1. In the dialectic of reason and feeling which we have been tracing, Westermarck was supposed for a time to show the furthest possible swing of the pendulum toward the reduction of moral judgment to feeling. In fact there was a still further swing to come, as we shall see. But he had barely launched his theory before the pendulum began to swing back again toward ethical rationalism. The Origin and Development of the Moral Ideas appeared in 1906, and a second edition was called for in 1912. In the latter year an article appeared in Mind by H. A. Prichard with the arresting title, ‘Is Moral Philosophy Based on a Mistake?’ It attracted no great notice at the time, but it planted seeds of reflection that grew steadily and produced rich fruit many years later in the work of Sir David Ross and E. F. Carritt.1
Prichard's answer to his own question was that moral philosophy was based on a mistake. It had assumed ever since Socrates that the judgment of right or duty called for a defence, and a defence of a particular kind. We must show that the act alleged to be right conduced to someone's advantage. The advantage might be to the agent himself or to others; and it might be in increased happiness, in enlarged knowledge, or in many another good. But if our approval was to be justified, some advantage there must be. Here, Prichard insisted, lay the mistake. No such justification was necessary or possible. The reason why we should keep a promise or avoid a lie is not that someone is going to reap a profit from it. There is really no reason at all outside the fact of duty itself; the reason for acting in this way is that it is self-evidently right and obligatory. We should do our duty because we see that it is our duty, just as we should believe what is true because we see that it is true. No more ultimate justification can be given in either case. What is right need not be right because it is contributory to the good, for it may still be right when it produces no discernible good at all, even when it produces, for all we can see, a surplus of evil. It is because of this emphasis on right as not derivative from good that these philosophers have received their name. They are called ‘deontologists’ because their emphasis is on το δεον that which is right or binding, as distinct from that which is good.
2. We have described them as returning to rationalism in ethics. The ground for this will be evident from this characteristically clear passage from Sir David Ross:
‘That an act, qua fulfilling a promise, or qua effecting a just distribution of good… is prima facie right, is self-evident; not in the sense that it is evident from the beginning of our lives, or as soon as we attend to the proposition for the first time, but in the sense that when we have reached sufficient mental maturity and have given sufficient attention to the proposition it is evident without any need of proof, or of evidence beyond itself. It is self-evident just as a mathematical axiom, or the validity of a form of inference, is self-evident. The moral order expressed in these propositions is just as much part of the fundamental nature of the universe (and, we may add, of any possible universe in which there were moral agents at all) as is the spatial or numerical structure expressed in the axioms of geometry or arithmetic. In our confidence that these propositions are true there is involved the same trust in our reason that is involved in our confidence in mathematics; and we should have no justification for trusting it in the latter sphere and distrusting it in the former.’2
This leaves us with no doubt that in Ross's view the judgment that an act is right is an expression of reason. But what is it exactly that reason is apprehending in such a judgment? Is it the rightness of some particular act? Is it the validity of some rule of action, like ‘keep your promises’ or ‘don't lie’? Or is that which is self-evident, as Kant thought, some very abstract rule like ‘be consistent’, which we can apply as a test to rules of less abstractness? Ross accepts none of these views.
3. What he takes as self-evidently binding is prima facie duties. And a prima facie duty is a claim on us resting on some circumstance of the situation we are facing. Sometimes this circumstance is a previous act of our own. Suppose we have promised to return a book on a certain day. Then in virtue of this promise, the return of the book has a self-evident claim on us. It is not a claim that can never be outweighed by any other claim, so that it is our duty to keep our promise though the heavens fall; that is fanaticism. But the man who does not feel that making a promise lays him under any degree of restraint must be morally blind. Similarly with other features of the situation. Sometimes what gives rise to my prima facie duty is a previous act of someone else. Suppose someone has saved my life at the risk of his own. Does this leave me at liberty to treat him as if he had never raised a finger in my behalf? Clearly not. Sometimes, again, what gives rise to a duty is the possibility of injustice. If I am distributing provisions at a school picnic, should I be justified in passing one boy over because he had a wart on his face? Should I be justified in doing so even if I could convince myself that by giving his share to some particularly greedy youngster, I could produce as much pleasure on the whole as if I had acted justly? Ross would say no. Justice has a claim on us merely as justice, and quite apart from any goods that it may lead to. How many sorts of circumstances give rise to these special obligations? Not many, apparently;3 but what there are are of very common occurrence, and taken together they fill a large space in the field of our duties.
