You are here

VIII: The Knowledge of What Is Right

I TURN next to the epistemological questions connected with duty. Can we be said to know our duty? And if we can, how do we acquire this knowledge? I will start with a simple case. I am walking along the street, and I see a blind man at a loss to get across the street through the stream of traffic. I probably do not ask myself what I ought to do, but more or less instinctively take him by the arm and pilot him across. But if afterwards I stop to ask whether I have done what I ought, I shall almost certainly say ‘Yes’; and if for any reason I ask myself, before doing the act, whether I ought to do it, I shall give the same answer. Now it is clear that it is in virtue of my thinking the act to have some other character that I think I ought to do it. Rightness is always a resultant attribute, an attribute that an act has because it has another attribute. It is not an attribute that its subject is just directly perceived in experience to have, as I perceive a particular extended patch to be yellow, or a particular noise to be loud. No doubt there are causes which cause this patch to be yellow, or that noise to be loud; but I can perceive the one to be yellow, or the other to be loud, without knowing anything of the causes that account for this. I see the attributes in question to attach to the subjects merely as these subjects, not as subjects of such and such a character. On the other hand, it is only by knowing or thinking my act to have a particular character, out of the many that it in fact has, that I know or think it to be right. It is, among other things, the directing of a physical body in a certain direction, but I never dream that it is right in consequence of that. I think that it is right because it is the relieving of a human being from distress. Now it seems at first sight to follow from this that our perception of the particular duty follows from the perception of a general duty to relieve human beings in distress. And, generalizing, we might feel inclined to say that our perception of particular duties is always an act of inference, in which the major premiss is some general moral principle. And no doubt my grasp of the principle that I should relieve human beings in distress precedes my grasp of the fact that I should relieve this blind man, since up to this moment I may not have known of the existence of this man, and certainly did not know of his desire to cross this particular street; while I certainly had at least a latent awareness of the general principle, an awareness which the occurrence of any instance falling under the principle might call into activity—just as I have a latent knowledge of the laws of arithmetic or of English grammar before I proceed to make up my accounts or to write a letter.

Yet it will not do to make our perception of particular duties essentially inference from general principles. For it may, I suppose, be taken for granted that man was a practical being before he became a theoretical one, and that in particular he answered somehow the question how he ought to behave in particular circumstances, before he engaged in general speculation on the principles of duty. No doubt there was an earlier stage still, when men in fact did right acts without ever asking whether they were right, when, for instance, they helped one another in distress without thinking of any duty to do so. We see disinterested help being given by men to one another every day, without any thought of duty. Aristotle puts the point simply:

‘Parent seems by nature to feel friendship for offspring and offspring for parent, not only among men but among birds and among most animals; it is felt mutually by members of the same race, and especially by men.… We may see even in our travels how near and dear every man is to every other.’1

Butler puts the matter more eloquently:

‘There is such a natural principle of attraction in mart towards man, that having trod the same tract of land, having breathed in the same climate, barely having been born in the same artificial district or division, becomes the occasion of contracting acquaintances and familiarities many years after: for any thing may serve the purpose. Thus relations merely nominal are sought and invented, not by governors, but by the lowest of the people; which are found sufficient to hold mankind together in little fraternities and copartnerships: weak ties indeed, and what may afford fund enough for ridicule, if they are absurdly considered as the real principles of that union: but they are in truth merely the occasions, as any thing may be of any thing, upon which our nature carries us on according to its own previous bent and bias; which occasions therefore would be nothing at all, were there not this prior disposition and bias of nature.’2

Aristotle's reference is perhaps the more interesting, in two respects. In the first place, it takes the practice of disinterested aid further back in time, by asserting its existence not merely among men, but among animals. He opens up the vista of the development of disinterested action, as it exists in man, from the instinctive co-operation of the members of an animal community. And secondly, he points to what is much the most striking and universal example of disinterested action, the operation of parental love, from which perhaps all disinterested action may be supposed to have developed.

In such action, in its earliest form, there was no thought of duty. We must suppose that when a certain degree of mental maturity had been reached, and a certain amount of attention had been, for whatever reason, focused on acts which had hitherto been done without any thought of their rightness, they came to be recognized, first rather vaguely as suitable to the situation, and then, with more urgency, as called for by the situation. Thus first, as belonging to particular acts in virtue of a particular character they possessed, was rightness recognized. Their rightness was not deduced from any general principle; rather the general principle was later recognized by intuitive induction as being implied in the judgements already passed on particular acts.

