perceive the agreement or correspondence between proposition and fact, and in order to do that, in turn, we must be able to see the arrangement in fact to which our proposition corresponds. Unless we can do this, we can have no knowledge, properly speaking, of the world at all. But Hume admits that we can grasp these arrangements in fact; in the very passage in which his argument is stated he admits that propositions may be compared with the real relations whose existence they assert, and he would not deny that reason can perceive the numerical relations among fingers and toes. But if it can do such things, it is idle to say that it is confined to the sphere of propositions or truth as distinct from fact, and idle too to say that if rightness is rationally grasped, it must fall outside the realm of fact in the limbo of the merely ideal. Thus Hume's argument proves too much. If the mere knowability of a relation is sufficient proof that it does not belong to the nature of things, we can never know or hope to know anything; and this involves a scepticism from which even Hume recoiled.
13. (3) But his chief reliance in the argument against rationalism lay in a consideration of a different kind. This was the failure of the rationalists to make clear what they meant by the rightness grasped by reason. Rightness, they said, consisted in the fittingness of act to situation, and this fittingness either was, or depended on, a relation of some kind; but then of what kind? Nero murdered his mother; there was an act that everyone would agree to be wrong; the rationalists would say that it was self-evidently wrong, and that the wrongness lay in some relation that it set up between agent and patient. But when they attempted to abstract this relation and place it in a clear light, it belied their claims, for it always turned out to be a relation that was morally neutral. In the case just mentioned it is obviously so. Abstractly considered, what is the relation established by Nero through his action? It is that of an offspring's exclusion from existence of its parent. Now if it is this relation as such that makes the action wrong, it must do so wherever it appears. And this it plainly does not. Here Hume introduces his famous analogy of tree and sapling:
‘To put the affair, therefore, to this trial, let us choose any inanimate object, such as an oak or elm; and let us suppose that, by the dropping of its seed, it produces a sapling below it, which, springing up by degrees, at last overtops and destroys the parent tree: I ask if, in this instance, there be wanting any relation which is discoverable in parricide or ingratitude? Is not the one tree the cause of the other's existence; and the latter the cause of the destruction of the former, in the same manner as when a child murders his parent?.… the relations are the same: and as their discovery is not in both cases attended with a notion of immorality, it follows that that notion does not arise from such a discovery.’9
This result, Hume argues, is typical. When the relation in which the rightness or wrongness is supposed to consist is set forth nakedly, it always turns out to be one that holds also among objects where no morality exists. It cannot, therefore, be the locus of rightness or wrongness where morality does exist.
This again is quite unconvincing. The difference between the two cases far outweighs their likeness. There is, to be sure, an abstract biological nexus between sapling and oak which appears also between Nero and his mother, and therefore each can be described as the destruction of parent by offspring, but no rationalist ever held that what made Nero's murder of his mother wrong was the thin abstraction gained by omitting all the differences between these cases. The relation between a son who knows what he is doing and a mother with memories, affections, and hopes, is so utterly different from that of a non-sentient sapling and tree, that for moral purposes the analogy between them is simply non-existent. The relation between the human beings is different because the terms of that relation are different, almost as different as terms can be.10 To call it the same as that which holds between the trees would be like saying that when a man and a flower both seek the sunshine, or nod their heads, or curl up and sleep, they are both engaged in the same activities. Rationalists hold, it is true, that there are types of virtue and vice which embrace many resembling cases, so that if it is wrong for A to deceive B under circumstances x, y, z, it is wrong for C to deceive D under similar circumstances. But the theory that right and wrong are so independent of circumstances that if it is wrong of A to deceive B, it must also be wrong for him to catch a fish by deceiving it or to tell a lie to the lamp-post, is a theory that no sane rationalist would for a moment acknowledge as his own.
14. Hume considered these arguments, and a few others of less weight, to be decisive against the view that we perceive the rightness of conduct by reason. We have seen that they are very far from decisive. But we must now insist on the other side of Hume's attitude toward rationalism. He never meant to say that reason has no place at all in moral judgment. It does not, to be sure, have the exclusive place; it does not even have the chief place; but it does have a humble and useful office of its own. What is that office? It is to place an act in perspective so that the judgment finally rendered on it may be circumspect and well informed. When a man is choosing a profession, the final vote is cast, Hume would say, by his desire or sentiment. ‘But in order to pave the way for such a sentiment, and give a proper discernment of its object, it is often necessary, we find, that much reasoning should precede, that nice distinctions be made, just conclusions drawn, distant comparisons formed, complicated relations examined, and general facts fixed and ascertained.’11 This is reason's work. Nearly every choice we make has its effects for better or worse on our own or others’ lives; it is the business of the responsible man to foresee these effects, so far as important and relevant, and to make his choice in the light of them. Only by the use of reason can this provision be achieved. Thus, while the rationalists were mistaken in holding that reason itself pronounces the moral judgment or achieves the distinctively moral insight, they were correct to this extent, that reason is the indispensable adjunct and aid to such judgment. I recently saw a Scottish farmer appear in a large field dotted with sheep among which he wanted to make some sort of selection. He had at his heels a well-trained sheep dog, and he had only to give this trusty lieutenant a word to produce a performance astonishing in its thoroughness, speed, and expertness; in a very few minutes the scattered flock was rounded up at his feet. Reason in Hume's theory is the sheep-dog of the moral faculty. Its business is to use all possible skill in rounding up the relevant facts through memory, analogy, and prediction, and presenting them before the judge as the material for his decision. Hume would not have denied that this was an extremely important service. He would only have added that the sheep dog makes itself absurd if it tries to usurp the place of its master and do the judging also.