4. Now to recognize a rule as self-evidently and universally valid has in the past meant that in practice it had no exceptions. Kant held that if the rule of truth-telling passed the test of rightness at all, it allowed of no exceptions whatever. Ross has abjured such rigorism. He holds both that the rule is self-evidently valid and admits exceptions in practice. How is this possible? His answer is that what is self-evident is the claim, not the duty in a concrete case. An act is a concrete event, and such an event is never simple; it always has a diversity of aspects. Considered under one aspect, it has a certain claim upon us; considered under another, it has a different and perhaps a stronger claim. My duty is to weigh these claims against each other, and to choose that line of action whose total claims are greater. Suppose I am considering whether to repay a debt. On one side, my payment of the money will be the fulfilment of a promise, and that has an obvious claim upon me. On another side, my family may be in extreme need, and then my repayment of the money has the aspect of a failure to alleviate their need. Ross would recognize a self-evident obligation to seek the good of one's family as well as to fulfil one's promises. Both features lay one under a self-evident obligation, though the obligations cannot both be met.
There is conflict here, but no contradiction. And the conflict can normally be solved, for the different features of the situation will impose obligations of differing weight. Moral laws are related to concrete cases very much as physical laws are. We hold that the law of gravitation is universal; when we see a balloon rise, we do not say that the law is being contradicted; we say that gravitation is still acting on the lighter-than-air gas in the balloon, but that the upward pressure of the heavier air around it is stronger than this downward pull, and hence the balloon goes up. What is really universal is not the falling of bodies, for sometimes they do not fall, but the tendency to fall; this will still be present, though temporarily overcome. It is the same with moral laws. The fulfilment of promises, taken in the abstract, is always, and as such, right; and so far as any particular act is the fulfilment of a promise, it also tends to be right. Similarly, the securing of good to one's family is, as such, right and obligatory. But in a particular case the first of these claims may be much stronger than the second and hence outweigh it as a ground of actual duty. The relative force of the obligations may be extremely hard to estimate; the deontologists hold that we can never be sure that we have made the right choice, since we can never be sure we have exhausted the relevant aspects of our action or done justice to their several claims. Prichard used to say that our chances of ever doing the precisely right action were very low, and that if we ever succeeded, it was by luck. But our estimate of comparative claims, however fallible, is still a function of reason, and if our powers were equal to the task, we could see the solution of concrete problems as clearly and certainly as a competent mathematician now grasps the solution of a complex equation.
So far we have been dealing with what is asserted in a judgment of right. According to Ross, what is asserted is either that the agent has an obligation grounded on some aspect of his situation, or that the situation as a whole is such as to render a certain action (or perhaps one of several alternative actions) obligatory. The first judgment expresses a necessary insight, as necessary as anything in mathematics. The second contains subordinate insights of this kind, but it is not as a whole necessary, because the concrete is inexhaustible.
5. What of the other principal judgment of morals, the judgment of good? Ross is willing to go a certain distance with the subjectivist. He goes so far as to say: ‘the only common fact that is present whenever we use the term “good” is that in each case the judger has some feeling of approval or interest towards what he calls good’.4 But though this is an interesting fact of psychology, it does not imply that being approved is all we mean by good;
‘it is surely a strange reversal of the natural order of thought to say that our admiring an action either is, or is what necessitates, its being good. We think of its goodness as what we admire in it, and as something it would have even if no one admired it, something that it has in itself. We could suppose, for instance, an action of self-denial which neither the doer nor any one else had ever admired.’5
Again, though goodness belongs only to states of mind, it is not dependent on being known; ‘goodness is entirely objective and intrinsic to the things that are good’.6 Ross holds, as Sidgwick and Moore did, that goodness is a non-sensible, indefinable quality which must be apprehended, if at all, by intelligence.
How is it related to the things that have it? We are of course talking now not about things that are good as means, but things that are good in themselves, like happiness, knowledge, or dutifulness. When we say that happiness, for example, is intrinsically good, the relation we have in mind is certainly not causal, for a cause precedes its effect, while happiness and the goodness that belongs to it are not separated by time; nor are we reporting a mere empirical togetherness, as we do when we say grass is green, for though grass could be grass without being green, happiness could not be what it is without also being good. The connection, therefore, is a necessary one. When we apprehend it, we see that anything having the nature X must, in virtue of that nature, also have Y.7 Such necessary insight may appear again in our comparison of values. We say, for example, that our delight in last night's talk with an old friend was worth more than the boredom of the night before over a recent ‘thriller’; but we should not offer this as an inductive or deductive conclusion from anything; we should expect it to be self-evident to anyone whose experience was really like ours.
So far, in his dealing with good, Ross is at one with Sidgwick and Moore. But at the next step he parts company with Sidgwick, and, a step later, with Moore also.