The question may, however, be asked: ‘Once the general principles have been reached, are particular acts recognized as right by deduction from general principles, or by direct reflection on the acts as particular acts having a certain character?’ Do we, without seeing directly that the particular act is right, read off its rightness from the general principle, or do we directly see its rightness? Either would be a possible account of what happens. But when I reflect on my own attitude towards particular acts, I seem to find that it is not by deduction but by direct insight that I see them to be right, or wrong. I never seem to be in the position of not seeing directly the rightness of a particular act of kindness, for instance, and of having to read this off from a general principle—‘all acts of kindness are right, and therefore this must be, though I cannot see its rightness directly’.

It appears to me that we apprehend individual facts by deduction from general principles in two kinds of situation, and in no more. (1) We may have no real insight that the attribute A implies the presence of the attribute B. But we may have accepted on what we believe to be good grounds the belief that A always implies B, and we then may say to ourselves, ‘This is an instance of A, and therefore it must be an instance of B; I cannot see it for myself to be so, but I think it must be, because of a general principle which I have for good reason accepted’. Or (2) the general principle may be one that is not self-evident, but known as the consequence of a proof; and we may remember the principle while we have forgotten the proof. There again, we shall not see with self-evidence that the particular A is also a B, but we shall read this conclusion off from the remembered general principle ‘all A is B’.

Both these situations actually occur in morals. (1) In most people's lives there is a stage at which they accept some moral principle on authority before they have really come to recognize its truth for themselves; and in such a case the rightness or wrongness of the particular act is not apprehended on its own merits but read off from the general principle. The suggestion is indeed sometimes made that we never pass beyond this stage of acceptance of moral principles on authority to a fresh original recognition of them. But the difficulty at once arises, that the reference to authority either lands us in an infinite regress, or leads back to some one who recognized the principle for himself. A may believe it because B said it was true, and B because C said it was true, but sooner or later we come to some one who believed it on its merits. Further, I think we can by careful introspection distinguish the acceptance of a moral principle on authority from its acceptance on its own merits, as we can distinguish the stage at which we accepted mathematical principles on our teacher's authority from that at which we came to recognize their truth for ourselves. It is probably the case that many people all through their lives remain in the condition of accepting most of their moral principles on authority, but we can hardly fail to recognize in the best and most enlightened of men an absolutely original and direct insight into moral principles, and in many others the power of seeing for themselves the truth of moral principles when these are pointed out to them. There is really no more reason to doubt this than to doubt that there are people who can grasp mathematical principles and proofs for themselves.

(2) The other situation in which we read off the rightness of particular acts from some general principle also arises. The general principle may have been accepted not on authority but on its merits, but it may have involved for its recognition a fairly elaborate consideration of the probable consequences of a certain type of act; this would be true of such a principle as the principle that indiscriminate charity is wrong. In such a case the rightness or wrongness of an individual act falling under such a description is by no means self-evident. It would involve for its recognition a tracing out of the probable consequences, which we in fact do not perform; but we remember the general principle, while we have forgotten, or do not take the time to recollect, the arguments for it; and so we read off the rightness or wrongness of the particular act from it.

Our insight into the basic principles of morality is not of this order. When we consider a particular act as a lie, or as the breaking of a promise, or as a gratuitous infliction of pain, we do not need to, and do not, fall back on a remembered general principle; we see the individual act to be by its very nature wrong.

So far I have considered the type of case in which the thought of a conflict of duties does not occur to us, but we regard an act straight off as right or wrong in view of some obvious character that it has. It must be admitted that in a great part of our lives we think and act so. When we are asked a question, we do not as a rule doubt whether it is our duty to give a true answer. When we have made a promise, we do not as a rule doubt whether we ought to keep it. When we see an opportunity of relieving pain or distress without, so far as we can see, producing any bad ulterior results, we do not doubt whether we ought to do so. Yet in fact all these acts of ours will produce further consequences, and the probability is that any of them will produce some bad consequences. It may be asked whether we are justified in habitually ignoring this possibility. We cannot take Kant's line, that of holding that the act is so right in virtue of being a telling of the truth or a keeping of a promise that no further consequences it has can possibly make it wrong. For apart from the paradoxical consequences that this simple faith leads to, it is clear that the problem of conflict of duties breaks out even among the duties of perfect obligation, which Kant treats as absolute, and even within a single one of these duties. I may, for instance, be unable to keep one promise without breaking another.