15. The question then presents itself, who is this judge and master? Here Hume sides with the ‘sentimentalists’. But just as his rejection of Clarke was carefully qualified, so was his acceptance of Shaftesbury and Hutcheson. They had said that the goodness of a man or of his conduct—and they tended to identify the two—was a special kind of quality, not sensible like a colour or sound, and not conceptual either, if that meant apprehensible by reason, like circularity. As we have seen, they thought its closest analogy was with the beauty of a picture or a statue. Hume will have none of this. There is no such quality as beauty. There is no such quality as goodness or rightness.
‘Euclid has fully explained all the qualities of the circle; but has not, in any proposition, said a word of its beauty. The reason is evident. The beauty is not a quality of the circle. It lies not in any part of the line whose parts are equally distant from a common centre.’12
It is the same with moral qualities.
‘Take any action allowed to be vicious, wilful murder, for instance. Examine it in all lights and see if you can find that matter of fact, or real existence, which you call vice. In whichever way you take it, you find only certain passions, motives, volitions, and thoughts. There is no other matter of fact in the case. The vice entirely escapes you.…’
One might be tempted to discount Hume's argument here as one that would be made automatically by anyone who held his theory of knowledge, since according to that theory, all our ideas are copies of what comes to us through sensation; hence if anyone alleged an idea of what was not sensory in character, it would be ruled out a priori. This mechanical process may have operated in Hume's mind, as one is quite sure it has in the minds of some of his successors. But it would be unfair to dismiss his argument on such grounds, for we find it startlingly strong on its own merits. Try to point out the beauty of a rose as a character distinct from the factual qualities and relations that compose it, and you find the sought-for character vanishing perpetually in mist. Try to isolate in your thought a distinct quality of goodness belonging to an experience, or a distinct quality or relation of rightness belonging to an act, and again you are puzzled, tantalized, and baffled. We seem to assume that a rose is beautiful very much as it is red, and an act right in the same obvious way as it is brief or violent; but whereas we can isolate these other qualities, on reflection, with complete clearness and ease, we find the beauty, goodness and rightness so hard to pin down, so impalpable and elusive, that we are likely to end by capitulating to Hume in helpless bewilderment. It seems strange to say that qualities which artists and moralists have talked about, or supposed themselves to be talking about these two thousand years and more are so many ghosts that have never existed at all. But hard as it is to believe this, it seems even harder to meet Hume's challenge to capture one of these ghosts and produce it for inspection.
16. Of course Hume does not deny that we say anything at all when we talk about goodness and beauty. What, then, are we saying? His own answer, as usual, is admirably clear. We are not declaring something about the make-up of roses or persons or actions; we are reporting our feelings toward them. We are talking less about them than about ourselves. Murder is wrong, we declare roundly, and then, under an ancient and general delusion, we may start examining the act to find where the wrongness lies. But this is hopeless.
‘The vice entirely escapes you as long as you consider the object. You never can find it till you turn your reflection into your own breast, and find a sentiment of disapprobation which arises in you towards this action.… It lies in yourself, not in the object. So that when you pronounce any action or character to be vicious, you mean nothing but that from the constitution of your nature you have a feeling or sentiment of blame from the contemplation of it.’13
To this the rationalist may retort, ‘You are really conceding, after all, that the moral judgment is a “pronouncement”, it’ reports the occurrence of approbation or disapprobation, and hence is capable of truth, and hence again is more than a feeling; it is, or may be, knowledge’. Hume would not be troubled. He would agree that the ‘pronouncement’ was a judgment, but would point out that the act which it reports, the actual moral estimate or appraisal, is not a judgment, is not true or false, and has nothing rational about it; it is a response of pure feeling. And it is this feeling that is the essence of the matter. The judgment only echoes it, and without the feeling would be empty, while the feeling could, and often does, occur without the judgment. The ‘insight’ or ‘perception’ that an act is right is simply that warm feeling about it which we call approbation. And the object of this approbation is the act itself, not some illusory quality of rightness in it; the rightness has now died as a quality, and been resurrected as an attitude of the beholder.