6. Sidgwick held that it was self-evident to the reflective mind not only that pleasure is good, but that it is the only good. We can see, that is, when other alleged goods are offered us, such as knowledge, friendship, reputation, power, that they would be worthless unless they brought pleasure with them, and that such value as they have belongs to this attendant pleasure. Ross agrees that pleasure is good and self-evidently so, but he thinks it plain on examination that the goodness we attach to experiences does not vary exclusively with the amount of pleasure in them, and indeed that some things other than pleasure are just as clearly valued for their own sakes. He recognizes three of these; virtue, knowledge, and the just allocation of pleasures and pains. That these really are intrinsically good can be shown, he thinks, by ideal experiments. Consider virtue. This means action, or the disposition to act, from such motives as the desire to do one's duty or to make others happy. One can imagine a universe in which such virtue did not exist, but in which the total pleasure or happiness was the same as in another where it did exist. Would the first lose nothing by reason of its lack of virtue? Ross thinks we can see this to be false. Imagine, again, a world in which virtue, happiness, and everything else regarded as good, existed in the same amounts as in some other, the only exception being that all the people in the first were ingorant while all the people in the second were enlightened; would not the second be better than the first by reason of the enlightenment? Ross thinks it self-evidently would be. Finally, take two worlds in which, with equal amounts of happiness and unhappiness, these amounts were so distributed that in the first the virtuous people were miserable and the unvirtuous happy, while in the second the allotments were reversed. This is a quite conceivable situation, however improbable in fact; and Ross thinks that the second pattern of allotment, in which virtue was attended with happiness and vice with unhappiness would be intrinsically and self-evidently better than the other. Ross thus goes beyond Sidgwick in recognizing that goodness is a necessary property of three other forms of experience besides pleasure. But there is still nothing novel about this position. With some differences regarding the number of goods recognized, it is the well known position of Rashdall and Moore.
7. From this point on, however, the deontologists strike out a new path. Since it was in reaction against the views of Moore and Rashdall that the most novel and significant part of their theory was developed, it may be well to pause for a moment over the theory against which they were in revolt.
The appearance at Cambridge of Moore's Principia Ethica (1903), followed a few years later at Oxford by Rashdall's Theory of Good and Evil (1907), had raised ‘ideal utilitarianism’ into something like an ethical orthodoxy. This theory was attractively simple. Of the two classical questions of ethics, What is good? and What is right?, it said that the first was by far the more important, since you could not answer the second without answering the first, while if you did find an answer to the first, you could answer the second by deduction from it. If you knew what sorts of things were intrinsically good, you could say that conduct was rendered right by the production of these things. Those who adopted this way of thinking were called utilitarians because they believed, as Bentham, Mill, and Sidgwick did, that what made an act right was its consequences; they were called, a little clumsily, ideal utilitarians to distinguish them from this earlier school, which had accepted pleasure as the only good; their summum bonum was what Rashdall called ‘an ideal end or good, which includes, but is not limited to, pleasure’.8 They recognized a variety of intrinsic goods, but held that all of these could be given places on a single scale of value. This made possible a single, simple rule of all conduct, which was taken as self-evident and holding without exception: ‘So act as to produce the greatest net good.’9
Now it is precisely this cardinal principle of utilitarians of all schools that the deontologists reject. They think there are cases where, on a mere calculation of goods, one should clearly do act A, while, by the general agreement of conscientious persons, one should do act B instead. You have borrowed money from a well-to-do man, for example, and promised to return it on a certain day. On the way to return it, you meet a friend who is unemployed, ill, or distressed, and who would, in all probability, use the money to much better advantage than the man to whom you owe it. Is it obvious that you should turn the money over to the needy person and break the promise to your creditor? Certainly most men would not accept that. Or suppose that two neighbours of yours are in need, and you are considering to which of them, with your limited means, you should offer help. So far as you can see, their needs are equal, and they would put such help as is offered them to equally good use. In your relations with them there is only one significant difference. You recall that, on a skating party last winter, when you foolishly ventured on to thin ice and went through, it was one of these men, Jones, who, at some risk to himself, pulled you to safety. Would the fact that the consequences of offering aid would be equally good in the two cases really leave it indifferent which man you offered it to? A utilitarian of either type, it is said, would have to answer Yes, whereas anyone who really acted on that view would be thought callous to the claims of gratitude. Or suppose, once more, that the government of a certain community finds itself faced with an outbreak of lawlessness which it believes can be effectively checked only by stern measures. Unfortunately it cannot lay its hands on any of the ringleaders. It has in jail, however, a man of criminal record, who would certainly be no loss to the community, and whose connection with the recent outbreaks could be easily made out by a little skilful and entirely secret manipulation of evidence. The good to the community of getting rid of him and of simultaneously deterring future outbreaks would seem greater than that of keeping him alive to be a lasting nuisance. So he is formally condemned and hanged by way of public example. Such action might conceivably be called for by utilitarian calculations. None the less, say the deontologists, it would be obviously and monstrously wrong.