Sometimes our simplification of the moral problem by viewing an act only under one category is plainly unjustified. A very little reflection would reveal probable consequences which make the act which we take to be right plainly wrong. Where the simplification is justified, it is justified by such considerations as these: An action which presents itself prima facie as right in virtue of some character it possesses—say, that of being the keeping of a promise or the relieving of another's pain—starts with reasons in its favour which go beyond what is expressed in describing it as the keeping of a promise or the relieving of another's pain. When we keep a promise we do more than keep faith with another person; we usually do something to strengthen the whole system of mutual confidence on which society is built up. When we break a promise, we do something to weaken this. So, too, when we tell the truth or tell a lie—which are in fact particular instances of keeping or breaking faith with another person.3 Again, if the immediate and most striking effect of an action is to relieve the pain or improve the character of another person, the argument for doing the act is not exhausted by that; for we know that happiness tends to radiate outwards from any one who is made happy, and goodness to radiate outwards from any one who is helped towards goodness, while pain and badness also tend to spread and radiate from one person to another. Thus an act which presents itself most obviously as conforming to one of the basic principles of morals starts with strong arguments in its favour; and we usually and justifiably suppose that unless some probable bad consequence reveals itself on a fairly brief inspection, the bad consequences are not sufficiently probable, or if sufficiently probable are not sufficiently weighty, to upset the strong prima facie argument in its favour.

Not only is it often justifiable to accept the fact that an act falls under one of the basic principles of morality, as sufficient reason for regarding it as right (or wrong) without further consideration. It is often justifiable to accept in the same way the fact that it falls under one of the media axiomata of morality. For mankind has for more generations than we can tell been exploring the consequences of certain types of acts and drawing conclusions accordingly about the rightness or wrongness of types of acts, and the media axiomata are the crystallized product of the experience and reflection of many generations. Suppose there is a medium axioma that actions of type A are wrong. Then any one who lightly does an act of this type, because he thinks some particular good result is likely to come of it and does not foresee equivalent bad results as likely to come of it, is in effect setting up his own very narrow experience against the experience of countless generations. He is in principle committing the same error as a child does who sets up his own very limited experience and immature judgement against the experience and judgement of older people. In the last resort we must use our own judgement as to what is right and what is wrong; but one of the factors of the situation which should very seriously affect our judgement is the fact that the orbis terrarum, although for reasons which may not be entirely clear to us, judges thus or thus about the type of act we propose to do.

It often happens, however, that no course of action presents itself as obviously called for by any basic moral principle or even by any medium axioma, or that incompatible actions present themselves as so called for. In such a case there is no escaping from the task of thinking out what it is that we ought to do. This task is one of greater or of less difficulty according to whether, when we come to reflect, some principle of special obligation, such as that of fulfilment of promise, is or is not seen to be involved. The latter is the less difficult case, and with it I will begin. Here the only principle of duty that we see to be applicable is that which bids us set ourselves to produce the greatest good. Our problem, then, is a twofold one: (a) to forecast the consequences of alternative actions, and (b) to estimate the comparative goodness of these consequences. In considering (a) a very strange fact at once presents itself. Generally speaking, no wish to produce a certain remote result leads to action which is effective in producing the result unless it is accompanied by knowledge of or opinion about some means which will effect the result. We do not need to know or have opinions about all the causal links that intervene between the means we set ourselves to produce and the final result we wish to produce. When I press the accelerator of my car, I produce the result that my car accelerates, though I may know nothing of the elaborate mechanism that produces this result. I may merely have discovered empirically that pressing the accelerator produces this result, or have learned from authority that it will. But at least, to make the car accelerate, it is not enough to wish or even to try to make it do so; I must set myself to press the accelerator. But at the very beginning of the causative process starting with an act of will quite a different state of affairs presents itself. If the teachings of physiology are correct, movements of members of the body, such as arms, or legs, or tongue, depend on movements of the controlling muscles; these depend on the stimulation of nerves passing from the brain to the muscle, and this in turn on some alteration in the brain. Thus, if the general order which I have stated held good in this case, we could effect a movement in a member only by setting ourselves to effect a certain change in the brain which we know or think will effect a certain change in the member we wish to move. But the fact is that, while any one who studies physiology may come to know this causal sequence, as ordinary moral agents we know nothing of it. If we have enough of a smattering of physiology to know in general that such a sequence exists, we certainly have not the remotest idea what sort of change in the brain will produce the wished for movement of the member. Thus we have something happening within the body that never happens outside the body, viz. that we can at will produce a certain result without having any knowledge or even opinion about any of the changes which are necessary preliminaries to this result. So far as our own awareness goes, we skip as it were the intermediate stages, and it seems to ourselves as if the mental effort to move the limb directly produced the movement.