17. Must we say, then, that no moral approval is more reasonable or unreasonable than any other? Strictly speaking, yes. Hume is willing to admit, however, that approval may be reasonable in a metaphorical sense. The act that one approves may be conceived thinly or thickly, fragmentarily or in the context of causes and effects, and its degree of ‘reasonableness’ varies accordingly. The judge who presides at a murder case studies the act in the context of the agent's motives and provocations, and the record and character of the victim; his final response is to the act in an extensive matrix of circumstances; and it is in that sense, therefore, a far more reasonable response than a first impulsive reaction.
Still, the fact that all this is taken into account does not make the moral response, when it comes, one whit less purely emotional. The approval or disapproval of an act by the most reflective of justices is itself as entirely a matter of feeling, and as completely non-rational, as the gurgle of a child over a bouncing ball. We can thus see what Hume meant when he said that in moral matters ‘reason is, and ought only to be, the slave of the passions’;14 it would be helpless by itself, and there is nothing else for it to serve. We can see too that if no moral choice is an affair of reason at all, it clearly follows that no such choice is, strictly, more reasonable than any other. Hume not only draws this conclusion, but drives it home relentlessly.
‘It is not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. It is not contrary to reason for me to choose my total ruin to prevent the least uneasiness of an Indian, a person wholly unknown to me.… In short, a passion must be accompanied with some false judgment in order to its being unreasonable; and even then it is not the passion, properly speaking, which is unreasonable, but the judgment.’15
This position is somewhat shocking to common sense; and to Hume's reputation as arch-destroyer in metaphysics it added a like reputation in ethics. But he was far too massive a presence to be exorcized by a lifting of eyebrows. His case, when it came to be examined, proved disconcertingly strong. Indeed, some of the clearest-headed philosophers now living believe that it has not yet been effectively answered, for the very good reason that it is in essentials unanswerable. Whether we agree with this or not, it would be holding Hume at less than his true worth if we took as the antithesis to him, in the pendulum movement we are tracing, any of the legion of little figures who rose to shake their fists at him. Price, to be sure, was not a little figure, and he has been unduly thrown into the shadow by Hume. Still, we shall do well, I think, to look farther forward to a critic who could meet the formidable Hume on more even terms. Such a critic we undoubtedly have in Henry Sidgwick.
18. C. D. Broad startled some readers by his judgment that Sidgwick's Methods of Ethics is ‘on the whole the best treatise on moral theory that has ever been written’,16 but it is a judgment with which, so far as my own reading goes, I should concur. Some of the conclusions of this work, most notably its hedonism, have not carried conviction to most students, certainly not to me. But for combined subtleness, thoroughness, lucidity, and fairness, I know of no equal to it in ethical literature. Writing on ethics has often gone to one or other of the two extremes of homiletics and logic-chopping, of so strong a desire, on the one hand, to make the good prevail as to lead to a neglect of rigorous analysis and, on the other, of so strong a preoccupation with analysis that one loses sight of ethics in a tangle of logic and linguistics. Sidgwick's acuteness was equalled by his sanity and moral seriousness; and for judicial detachment—the somewhat bleak, but clear, full light in which he sees things—he stands quite alone, so far as I know, in philosophic history. It is a sad thought, but an inevitable one, that the interesting people of the world are as a rule unreasonable ones, and that a man who never raises his voice, or allows impulse to take him a hair's breadth out of the line of fairness and good sense, is bound to seem dull. To most types of reader, Sidgwick is irredeemably dull, while writers with far feebler intellectual powers who speak in tones of prophesy, like Nietzsche and Kierkegaard, are listened to with a respect they ill deserve. But for those who want to know simply what ethical theories make sense and what do not, and who are bored with attempts to make the subject interesting, Sidgwick's book is supreme.
It is a main thesis of this book that moral judgments are an expression of reason. Most men assume this without question, and though they may be mistaken about this as about other things, Sidgwick holds that one who differs from them must assume the burden of proof. That they do hold this conviction is clear enough; ‘we commonly think that wrong conduct is essentially irrational, and can be shown to be so by argument; and though we do not conceive that it is by reason alone that men are influenced to act rightly, we still hold that appeals to the reason are an essential part of all moral persuasion…’17 In a sense, of course, even Hume would agree with this. He admits, as we have seen, that reason is an indispensable tool for bringing before our minds both means to our ends and alternative ends themselves. Sidgwick considers whether this is all that reason can do in such judgment, and concludes, that it does much more. His own views on the office of reason are scattered widely through the five hundred pages of his book, but it will not be unfair, I think, to bring them together under two heads, his grounds for rejecting Hume's emotionalist view, and his own alternative to it.