8. How would the ideal utilitarians reply? They would certainly begin by saying that on the practical issue their conclusion in all these cases would be the same precisely as that of the deontologist and the plain man; they differ only as to the ground on which this common conclusion is based. In Ross's view, it is based on a prima facie duty to keep promises, to pay debts of gratitude, and to do justice, independently of any good that such action may bring; and in these cases the line of duty on which everyone would agree is supposed to be indefensible on any other basis, the appeal to consequences requiring the opposite verdict. To this the utilitarian answers that it is only by irresponsible book-keeping that consequences are made in such cases to cast their vote against what common sense would say. Make your calculation more complete, and you will find that it supports rather than opposes common sense. Take the case, for example, where you have to choose between paying a debt as you have promised and giving the money to someone in need. You are not playing the game of consequences fairly if you consider only that the creditor does not need the money and the other man does, and that you will therefore do more good by breaking your promise. A breach of promise does not stop with itself; it has its own train of manifold and far-reaching consequences. They include not only the deprivation and disappointment of the creditor, but also the loss on your part of his confidence, the weakening of your habit of keeping engagements—a habit that must be firmly maintained if the ever-present temptation to relax it in one's own favour is to be avoided—and, above all, the small but appreciable lowering of public confidence that follows any deliberate breaking of such an engagement. This last is no light matter. The business of life depends at a hundred points on people's readiness to trust the promises of others, and every breach of that trust has its repercussions. Subtract from the advantages of promise-breaking all these serious ills, and you will find them clearly outweighed by the goods of loyalty to one's word. Rightly viewed, ideal utilitarianism thus has the support of common sense rather than its veto.
9. It is now the deontologist's turn. What would he answer? He would admit that promise-breaking does have consequences of this kind, and that before a decision is arrived at in any particular case they must be taken into account. He is not denying that consequences are important. He is a rationalist, but he is not a rationalist like Kant, who tried disastrously to make right-doing independent of consequences; he recognizes the producing of good consequences as one of his self-evident prima facie duties. What he is denying is that consequences alone are important. He is insisting that some kinds of action have an obligation not derived from them, and that, in a particular case, this may outweigh the obligation that is derived from them. Now there is no hope of settling this issue if, whenever such a case is offered, the other side is allowed to contrive additional consequences ad hoc and throw them into the scale. One must take a case where the evidence is agreed upon and where it stays fixed. Can any such cases be found?
The deontologist would say that, ideally at least, they can; he would say, I think, that they can be provided in either of two ways. First, one may imagine a situation in which all the goods and bads involved in keeping a promise and in breaking it are exhausted as far as the eye can see, and the net good on the two sides is admitted to be the same. (This is clearly a possibility, and if anyone says it is not, on the ground that promise-keeping will always have the best results, he is claiming quite unreasonable powers of prophecy.) Very well, in such a case, where there is admittedly nothing to choose in point of consequences, would the ordinary conscientious man say that it was just as permissible to break a promise, or suppress his gratitude, or punish an innocent man, as to do the opposite? The deontologists are clear that he would not. Hence they say that, in point of fact, he thinks of such actions as promise-keeping as having a rightness of their own, which they owe not to their consequences but rather to being what they are. Here I think the deontologists have the better of the argument.
They would proceed to ask, secondly, how it is that, if the rightness of our conduct does depend wholly on consequences, we should so often be sure, before knowing the consequences, that certain actions are wrong. Would a judge, or any other reflective man, to whom the suggestion was made that an innocent man be hanged, really have to stop and work out the consequences pro and con before having an opinion about it? Surely his opinion would come out instantly, and with a confidence that would be inexplicable if it depended on calculating results. Such considerations do show, I think, that many of our moral judgments are made with no thought of consequences. Still, the argument is less decisive than it seems. For, as Sidgwick so conclusively showed, it is possible for a rule that was utilitarian in origin and is justifiable only in the utilitarian way to come, in course of time, to seem evident intuitively. Nevertheless, the argument does have some weight. Most of us would not only say before thinking of consequences that the hanging of an innocent man was wrong; we should stick to our opinion, even when faced by evidence we could not refute that society would profit by the deed. Such tenacity strongly suggests that in justice of this kind we find an obligation that cannot be wholly accounted for by any consideration of prospective gain, personal or social.