I mention the problem, not because I have any light to throw on its solution, nor because I think it ethically important, but because it is interesting in itself and takes us deep into the whole problem of the relation of mind and body.

Whatever be the explanation, we start, then, not indeed with the knowledge that the mental effort to move a certain limb will in fact move it (for some lesion in brain or nerve or muscle may prevent this), but with what for practical purposes is generally as good as knowledge. And further, within certain limits of accuracy we may be said to ‘know’ the kind of movement that our effort will produce, that by trying to move an arm forward we shall in fact move it forward and not backward. As regards the further effects of an act of will, on bodies and on minds, we depend on analogical reasoning. We have a good deal, if not of knowledge, at least of highly probable opinion, as to the present condition of many of the bodies in our immediate vicinity and of the minds connected with them. The condition of the things in my environment and the final intra-corporeal effect of my act of will are the joint causes which will determine the first extra-corporeal effect; and using such probable opinion as we have of the condition of the things in our environment, and our experience of what effects similar intra-corporeal changes have had on bodies and minds similarly conditioned in the past (and we have amassed a good deal of such experience before we begin to think morally), we can form fairly probable opinions as to the first extra-corporeal effects of our act of will.

Our knowledge of its later effects is very much less, or rather our opinions about them are much less likely to be right. We can see that in this way. Let us first suppose, merely for the sake of argument, that there is no other agency at work except oneself causing changes in the world, or at least in one's environment. Then we may suppose that an action of ours will affect some of the things (minds and bodies) in our environment, while leaving others approximately unchanged. Let us denote things (substances) by capital letters and their successive states by attached numbers. Then by our action a set of things A1 B1 C1 D1 E1 F1 will be so changed as to become a set of things A2 B2 C2 D1 E1 F1. Then by the interaction of these things a further state of things will be produced, in which again some of the things will have been changed by their interaction, while others will be approximately unchanged. This state we may denote by A3 B3 C2 D2 E1 F1 Now we can perhaps anticipate with reasonable accuracy that the immediate result of our action will be to change A1 B1 C1 D1 E1 F1 into A2 B2 C2 D1 E1 F1; but we could foresee that its later result would be to produce the condition of affairs A3 B3 C2 D2 E1 F1, only if in addition to anticipating the change from A1 B1 C1 D1 E1 F1 to A2 B2 C2 D1 E1 F1, we could foresee the further change from this latter state to A3 B3 C2 D2 E1 F1. If our chance of being right about each change separately is one in two, for example, our chance of being right about the final result of both is much less. And it will diminish as we try to forecast effects further and further from us in time.

But the position is much worse than this in fact. We have simplified the problem immensely by supposing ourselves to be the only active agency at work. In fact, there are many other agencies, bodies and minds, at work altering our environment. We can foresee the first change with some approach to accuracy because the state of things A1 B1 C1 D1 E1 F1 is already in existence and more or less open to our observation. But we cannot anticipate with certainty that at time 2 the state of affairs will be A2 B2 C2 D1 E1 F1, as we could if we were the only agency at work and if we could rightly estimate the effect of our agency. By that time other agencies will or may have produced other changes in some or all of the substances in question. People, for instance, who are now alive and whom we may expect to be affected in some way by our action may by that time be dead or at a distance. Thus the difficulty of forecasting the future increases more rapidly than we have above suggested, as the future we try to forecast is a more and more distant future.

In the attempt to forecast the effects of our action, we are not limited entirely to reasoning by analogy from experience. To a very limited extent perhaps even bodily change may be anticipated a priori; it seems probable that a few simple laws of dynamics are known a priori to be true. But in forecasting effects on minds we can use a priori reasoning much more. We can anticipate, even apart from experience, that the announcement of a forthcoming pleasure will itself produce pleasure, and the announcement of a future pain, pain, that the news of some one's success will cause pain to his enemies, that that which is enjoyed while it is possessed will be to some extent missed when it is taken away. We have, I think, far more a priori insight into mental causation than into physical. But if many of our major premisses are won by insight and not by experience, the minor premisses which we must fit on to these if we are to draw conclusions about the future must be borrowed from experience. It is only by the help of experience that we can know that A is B's friend or that C is D's enemy, that E has enjoyed experience F in the past and will therefore be glad to be promised the future enjoyment of it, or that G has found experience H painful and will be sorry to hear that he is to have it again. All things considered, the difficulty of forecasting the future is so great that the slenderness of our insight into it is not to be wondered at. It is perhaps more surprising that wise men can often form such shrewd forecasts as they do.