19. (1) If Hume's view is correct, approval is a feeling, and the judgment that an act is right or a duty is a statement that the feeling is actually felt toward it—usually in the light of its consequences—by oneself or others. According to Sidgwick this misreports what we mean to say. Roughly, what we are really asserting is an ought, and this takes us to be asserting an is. Suppose that three hospitals ask me for a contribution, and I try to make up my mind which, if any, to contribute to, and how much. I conclude that I should give a certain amount to C, and none to the others. Now if Hume is correct, what I have been trying to find Out is the actual state of my feelings toward various courses, and when I conclude that giving a certain amount to C is right, I am reporting the state of my feeling about it. Sidgwick thinks that by merely inspecting our meaning we can see that this is not so.
He would say, I think, that Hume's interpretation mistakes both the subject and the predicate of our actual judgment. Our subject is the suggested act, not ourselves or our feelings about it. If we were talking about our feelings, the report of an expert psychologist as to what is actually felt would be grounds for calling the judgment true or false, whereas in fact we should take such a report as irrelevant. Again, the predicate we ascribe to the act is not descriptive of any existent thing. Hume was correct in saying that rightness was not this kind of predicate; he was mistaken in saying that if a predicate was not of this kind, it was not a predicate at all. Sidgwick held that ‘the fundamental notion represented by the word “ought” or “right” is essentially different from all notions representing facts of physical or psychical experience’.18 It was the notion of what more recent ethicists would describe as a ‘non-natural characteristic’, one that made no part of nature as investigated by the natural sciences. And unfortunately it was ‘too elementary to admit of any formal definition’.19 This may raise suspicions about it. But so far as these suspicions are based on an antecedent conviction that no such character is thinkable, Sidgwick would be little moved by them. In this matter he was empirical without being an empiricist; that is, he was trying to find what actually was in our thought, without any preconceived theory, drawn from the supposed origin of knowledge, as to what this must be. His first argument, then, against the emotionalist view was that if it were correct, a moral judgment would be an assertion of fact—in this case of psychological fact—and that in such typical moral judgments as that of the rightness of some act yet in prospect, we were not asserting about fact at all.
Sidgwick offered other arguments, applying some to judgments of right, and others to judgments of intrinsic good. Since for him both types of judgment were in the same sense rational while for Hume they were both non-rational, we may consider these arguments without specifying in each case the context in which they were offered.
20. (2) If Hume is correct, when I judge ‘that act is right’, I am reporting not a character of the act, but a feeling of my own. (He also held that I am reporting the feeling which prevails in my society, shifting from the one view to the other as suited his argument, without seeing, apparently, how different they were).20 Then when I say, ‘truth ought to be spoken’, and you say ‘truth ought not to be spoken’, we are not expressing conflicting opinions, but only reporting two coexisting facts about our feelings. There is no conflict of opinion at all. Sidgwick thinks this paradoxical and untrue.
21. (3) Indeed he thinks it so obvious that we are really expressing conflicting opinions that he believes the emotionalists must have something else in mind than what they say. They can hardly intend that we do not mean to express opinions about truth-speaking at all, for we do very plainly mean to; they must be saying that, if we knew what we were about, we should see that we could not sensibly mean to, that the ‘subjective fact of my approbation is all that there is any ground for stating, or perhaps that it is all that any reasonable person is prepared on reflection to affirm’.21 Is this, then, a tenable view?
No, Sidgwick answers, it is still untenable. There are cases, to be sure, regarded by writers like Hume as similar, where this sort of analysis might serve. If I call a certain food disagreeable, and my remark is challenged, I may content myself with insisting only that I dislike it. ‘But there appears to me to be a fundamental difference’, writes Sidgwick, ‘between this case and that of moral feeling.’ What is the difference? It is this, that whereas in the case of the food, my meaning is, or may be, exhausted in the report of how I feel, in the moral case it is never so exhausted; besides the feeling, I am pointing to something in the object that is taken to justify the feeling. I am not simply asserting that I happen to feel in a certain way; I am asserting—if indeed there is any reference to feeling at all—that I feel in this way because, independently of my feeling, the act is right. ‘The peculiar emotion of moral approbation is, in my experience, inseparably bound up with the conviction, implicit or explicit, that the conduct approved is “really” right—i.e. that it cannot, without error, be disapproved by any other mind.’22 In short I do not call the action right because I feel in a certain way; I feel in this way because I think the action right. And Sidgwick goes on to show that where I do not think the action right, this is reflected in a difference in my feeling about it. If one became convinced that on a certain occasion—on which, for example, one was invited to disclose the secrets of an atomic bomb to an enemy—telling the truth was wrong rather than right, one's feeling about it would be ‘quite different in kind and degree’. Feeling is distinct from, but follows and provides an index to, the intellectual conviction.