10. Granting to the deontologists that we do feel this obligation, the question now is whether they are right in their way of explaining it. Their way of explaining it, as we saw, amounts to saying that it is ultimate and inexplicable. ‘The sense of obligation to do, or of the rightness of, an action of a particular kind’, says Prichard, ‘is absolutely underivative or immediate’; ‘our sense of the rightness of an act is not a conclusion from our appreciation of the goodness either of it or of anything else’.10 Ross writes: ‘It seems, on reflection, self-evident that a promise, simply as such, is something that prima facie ought to be kept…’11 ‘If any one ask us,’ says Carritt, ‘“Why ought I to do these acts you call my duty?”, the only answer is, “Because they are your duty”, and if he does not see this we cannot make him, unless by informing him about matters of fact; if he sees they are duties, he can no more ask why he ought to do them than why he should believe what is true.’12
Now we have granted that the duty of promise-keeping, for example, does not rest merely on consequences; it is better that promises should be kept, even if no later advantage accrues from it. And it may be thought that this is what the deontologists too are saying. That would be a mistake. What they are saying is that promise-keeping is our duty though in fact there is no good in it at all. According to Ross, ‘If I contemplate one of the acts in question, an act, say, in which a promise is kept… and ask myself whether it is good, apart both from results and from motives, I can find no goodness in it. The fact is that when some one keeps a promise we can see no intrinsic worth in that.…’ And again: ‘we can see no intrinsic goodness attaching to the life of a community merely because promises are kept in it.’13 It is our duty to keep promises, not because, even with other things equal, the life of a community is better for promises being kept in it, but because… the sentence cannot be completed. It is our duty, but there is no reason why. Our obligation is read off directly, and with self-evident necessity, from a set of neutral facts. So also of such duties as repaying debts, and telling the truth. Indeed most of the prima facie duties recognized by the deontologists rest not on the goodness of any state of things, but on the neutral and factual character of the act itself. For this reason, the ancient search of the philosophers for some single characteristic of right acts which serves to make them right is set down as misguided. There is no one thing that makes right acts right. Sometimes they are right because they are the keeping of promises, sometimes because they are the paying of debts. But between the rightness of an act and its tendency to bring into being any kind or degree of good there is no general relation at all.
11. Here I must dissent. This conclusion does not seem credible. We are being told that it may be a self-evident duty to choose one rather than another state of affairs even though, in respect to goodness, there is nothing to choose between them. But more; we are being told that state of things A may be definitely and admittedly worse than B, and that it may still be our duty to bring A into being. With a choice before us of making the world worse or making it better, we may have a moral obligation to make it worse. This is very hard to accept. A strong case has been carried too far. When the deontologists said that duty is not based always on a goodness that follows the act in time, but sometimes on the character of the act itself, they carried us with them. They did so because it seemed clear that a state of things in which promises were kept, gratitude recognized, and truth told, was a better state of things than one in which these were not done. But when we are now told that such obligations have nothing to do either with the intrinsic goodness of the acts, or of the state of things they institute, let alone the goodness of their consequences, we feel as if the mat on which we had been approaching this school had been pulled out from under our feet. The obligations that were presented to us as rational insights take on an air of caprice.
Can we offer any evidence that we are correct about this and the deontologists not? I think we can, though argument on such ultimate issues is notoriously hard.
12. (1) Perhaps an ad hominem argument may therefore be permitted. Sir David Ross ‘can see no intrinsic goodness attaching to the life of a community merely because promises are kept in it’; goodness attaches only to the motives or consequences of such conduct, not to the state of affairs constituted by its general practice. But when we come to justice, in one of its important forms, Ross is emphatic that this does have intrinsic goodness, regardless of motives or consequences. A community in which virtuous men are happy and wicked men unhappy he considers much better than one in which these allotments are reversed, even though the totals of happiness and unhappiness are the same.14 Now the goodness admitted to be present here is not the property of anyone's experience, but of a set of arrangements between experiences. If intrinsic worth can be owned by such a set of arrangements, why should it not also be owned by that other set of arrangements in which promises are kept, or truth told, or debts repaid? If there is any fundamental difference between these situations, which would justify saying that one is self-evidently good and the other self-evidently valueless, I have failed to catch it.
13. (2) Is it really the case that to ask why something is our duty is meaningless? Mr Carritt says that if we see it is our duty to keep a promise, it is as pointless to ask why we should keep it as to ask why we should believe what is true. In this he is no doubt right. Once we have seen that the keeping of a promise is our duty, it would be idle to ask why we should do it, since we have in our possession already the most conclusive answer that could be given. But then this is not the situation in which the question would be asked. It would be asked, rather, by someone to whom his duty was not apparent and who wanted guidance about it. He might see that keeping a promise would bring pain to someone, and, reluctant to give this pain, he might ask, ‘What ground is there, after all, for saying that to keep a promise is obligatory? You tell me that there is no ground, that if I attend and am clear-headed I shall see the obligation immediately. But for the alternative course of breaking the promise there clearly is some ground, namely that it will avoid pain. Am I to believe that it is my duty to do something that involves that pain, yet carries no good to counterbalance it? That does not seem to me, on the face of it, reasonable. Why should I take it as obligatory to produce that which you own to be without value? I cannot think this question meaningless, nor the asking of it a sign of obtuseness or failure to attend. I suspect that most people who do attend would agree that it is significant, and also that it has a natural and intelligible answer; namely, ‘It is your duty to keep promises because the state of things in which they are kept is a better state than one in which they are broken’. A critic may rejoin that this is no more obvious to the obtuse than the proposition for which it is offered as a reason. That may or may not be true. But the point is that it is significant and does give a reason. It is one thing—and a not very convincing thing—to say that the keeping of promises is a duty because its obligatoriness is self-evident and another thing to say that it is a duty because it is good. If the first account is true, the reason given in the second must be an impertinence—unneeded, untrue, adding nothing to the case. And I do not think it is.