When we turn (b) to estimating the goodness of the results of alternative actions, further difficulties confront us. These would be great enough even if pleasure were the only good; for not only is it extremely difficult to compare the intensity, and therefore the pleasure-value, of pleasures of very different quality, such as those of pushpin and poetry, to take Bentham's instances; it is extremely difficult to compare with accuracy the pleasure-value even of similar experiences. Yet it seems that in comparing somewhat similar pleasures we often have no difficulty in recognizing that one is more intense than another. And if we pass from a pair of similar pleasures to a pair of less similar pleasures, and so on, there does not seem to be in principle any point at which we should be justified in drawing the line and saying ‘up to this point comparison is possible; here it becomes impossible’. Thus in principle it seems to me that all pleasures fall on one scale in respect of intensity and are comparable in respect of it, though when the pleasures are very different in character it is only a very considerable difference in intensity that one can detect.

If we recognize, as I think we should, other goods than pleasure—virtuous emotion and action, and the exercise of intelligence—the difficulty of comparison becomes much greater.

Two views seem to be here possible. It may be held that all these things, including pleasures, are good in the same sense of ‘good’. Then the position will be that, just as it is easier to compare two similar pleasures in respect of intensity (and therefore of goodness) than two dissimilar ones, and yet dissimilar pleasures must be in principle comparable, so it is easier to compare two similar activities (e.g. two virtuous actions) in respect of goodness than to compare a virtuous action with an exercise of intelligence or with a pleasure, and yet in principle all three are comparable. If, on the other hand, as I think to be the case, good actions are good in a different sense of ‘good’ from that in which any pleasures as such are so, then good actions will not be comparable in respect of goodness with pleasures as such. Then, when we try to decide whether we ought to set ourselves to produce some good activity or some pleasure, the two things to be produced will not fall on one scale of goodness, but the two prima facie duties will still fall on one scale of obligatoriness and will be comparable thereon.

The choice between these two views must be deferred to the chapter in which the meanings of ‘good’ will be discussed.4 Meantime, however, we may discuss the position with regard to any good things which are good in the same sense of ‘good’, as, for instance, two virtuous actions may be properly held to be.

It has been suggested5 that there are not amounts, but only degrees, of goodness, and that in consequence all that we are entitled to assign to different goods is not cardinal numbers, implying that each good contains a certain number of units of goodness, but only ordinal numbers, implying that the two goods occupy different places on a scale of goodness, or are unequally far removed from the zero-point of indifference. Now such a state of affairs would be all that is needed if we had, in choosing which of two actions we should do, to compare a single good which will be produced by one with a single good which will be produced by the other. But this is not usually the case. Far more often we have to recognize that one or both of the two actions will affect for good or evil more than one person; and in such a case we are bound to attempt some summation of the goods and evils to be effected by each action. Let us for simplicity's sake suppose that only good effects are anticipated, and that only three goods are involved, whose order on the scale of goods is A, B, C (A being the nearest to zero), and that we have to choose between two actions, one of which will produce one, and the other the other two, of these goods. Then if (as the theory in question supposes) we knew only the order, but had no notion of the amount of goodness in any of the three goods, we should know that it was preferable to produce A+C rather than B, and B+C rather than A, but we should have no notion whether it was better to produce A+B rather than C. Similarly, if four goods were involved whose order on the scale of goods is A, B, C, D (A being the nearest to zero), and if at least one of the two actions will produce at least two of the goods, then we could (if goods had only ordinal and not cardinal numbers answering to them) in most of the cases6 decide which action would produce more good, but we should be quite unable to deal with the cases in which the effects of the two actions were to be as follows:

First action Second action

Now, in practice we are not conscious of this particular limitation. It certainly sometimes happens that when we think one action will produce one single good and the other a combination of lesser goods, we judge without hesitation, in some cases that the action which produces the single great good is rather to be done than the other, and in other cases that the other is rather to be done than it. I do not suggest that this is always so; it is perfectly clear that very often we should in such a case find it quite impossible to say whether the single good or the combination was to be preferred. But in principle, if we ever are justified in thinking that a certain combination of lesser goods is more worth (or that it is less worth) producing than a single greater good, we must know more about the goods than that they fall in a certain order on the scale of goods.