22. (4) Hume was in a sense a utilitarian; he thought we called an action right in virtue of the pleasure we believed it normally produced. But if we asked him what we meant by calling the pleasure good, he would say that we were referring not to some quality of worth or value in the pleasure, but, once more, to our liking or desire for it. This liking or desire was the goodness, in the only intelligible sense, and hence its variations in intensity gave the measure of goodness. Sidgwick saw that this theory could not be worked out consistently with our actual ways of thinking, and he criticized the theory through that very aesthetic analogy which was so favoured by emotionalist writers.
We call a musical composition good, for example. Do we call it so in virtue of any kind of pleasure that we may happen to like, e.g. in blasts of loud noise or tom-tom regularity of beat or the faithful imitation of a cuckoo, or, on the other hand, in virtue of a more special satisfaction, a satisfaction derived from music that conforms to a certain ideal? And if the answer is the latter, as it apparently must be, can we rule out conformity to the ideal and fix on our feeling alone as contributing to the goodness? If pleasure, or more strictly, the satisfaction with which we contemplate it, is to serve as the index of aesthetic goodness, it must be the kind of satisfaction that is taken in a particular kind of object; and the same holds of our satisfaction with moral goodness.
Again, when I am talking about my own feeling, it is clear that I am myself the final judge; if I say that I take satisfaction or pleasure in something, I have no doubt of what I am saying, and should think it absurd to suggest that there was someone else who knew better. But when I venture to say that a new work of Stravinsky is good, I am surely making no such claims. I am quite ready to admit that, whatever my feelings about it, I may very well be mistaken. Hence to say that something is good is plainly different from saying that I take pleasure or satisfaction in it.
Nor is it true that the degree of goodness attributed corresponds to the intensity of feeling; ‘connoisseurs of wines, pictures, etc., often retain their intellectual faculty of appraising the merits of the objects which they criticize, and deciding on their respective places in the scale of excellence, even when their susceptibilities to pleasure from these objects are comparatively blunted and exhausted’.23 Indeed, there may be savages who take as intense a delight in their tom-toms as ever Schweitzer in Bach. But we should hardly take that as conclusive evidence that what they delight in is equally good. A similar line of argument applies to moral appreciations.
Sidgwick makes further relevant comments when he comes to Hobbes's suggestion that ‘whatsoever is the object of any man's desire, that it is which he for his part calleth good, and the object of his aversion, evil’. In essentials, this is Hume again, since it lodges goodness in the attitude rather than the object. The obvious difficulty with it is that a man sometimes desires intensely what at the same time he knows or believes to be bad, for example, ‘revenge, when he knows that his true interest lies in reconciliation’. If being desired is all we mean by ‘good’, how can we call this bad while aware that we desire it desperately? Furthermore, something that we very much want ‘may not be found good when fruition comes’; ‘it may turn out “a Dead Sea apple”, mere dust and ashes in the eating’. Is the same thing, then, both good and bad, or good for a while and then bad for a while? It certainly accords better with ordinary thought to say that at first we thought it was good, but that it turned out to be really bad. Finally, many things are out of our power and we should consider it rather foolish to go on desiring them when we know they are unattainable—such things as perfect weather and perfect health. But we do not necessarily regard them as less good because our desires for them have lost the intensity they once had.
23. For these reasons Sidgwick rejects the view that judgments of right and good are either feelings or reports of feeling. Moral judgment always contains a rational element. What is this element? Here we come to the theory which Sidgwick would offer as his alternative to Hume. His suggestions, as before, are scattered widely in his book, but they are of such interest and importance, that we must try to assemble and summarize them. Professor Broad has said that reason
‘includes three cognitive powers, viz. (a) the power of forming a priori concepts, i.e. concepts of characteristics which are not sensibly manifested in any instance and are not composed of characteristics which have separately been sensibly manifested in various instances; (b) the power of recognizing that a conjunction of attributes is an instance of a necessary connexion between these attributes, i.e. the power of Intuitive Induction, as Mr Johnson calls it; and (c) the power of inferring conclusions from premises.’24
Sidgwick was a rationalist in the sense that he believed that reason in all three of these meanings entered into our moral judgments. All moralists would probably agree that in forming such judgments we use reason in the third sense, in which it is equivalent to reasoning. It is Sidgwick's insistence on the other two that distinguishes his rationalism.