Since the inability to give a reason for duty is ascribed to the ‘immediacy’ of our insight, it may be well to look a little more closely at this immediacy. Take a necessary proposition of an ordinary kind, say ‘whatever has shape has size’. This is an immediate insight because the first attribute entails the second directly; the suggestion that the second follows only through the intermediary of another and omitted term would seem absurd. But the parallel suggestion about promise-keeping and obligatoriness would not seem at all absurd; indeed the absurdity, if there is one, seems rather to lie on the other side, in the suggestion that promise-keeping, described as wholly valueless, should as such be obligatory. And whereas in the shape-size proposition the suggestion of a middle term is a gratuitous complication of what is plain already, the introduction of a middle term in the other clears up a puzzle. ‘Act A, worthless in itself, should be done’; that seems very dubious indeed. ‘Act A, because it is intrinsically good, should be done’; that is a different matter and makes sense at once. We are not questioning that the obligatoriness of a kind of action may be seen as rational and necessary. We are only contending that the line of necessity does not run straight from a state of things that is utterly grey, so far as worth is concerned, to a duty to produce that state; what carries the obligation is the fact that the state would be worth while, or of some positive value—in short that it would be good.
14. (3) A further point may be mentioned which has troubled many moralists about this form of intuitionism: the duties with which it presents us are ‘an unconnected heap’. Philosophers and scientists alike have generally felt that they should use Occam's razor whenever they can; it is their business to reduce apparent disorder to law, and diversity of fact to unity of principle. Moral philosophy has proceeded on the assumption that, if we searched resolutely enough, we should discover behind the great variety of acts that we call right and obligatory some unifying principle that made them so. Ethics has consisted very largely of the search for that principle. To be sure, moralists have disagreed in disappointing fashion as to what the principle was; we have been variously advised to order our conduct so as to secure survival, wisdom, self-realization, power, pleasure, the beatific vision, and much else. But the difficulty and disagreement in answering the question have not destroyed the conviction that some principle is there to be found, even if only a very abstract one like the rule of producing the greatest good. Now the new intuitionism says that there is no such unifying principle. There is no common reason for calling actions right or obligatory. Sometimes they are right because they produce the greatest good; sometimes because, though they produce a smaller good, they have the character of promise-keeping; sometimes because they have the character of debt-paying; and so on. And the deontologists are surely right in saying that we cannot be sure that when we set out in search of unity we shall find it, or even that it is there to be found. The fact that a theory is simpler than another is no proof in itself that it is nearer the truth. The world has not been ordered for the ease of our understanding.
15. Are we to say, then, that there is no presumption in favour of a single ground of duty as against a heap of duties? That, I suggest, would be going too far. When we use the same word ‘right’ of a variety of actions, I do not think it is used either to refer to utterly different things, as the word ‘mother’ is to refer to a female parent and to a mould that appears in vinegar, or to some quality that is perfectly obvious to us but abstract, simple, and indefinable. ‘For it is a character of that peculiar sort, like truth, beauty, and goodness, that we can seek it, as Plato says, “divining it to be something”, before we know what it is…’15 The analogy to truth is worth considering. The truth of a proposition is not, as certain realists once suggested, a simple unanalysable quality that attaches to it ‘like the flavour of gorgonzola’, nor can we believe that the term refers to utterly different things—sometimes it means correspondence with fact, sometimes coherence with other beliefs, sometimes success in practice, sometimes prescription by authority. However hard it is to bring to light the nature of truth, it has been the conviction of most inquirers in this field that such a nature there must be. Unless there were, the conflicts of belief that trouble us most deeply would be insoluble and indeed hardly possible. When the biologist declared it true that the human mind has evolved, he would mean, for example, that it corresponded to fact, and when the theologian denied it, he would mean that it was not sanctioned by authority. The second would not be denying what the other asserted, and there would be no dispute. But we know perfectly well that the one does intend to assert the truth of the proposition in the sense in which the other intends to deny it; they mean by truth the same thing, even though one or other must have gone wrong in formulating what he means.