Nor will it be enough to know the size of the intervals that separate the goods. Suppose we know that B = A+M and that C = B+N. Then we know that A+B = 2A+M, and that C = A+M+N; but we should not know whether A is greater or less than N. We should need to be able to compare the intervals which separate the goods from one another with that which separates the smallest of them from zero. And this seems to me indistinguishable from recognizing each of the goods as containing a certain number of times a certain unit of goodness, i.e. from assigning to them cardinal as well as ordinal numbers.

If, then, we are ever able to say with confidence, comparing one greater good with the sum of a number of smaller ones, that it is more (or that it is less) worth producing than they, it is implied that each of the goods contains a definite number of times some unit of good. There is, of course, no natural unit of good. But we can arbitrarily take some small good and say that the goods we are comparing are twice, five times, &c, as good as it. Or, without having any particular unit of good in mind, we can say ‘whatever unit of good be taken, B would be worth twice as many of it as A, C five times as many as A’, and so on.

Now in fact we can never speak with as great precision as that. The position is rather this: the most that we can say with confidence is that B is worth not less than m times and not more than n times as much as A, and so on. It is clear that if we have this type of knowledge, then we shall sometimes be able and sometimes be unable to say of good C (for instance) that it is worth more (or less) than A+B. Suppose, for instance, that, taking some good G as unit, we can say

A = not less than 2G nor more than 3G
B = not less than 3G nor more than 4G
C = not less than 5G nor more than 7G’,

we shall not be able to say whether C is greater or less than A+B. But if we can say ‘C = not less than 8G,’ we shall know that in any case it is worth more than A+B.

This is, I believe, the kind of position in which we actually find ourselves. We should be justified, I think, in supposing that any good contains a definite amount of goodness, but since we cannot estimate this exactly but only as falling within certain limits, our knowledge is often not enough to enable us to compare one greater with two or more lesser goods. And, of course, the same difficulty often makes it impossible to say whether, of two single goods A and B, A or B is greater or A and B are equal.

The question may at this point be raised, whether such assessment of the goodness of the results of an action (or of the goodness of anything, for that matter) as we can reach is or is not reached by inference. The answer seems to be that it is not. If it were to be reached by inference, it would have to be either from premisses in one of which the term ‘good’ already occurred, or from premisses in which it did not occur. Now the latter is logically impossible; you cannot import a term into your conclusion which did not occur in one or other premiss. The former is not logically impossible. It would be logically possible that all judgements about the goodness of any particular results were deduced from premisses, of which one stated the goodness of a class of things and the other brought the individual thing under the class. But while this is logically possible, it seems to me, for reasons similar to those given before, with reference to rightness,7 not to be true in the case of our appreciation of goodness. If this view be correct, the apprehension of the degree of goodness of particular goods is logically immediate. But, of course, it does not follow that it is psychologically immediate. Goodness is a resultant attribute; it belongs to anything to which it does belong, because of the nature of the thing in some respect or other—because, for instance, it is a brave and not a cowardly act. And while even the vaguest apprehension of the goodness or badness of anything depends on some previous insight into the nature of the thing, an apprehension of the degree of its goodness will depend on close study of its nature, upon which the apprehension of the degree of its goodness supervenes, not as a logical conclusion but as a psychological result.

The psychological preliminaries to the judgement on the goodness of the results of an act will, of course, differ according to whether we are judging of an act already done, or of one not yet done. In the former case some of the results have probably already taken place and will be open to observation; but others lie in the future and require an effort of imagination for their envisagement. In the latter case all the results can only be apprehended by an effort of imagination. In both cases the imagination will presuppose reasoning—reasoning to the probable consequences of an act, based upon analogies drawn from previous experience. Thus in no case are the psychological preliminaries at all simple, and the more accurate our judgement of goodness is to be, the more careful must be our observation of achieved results and our imagination of results not yet achieved. It is hardly necessary to dwell on the difficulties of the analogical reasoning of which I have spoken—of the danger, for instance, of supposing that because one act has affected in a certain way the people mainly affected by it, a similar act will affect similarly quite different people.

A special complication is introduced into the judgement of goodness by the well-known principle of organic values, i.e. of values of wholes which are not equal to the sums of the values of their parts. But here the broad principle which I have stated above holds good—that the judgement of the goodness of the whole is logically immediate, but psychologically mediated by a study of the goodness of the parts. Take, for instance, the whole state of things constituted by a vicious act and the pain of the subsequent punishment. Here both elements, taken apart, are bad, but the whole has not a badness equal to the sum of the badnesses of the parts.8 We cannot, therefore, deduce its value from the values (using ‘value’ non-committally to cover badness as well as goodness) of its parts. Yet it is only if we envisage clearly the degree of badness of the vicious act and the degree of badness of the pain suffered that we can arrive at any definite view of the value of the whole which they compose.