(1) He believed, as we have seen, that neither rightness nor goodness was a character in the natural world, that is, the sort of character that could be sensed, or occupy space. Value is not existence; ought is not reducible to is. Right and good are a priori, not in the sense that we impose them on objects in knowing them, for they may belong to acts or objects without being known to do so, but in the sense that they are not given empirically, and must be intellectually grasped. It may be protested that such characters are mere wraiths, that there is nothing in the real world answering to such a description. But is this true? No one doubts that things are really similar or dissimilar to each other, but similarity and dissimilarity are not themselves sensed. There may be a dozen eggs in a nest, but twelveness is not an empirical characteristic. Sidgwick's convictions about the non-empirical nature of rightness and goodness was the more striking because of his hedonism; he held that what made an act objectively right was the fact that it produced at least as much pleasure as any possible alternative act, and that what made an experience good was the pleasure it contained; and both these points must be empirically determined. But you could not identify an act's being right with its producing pleasure, for then when we said that the act producing the greatest pleasure was right, we should be saying nothing more than if we said that the act producing the greatest pleasure produced the greatest pleasure, and Sidgwick was clear that we were saying more. He took a like view about good. These moral attributes were worths or values, and values, while attaching to empirical objects, must be non-empirically known.
24. (2) He held, again, that in morals we have a variety of insights that must be called rational because they are necessary and universal. These insights were of three kinds.
(i) First there were insights, some absolute and some comparative, about intrinsic goods. (a) Sidgwick thought that everyone makes absolute judgments about what is good, whether he admits this or not. If you ask a man why he buys a hammer, he may answer, ‘to drive nails with’; if you ask him why he wants to drive nails, he may say ‘to put up a shelf; if you go on asking why, he may answer successively ‘to keep some books conveniently at hand’, ‘to facilitate my work’, ‘to increase the pleasure in my work’. If you ask him why he should want pleasure in his work, he would regard that as a foolish question, because if ‘why?’ means ‘for what further end?’, as it did in the earlier questions, there is no answer to it; pleasure is the end of the line; it is wanted not because it is good for something, but because it is good in itself. Of course many other intrinsic goods have been alleged, and no one can avoid believing in some such good, for if we only wanted a for the sake of b, and b for the sake of c, and so on without end, there would be nothing that we wanted for itself, and the chain would end in the air. But though the question ‘why?’ has no answer here if it asks for some ulterior end, it is not merely pointless, for there is another way of justifying our wanting something besides showing what it leads to, namely showing that it is good in itself. We have this insight when we see, for example, that pleasure, by reason of being the sort of experience that it is, is good, and that intense pain, for the same reason, is bad. Sidgwick would say that it was a muddle to confuse such an insight with an empirical one, such as ‘This lump of coal is black’. In the empirical judgment there is no insight that the coal must be black, while in the judgment about the intense pain, we do see that being bad is entailed by the nature of the pain; and again, the empirical judgment would at best entitle us to say of another lump of coal that it would probably be black, while we are prepared to say of any pain as intense as this that it will be bad. In short we are seeing a connection which, because necessary and universal, is as truly an insight of reason as the perception that shape entails size or that the diagonal of a square divides it in half. Indeed Sidgwick thought we could see by intuition ‘that happiness is the only rational ultimate end of action’.25
(b) He would say that we have a similar kind of insight when we see that a is better than b. If one compares the pain of a finger caught in a door and the pleasure of a cool drink when thirsty, one sees that the second is better, more worth having, than the first, and here again, since, given the nature of the terms, the relation must be what it is, the insight is not contingent but necessary. At this point one may anticipate a certain kind of objection. Do not people differ in their valuations, and if they do, is it not the height of dogmatism for anyone to claim necessity for his own insight? This is a tiresome objection. For (a) there is no reason why one who holds that judgments of this kind can and do occur should hold that we never make mistakes about them, or that one's own judgments are infallible. Sidgwick was peculiarly alive to the difficulties of distinguishing what we may call certainty, in the sense of self-evident necessity, from certitude, in the sense of emotional conviction; he laid down an admirable list of requirements that must be fulfilled before he would concede the former status to a proposition; and no one could be more cautious or less dogmatic than he in intellectual practice. (b) It is worth pointing out that those who object to insights of this kind do not themselves dispense with them; they only find them in other places. They insist on empirical evidence or demonstrative proof, forgetting that empirical evidence can never render a general proposition more than probable, and that the rules of probability themselves rest on the kind of insight here in question; or else forgetting that demonstration depends on logical principles, and that these too must be established, if at all, by insight of the same kind.