Just as we must believe that there is some common character that makes true beliefs true, so we are constrained to believe that there is some common character that makes right acts right. To be sure, this may be an illusion. We may be using the same word to indicate the presence of quite different characters. But one who thus rejects the evidence of usage must at least accept the burden of proof. And most moralists would probably confess to an ‘invincible surmise’ that there is more to the story than an ‘unconnected heap’ of duties, that we can hardly have reached the end of the line when we say that what makes generosity right is its producing something of worth, while that which makes truthfulness right has nothing to do with worth, but lies in a suitability in itself worthless. That position is to be accepted only when the search for a common ground has broken down. It has been replied that this disunity is not escaped by making the right dependent on the good, since among the things to which we apply the term ‘good’ there is the same lack of common nature that appears among right acts. Without agreeing that this is true, we may point out that even if it is, those who hold to the primacy of good have the more intellectually satisfying case. For though they may admit that there is no common nature in what is good, the deontologists double this pluralism by saying that there is no common nature in what is right either.16
16. Is there any way in which the deontologists could admit a common ground for all right acts and yet keep this rightness independent of the good? One interesting suggestion, recently put forward, is somewhat reminiscent of Clarke. May it not be that all right acts are right because of their fittingness, that in one way or another they all fit the situation? We recognize that if a service has been done us, an act of gratitude is fitting, while an ignoring of the service or a returning of evil for good would be plainly unfitting. It is fitting that we should pay our debts and unfitting that we should renounce them, fitting that we should concern ourselves more for the welfare of our own family than of other people's families, fitting that we should tell truth rather than falsehood, fitting that we should acquit innocence of guilt and unfitting in the extreme that we should punish it. Even the rightness that depends on good consequences can be brought in under the formula, for in view of human needs and desires it is always fitting, so far as it goes, to add to the stock of human good. If we take this line, we can say not only that rightness has a common base, wherever it appears, but also that this base is objective and rationally apprehended, for fittingness is not a quality, still less a feeling, but a relation evident only to intelligence. Right will then remain irreducible to good, for the fittingness which makes an act right will be quite distinct from goodness.
This view does, I think, help to explain why we call some acts right and some wrong. The unsuitability or inappropriateness of treating harshly a person who has befriended us is surely a ground, whether the whole ground or not, of calling such conduct wrong, and the theory applies with special force to such cases as the condemning of an innocent man; the wrongness of such action, particularly on the part of a judge, seems to consist chiefly in its being so shockingly unfitting and incongruous. Every sound theory must, I think, incorporate this insight into its account of our calling such actions wrong. My one question about it is whether it has really escaped the basing of the right upon the good, whether the fitness upon which rightness is now based is not itself approved because it is good. Fitness or suitability is, of course, a very broad notion, and it is obviously not all kinds of fitness, but only certain special kinds, that are supposed to make actions right. If a man has committed a long row of murders, and finds a chance to make a great gain out of another with small chance of detection, there is a sense in which it fits his character and the circumstances perfectly to do it again. I knew an old lady once who, when asked her view what should be done with Hitler if he were caught, said that he should be put in a cage and taken about for persons like herself, whom he had made miserable, to stick pins into; and I must confess to having felt a certain fitness in the suggestion, however absurd. Kant, as is well known, thought that if the society on some remote island were being finally dissolved and certain persons guilty of murder were still living in jail, the appropriate thing to do with them was to execute them before the community broke up. Military men think that the appropriate way to treat enemy invaders is to kill as many of them as possible; Quakers have commonly thought that for a man to kill a brother man is unfitting absolutely. We are clearly in need of a test which will distinguish the suitability that makes an act right from other kinds of it.
Sir David Ross thinks there is such a difference, but that it is indefinable; ‘We can begin to define moral rightness, because we can say it is a form of suitability; but we cannot complete the definition, since if we ask what kind of suitability it is we can only say that it is the kind of suitability that is rightness.’17 Now I strongly suspect that if we were to examine the suitabilities that were pronounced right, we should find that all of them constituted, or belonged to, a state of affairs that we recognized as good, and that if we inspected the suitabilities pronounced wrong, we should find that they were similarly bound up with states of affairs thought to be bad. How, for example, could you ever decide Kant's problem whether it was more suitable to kill the men in jail before society broke up than to give them another chance, except by considering that the two acts would belong to two different states of affairs of which one was better than the other? Again, taken as mere abstract suitabilities, what possible choice could there be between the pacifist and the military view of how to treat aggressors? Both are eminently suitable in their different ways. If we are to arrive at an intelligent choice between them, we must compare not two abstract suitabilities merely, but the two vastly extended states of affairs which these acts would tend to engender, and ultimately, perhaps, the two opposed ways of life to which they belong. It may be said that this makes the problem enormously difficult. But then it is difficult; and when we see that to solve it satisfactorily we must judge the comparative worth of competing ways of life, we see why it is so difficult. Comparative fitness should be considered, by all means. But we shall soon find ourselves effectively blocked unless the schemes of things into which differing acts would fit may themselves be registered as better and worse.