It is not to be supposed that the existence of organic values vitiates any and every computation of the goodness of the total results of an act by summing the values of its individual results. Where the different effects of an act are effects on different persons, they do not, so far as I can see, coalesce into organic wholes, and it appears therefore to be safe to arrive at the goodness of the total results of an act by summing the goodnesses of the individual results—though, of course, the effects on any one person may form an organic whole whose goodness cannot be assessed by the process of summation.

One further complication remains. It will be remembered that in our consideration of the epistemological questions connected with the judgement of duty, we have so far considered only the duty of producing the maximum good. But there are other duties than this, the duty of fulfilling promises, the duty of making reparation for wrongs we have done, the duty of making a return for good we have received. Where such a special prima facie duty exists, as well as the general prima facie duty of producing the maximum good, our final judgement about our duty depends not on a comparison of goods but on a comparison of prima facie duties. But the same general principle reappears, that the final judgement is not a logical conclusion, but yet is something that presupposes preliminary mental acts, in which we study the situation in detail, till the morally significant features of it become clear to us.

Epistemologically, the position about duties of special obligation seems to me to be this. We all recognize their existence, but in two very important respects our judgements about them differ. (1) To most plain men these present themselves as duties independent of the duty of promoting the general good. To some philosophers they present themselves as merely derivative principles, flowing from the duty of promoting the general good, and ceasing to have any binding force whenever action according to them seems unlikely to promote in fact the general good. But (2) apart from this difference of view about the ground of the obligation to behave in these ways, people probably differ a good deal with regard to the degree of obligation which they think to attach to these principles of action. We may consider (a) how in this respect people who hold the teleological view will differ from those who hold the intuitionistic view. In general, the former will probably think that less obligatoriness attaches to the fulfilment of promises, for example, for they will think it is always out-weighed by the obligatoriness of any act which is likely to increase more the general good; while holders of the intuitional view will hold that some fulfilments of promise are more obligatory than some actions which are likely to increase more the general good. This is the general position; but it would be a mistake to expect that holders of the teleological theory will always take a laxer view about fulfilment of promise than holders of the intuitionist view. For teleologists, having to account somehow for the stubborn general disposition to regard fulfilment of promises as binding, are apt to explain this by referring to the tendency which breach of promise has to break down mutual confidence, and, in doing so, they are apt to exaggerate this tendency; so that a plain man recognizing an independent duty of fulfilling promises may easily think that much more good might be achieved by doing something else which involves breaking a promise, and that in this case the duty of promoting the general good outweighs the other, while a teleologist may have persuaded himself that the keeping of the promise will in fact bring more good into existence than any other act and that therefore the promise should be kept.

Apart from this difference between the attitude of teleologists and that of intuitionists towards promise-keeping, there are (b) no doubt considerable differences between intuitionists as to the degree of obligatoriness of promise-keeping. All that Intuitionism implies is the view that the duty of promise-keeping is independent and sui generis; it implies no particular view about the relative weight of this prima facie obligation compared with others. Within Intuitionism, we can have at one extreme the view of Kant that duties of perfect obligation always outweigh those of imperfect obligation. At the other end we might have people who think the duty of promise-keeping to be sui generis but yet to be one which very rarely outweighs the duty of promoting the general good. Thus in a particular case of conflict of duties of these two kinds, different inflationists (or different plain men) will give quite different answers. But this casts no doubt on the truth of the intuitionist view. It simply points to the fact that in this region our knowledge is very limited, that while we know certain types of act to be prima facie obligatory, we have only opinion about the degree of their obligatoriness. An exactly similar situation would reveal itself among teleologists, as soon as they began to face the question of the comparative goodness of different goods. Suppose them to agree that virtuous action, intelligent thought, and pleasure are goods; yet there is certainly no agreement about the comparative worth of these things. That casts no doubt on their being goods, and goods with different degrees of objective goodness; it only shows that our knowledge in this field is very limited. And so it is with regard to our knowledge of the relative obligatoriness of different prima facie duties.