25. (ii) Sidgwick thinks that we may have a similar rational insight into the connection of right with duty. Suppose that, seeing my neighbour's house on fire, I recognize that my duty is to try to save it; suppose I recognize, too, that the only way to save it is to call the fire department. Sidgwick would say that in this process reason is at work in two further ways. (a) First, if I recognize that in this situation an effort to save the house would be the right act for a normal man, I see also that it is my own duty to make the attempt. To see that the act would be the right one and that it is my duty to do it are not quite the same though the first entails the second. (b) Again, ‘it can hardly be denied’, Sidgwick says, ‘that the recognition of an end as ultimately reasonable involves the recognition of an obligation to do such acts as most conduce to the end’.26 Thus if I recognize that saving the house is the most desirable thing that at the moment I could do, it would be unreasonable of me not to adopt the means of doing it. But reason is involved here in a somewhat different way from that just noted. There, the fact that an action would be the right one entailed that I should try to do it. Here, does the fact that I should try to do it entail that I should try in this particular way? Certainly there is no seen necessity in saying that, if the house is to be saved, it must be saved through my calling the fire department. That would be a causal necessity, not a necessity of the kind in question. Sidgwick does not help us here, but he would probably say that the perception that I should try to save the house, jointly with the belief that calling the department was the best or only means of doing so, entailed that I ought to call it, or try to call it. If this seems complicated and pedantic, we may say instead merely that it would be inconsistent to recognize a duty and yet to reject the means of performing it. And to perceive an inconsistency is as truly rational an insight as the grasp of necessity. (c) There is still another rational insight that Sidgwick thinks we have about duty. We perceive that ought implies can. ‘I cannot conceive that I “ought” to do anything which at the same time I judge that I cannot do.’27 If something is really our duty, it is something we are able to do; if we are not able to do it, it cannot be our duty. These are two ways of stating the same necessary connection.
26. (iii) We come now to another and final set of insights which Sidgwick considers rational because universal and necessary, his axioms for distributing good. Suppose that, seeing pleasure to be good, and believing that in a certain situation we should try to produce as much of it as we can, we find that, to all appearances, we can produce the same amount of it in different ways. Take a homely example. If we are holding a children's picnic and have a limited amount of ice cream, we may consider that an equal amount of pleasure would be produced by giving equal shares to each, and by giving it all to a clamorous and ravenous sub-section without letting the others know. Most persons would cast a critical eye on this last procedure. If so, it is obviously not enough to know that we should choose the largest attainable good; we must also have some principles for distributing it. Do we have any such principles? Sidgwick thought we did, and that, so far as they went, they were self-evident and certain. He called them axioms of prudence, justice, and benevolence. (a) Prudence had to do with one's own good, and here he found two self-evident principles. (a) It is a duty to elect one's greatest good rather than any lesser goods. Some moralists, for example Kant, have thought it absurd to say that we ought to do what we should in any case do instinctively, and have taken the pursuit of our own happiness as the prime example of this. Sidgwick admits the oddity of speaking of a duty to be happy, and agrees that we usually think of duty as running counter to inclination, but he believes also that duty and reasonableness coincide, and that it is sometimes as unreasonable not to follow inclination as, at other times, to yield to it. (b) Suppose one must choose between a lesser but nearer good and a greater but more remote one? In such a case, one sometimes chooses the nearer on the ground that there may be many slips between the remoter cup and one's lip, and this is reasonable enough. But if the two are equally certain, it would be foolish to choose the smaller one merely because it is the nearer; it is self-evident to Sidgwick that ‘the mere difference of priority and posteriority in time is not a reasonable ground for having more regard to the consciousness of one moment than to that of another’.28
(b) He finds still other self-evident principles in the field of justice. (a) The Golden Rule is sometimes taken as such a principle, ‘Do to others as you would have them do to you.’ But he sees that this will not pass as it stands. There are many persons who would be glad to co-operate with you in thievery and have you co-operate with them in return. Still, the Golden Rule is feeling after a principle which, when precisely stated, does seem to be self-evident, though it is to be feared that it suffers as much in form when translated precisely as does the King James Bible. Sidgwick reformulates it as follows: ‘it cannot be right for A to treat B in a manner in which it would be wrong for B to treat A, merely on the ground that they are two different individuals, and without there being any difference between the natures and circumstances of the two which can be stated as a reasonable ground for difference of treatment.’29
(b) A similar principle crops up when we are considering, not what others may claim for us, but whether some action of our own is justifiable. Here the taking of an impersonal point of view may be enough to reveal to us that an act is indefensible; ‘we see that we should not think it right for another, and therefore that it cannot be right for us.’30 Kant held that the rule of subjecting our conduct to this impersonal test was the supreme rule of right-doing, and though Sidgwick demurs, he does think it self-evident that ‘we cannot judge an action to be right for A and wrong for B, unless we can find in the natures or circumstances of the two some difference which we can regard as a reasonable ground for difference in their duties’.31