17. Why should we trouble our heads over these impalpable distinctions? It is because they lie on the watershed between two conceptions of ethics, that which finds in right the primary notion, and that which finds it in good. The majority of great moralists have belonged to the latter school. Plato, Aristotle, the Epicureans, Hobbes, Hume, the utilitarians, the idealists, the evolutionists, though they had very various notions as to where the good lay, all took as the test of right conduct its conduciveness to good. But there have been moralists of weight on the other side. Christian ethics, indeed, has not taken a very clear stand, since its insistence on conformity to right, as given in divine commands, has been balanced, even in such exponents as Locke and Butler, by the teaching that this conformity is the necessary means to happiness hereafter. But in the Stoics, in Cudworth, More, and Clarke, in the immensely influential Kant, in Reid, Price, and Whewell, and now again in the recent and remarkable school of British intuitionists, we find the insistence on a right that is directly perceived by reason and owes nothing to its productiveness of good. What we have been considering is, in effect, the issue between these schools. If a single case can be adduced of an act that is seen at once to be obligatory and not to produce the greatest good, the deontologists—the advocates of an independent rightness—have won their case.18
18. My own position in the matter is this: if the good on which rightness is said to depend is confined to consequences caused by the act and following it in time, then the dependence of right on good is not made out. The reason is that in some cases what seems to promise such consequences, for example the punishing of an innocent man for public good, is plainly wrong. On the other hand, if it is contended that the rightness of an act need have no connection with good whatever, the contention seems to me false. And it will be the more clearly false if, besides the goodness of later consequences, we take into our account the goodness of the act itself, or, I would rather say, of the pattern of life generally, with its mass of personal relations, which this act implies. If this goodness—too often forgotten in utilitarian reckonings—is included in our review, I think that rightness is always dependent on good produced, and that a duty that would leave the world no better for the doing of it is an irrational duty, indeed no duty at all. The attempt by the new intuitionists to breach this position is not, in my opinion, successful.
Have we anything better to offer? We have shown that there are some elements of paradox in saying that the right is independent of the good, but we have produced no constructive argument to show that it is dependent on it. In theory, two such lines of argument are possible, one inductive, the other a priori. The first line was developed with typical caution and clarity by Sidgwick. He undertook to show of each of the main types of virtue that, with minor exceptions and borderline cases, it conduced so uniformly to good that the connection could not be accidental; we must have come to account actions right because of this conduciveness to good.19 His argument here is masterly and we shall not try to retrace it. The trouble with it is not so much in its account of the right as in its account of what is intrinsically good. Sidgwick found this in pleasure alone. But if the only thing that can be thrown into the scale as good is pleasure, the case of the deontologists against the utilitarians seems to me unanswerable. To say that in every case where we approve the keeping of a disagreeable promise or the telling of an unpleasant truth we approve it because it promises the greatest pleasure is to say something that is not only unprovable, but plainly contrary to fact. The argument becomes much more plausible, however, if the good admits other elements than pleasure, and more plausible still if it is allowed to include the good of the frame of life to which the act belongs. In that case the right and the good may not, after all, fall apart. When we come to discuss constructively the nature of goodness, an account will be offered that we hope will heal this breach.
The other way of arguing for the dependence of the right on the good is the way of direct insight. ‘No one would say’, wrote McTaggart, ‘that a man ought to will the existence of anything unless the thing willed was judged to be such that its existence would be good.’20 ‘It seems to me self-evident’, said G. E. Moore, ‘that knowingly to do an action which would make the world, on the whole, really and truly worse than if we had acted differently, must always be wrong.’21 I do not reject this appeal to intuition, and those who do reject it will usually be found to be substituting intuitions of their own; such an appeal is sooner or later inevitable. But it ought to come later rather than sooner. One of the requirements for self-evident truth laid down by Sidgwick was that the terms of the proposition should be unambiguously clear; and to say that the rightness of an action is self-evidently dependent on its production of good before one knows what one means by ‘good’ is courting danger. Claims of self-evidence should follow analysis, not precede it.
19. The deontologists have shown a fidelity to actual moral judgment that is probably closer than that of any other contemporary school. They have argued with great force that moral judgments are really judgments, not expressions of feelings only, and here—for whatever it is worth—common sense is undoubtedly on their side. If they were asked more particularly about the part played by reason in such judgment, they would answer, I think, that its work was threefold. (1) It may apprehend that certain experiences, or arrangements of experiences, are intrinsically and necessarily good. (2) It may infer the consequences of a suggested action. (3) It may perceive that certain acts are necessarily right in virtue of, and so far as, they have a certain character, such as that of truth-telling or promise-keeping. As to (1), I suspect that they are substantially correct, though if the notion of goodness turns out to be less simple than theirs—and Moore's—the insight that X is good will itself be less simple. As to (2), the function of deducing consequences, they are again surely right; indeed this is hardly a matter of controversy among present-day moralists. As to (3), the alleged insight that a kind of act is right by reason of a certain abstract character, I think they are mistaken, but not wholly mistaken. The act is seen to be right in virtue of producing goodness, if not in consequences, then through being the kind of act it is. And reason is involved not only in seeing it to be of a certain kind, but also in seeing the filiation of this kind with a way of life as a whole. Truthfulness and promise-keeping do not stand alone. They are implicated in a wider order from which they draw part of whatever goodness they possess. Since we thus hold that the rightness of a kind of action is not independently grasped at all, but always derives from goodness, it is upon the notion of goodness that we shall fix our attention from now on. The next swing of the pendulum, returning to the view that rightness depends on the goodness of consequences, will compel us to look more closely at this fundamental notion of good.