(c) Finally, Sidgwick finds that we have self-evident duties under the head of benevolence. (a) If I believe that the welfare of others is good, as well as my own, it will be unreasonable of me to fix my eye on my own exclusively; ‘I undoubtedly seem to perceive, as clearly and certainly as I see any axiom in Arithmetic or Geometry, that it is “right” and “reasonable” for me… to do what I believe to be ultimately conducive to universal Good or Happiness’;32 ‘as a rational being I am bound to aim at good generally’.33 (b) Strict impartiality requires us to say that the good of one individual is no more important than that of any other; the happiness of a king is not in itself more important than that of a peasant, and is not to be more thought of unless for some special reason, such as that it might be more productive of further happiness. These two principles jointly entail a further and highly exacting rule of duty, namely (c) that ‘each one is morally bound to regard the good of any other individual as much as his own, except in so far as he judges it to be less, when impartially viewed, or less certainly knowable or attainable by him’.34
Here is a mass of moral principles which, according to Sidgwick, ‘present themselves as self-evident; as much (e.g.) as the mathematical axiom that “if equals be added to equals the wholes are equal”.’35 ‘I regard the apprehension, with more or less distinctness, of these abstract truths, as the permanent basis of the common conviction that the fundamental precepts of morality are essentially reasonable.’36 Whether these principles will have commended themselves to readers as having the compelling force they had for Sidgwick, I do not know. But it can hardly be denied that the emergence of such a system was a significant fact. The most distinguished of British philosophers, Hume, had argued with great persuasiveness that in morals ‘reason is, and ought only to be, the slave of the passions’. Here was a philosopher, no less acute than Hume, and in ethics far more thorough, who held that not only the nature of the right and the good, but also our duties in particular cases, were rationally apprehended. The pendulum had swung back again from the emotionalist toward the rationalist extreme. Not, of course, that Sidgwick stood where Clarke had stood. Most of the rules of practice that Clarke had accepted as dictates of reason, such as those of truth-telling and promise-keeping, Sidgwick held to be utilitarian precepts, which, since they obviously allowed exceptions, could not be set down as necessary. And the maxims he did accept, as we have seen, were accepted very cautiously, and formulated with precision. On the place of reason in morals, his admirably lucid and balanced discussion seemed to many to have said the last word.
27. But it turned out, as it always does in these difficult, basic questions, that there was very much more to say. One day in the nineties there appeared in Sidgwick's study in Cambridge a young man from Finland whose interest was in anthropology. Like most other students of this subject, he had been much impressed by the diversity of moral codes that modern research had brought to light, and he wanted to know how Sidgwick reconciled the fact of such diversity with his view that moral principles were insights of reason. In cases where insights were undoubtedly rational, such as the multiplication table or the laws of logic, this diversity did not exist; Eskimos do not hold that seven and five make eleven nor the pundits of Tierra del Fuego that one can argue validly from affirming the consequent. Now if the principles of ethics were similarly insights of reason, one would expect that they too would be accepted unanimously. But it is notorious that they are not. How, then, could one combine a rationalist ethics with the anthropological facts? Sidgwick listened to the young man attentively and replied that in ethics as in other fields, education and intelligence varied. Even in mathematics, there were people who could hardly count beyond their fingers, and would be wholly helpless before an advanced demonstration. But this did not show that the demonstrations of the mathematician were not self-evident to him. It is no part of the rationalist's case that what is necessary and self-evident should appear so to everyone; indeed if this were true, the larger part of mathematics would have to be rejected, since to most people it is anything but plain. And if such variation is inevitable in mathematics, it is still more so in morals, where feeling is so much more likely to distort one's thinking. There is thus no sort of inconsistency between ethical rationalism and the anthropological facts, provided that you include among those facts the varying levels of men's intelligence.
The young man appears to have gone away shaking his head. He was unconvinced, even by Sidgwick. He dug still deeper into anthropology, read omnivorously, produced two massive volumes on The Origin and Development of the Moral Ideas, and became in time the first professor of sociology in the University of London. Running through all his work is his old conviction that rationalism in ethics is a superstition, that if it is approached from the side of anthropology, this theory will fade away like other superstitions. It became one of his main purposes to discredit the theory, and after he retired he published a book on Ethical Relativity in which the thought of a lifetime about it was summed up and used against it in concentrated fire. The name of this intransigent student was Edward Westermarck. With him we come to a philosophy that may be fairly called contemporary, and we must observe at closer range the swings of the dialectical pendulum. Westermarck's subjectivism was urged with an unprecedented mass of the anthropological learning which the penetration of primitive cultures had made possible. It calls for a chapter by itself.