It would perhaps be appropriate here to take account of an objection recently made to the kind of view I have been trying to state.9

‘If the most significant kind of morality is creative morality (and whether it is or is not can only be judged by the success of the application of the idea of creativeness to ethical concepts), then the ethics of intuitionism or deontology approaches from the wrong end. If we begin with the consideration of rational general rules we are bound to find out in time that the rules are inadequate to meet all cases and if we modify the theory and speak of “prima facie” duty versus “duty proper” or “actual duty”, we have still in the end to acknowledge a remainder, the surd of the individuality of the individual. We have to quote Aristotle again and say, “The decision lies with perception”. Indeed the root of the matter lies with perception, and at the best with a deeply imaginative perception linked to a consciousness of a larger good. This comes first, this is of prime importance, and what is left over, the considerable amount of life that is routine, may be dealt with approximately enough by rules and formulae. I for one, anyhow, believe that we get a fresher view of morality if we look at it from this angle.’

This objection, I think, rests on a misconception. If I have understood aright what Professor Reid means by creative morality, it is its aspect of spontaneity, of freedom from routine rules, that he wishes to emphasize. Now Intuitionism, in the form in which I hold it true, does not in any way condemn the moral life to routine. Such a charge might perhaps be brought against Kant's form of Intuitionism, in which it is held that the rightness or wrongness of an individual act can be inferred with certainty from its falling or not falling under a rule capable of being universalized. My criticism of this view is that it unduly simplifies the moral life. It ignores the fact that in many situations there is more than one claim upon our action, that these claims often conflict, and that while we can see with certainty that the claims exist, it becomes a matter of individual and fallible judgement to say which claim is in the circumstances the overriding one. In many such situations, equally good men would form different judgements as to what their duty is. They cannot all be right, but it is often impossible to say which is right; each person must judge according to his own individual sense of the comparative strength of various claims.

The criticism which Intuitionism as I hold it makes upon teleological ethics is that teleological ethics, in a different way from Kant's, over-simplifies the moral life; that it recognizes only one type of claim, the claim that we shall act so as to produce most good, while in fact there are claims arising from other grounds, arising from what we have already done (e.g. from our having made a promise, or inflicted an injury) and not merely from the kind of result our action will have, or may be expected to have. Intuitionism of this kind seems to me not to be hostile to creative morality in any sense in which creative morality is a good thing. I suppose that there could be no better instance of creative morality than the case of a man who, going beyond the routine of the duties commonly recognized by those round him, becomes convinced of some new duty and devotes his life to the discharge of it, as for example Wilberforce did when he devoted his life to the abolition of the slave-trade and of slavery. Creative morality involves not the denial or belittlement of the claims whose existence has long been recognized, but the coming to recognize new ones. If Intuitionism meant that people are to accept as absolute all the claims that are commonly recognized, and never to accept new ones, it would indeed be adverse to creative morality. But it means neither of these things. The general principles which it regards as intuitively seen to be true are very few in number and very general in character. With regard to all media axiomata, which are attempts to apply these general principles to particular types of situation, it preserves an open mind. It recognizes that new circumstances sometimes abrogate old claims and sometimes create new ones, and that we must be constantly alive to recognize such changes and to act on them.

So far I have been speaking of the problem of discovering which of the actions open to one would be objectively right, would discharge in the fullest possible measure the various claims or prima facie duties that are involved in the situation. This is what we should like to know; and it is clear, in view of the various difficulties I have pointed out—the difficulty of comparing the goodness of various results, the difficulty of balancing the duty of producing the greatest good against special obligations—that we can never know our duty in this sense, but can only reach more or less probable opinion about it. At the same time the difficulties are not so great as to make the attempt useless. We can by the use of analogical reasoning from experience, and of a priori reasoning, forecast with some confidence the nearer consequences of our acts; and in certain cases (as we have seen10) there is some reason to suppose that of two acts that which has the better proximate consequences will also have the better remote consequences. And again in comparing goods, and in comparing prima facie duties, while we are often in doubt which is the greater good or the more stringent obligation, in other cases, where the one good is much the greater or the one obligation much the more stringent, we seem to be able to grasp these facts with certainty. The fact that in many individual cases the people whose judgement we have learned most to respect in ethical matters will pronounce the same judgement on acts is some guarantee that objectivity has been attained.

We have, however, in an earlier chapter11 come to see that besides the objective duty of which we have been speaking, there is a subjective duty; there is an act which we think likely to be the maximum fulfilment of objective duty. In our attempt to discover objective duty, whether it succeeds or fails, we can at least discover our subjective duty; for we can come to know what it is that we think. At the same time, there are cases in which we do not even think any one act to be likely to be the completest fulfilment of our prima facie duties—in which we are quite doubtful as between two or more acts. But even so we are not completely ignorant; for we at least think that the right act is one of a limited number of acts. In such a case it is our subjective duty, and we know that it is, to do one or other of the acts, of which we think one or other to be our objective duty.

  • 1.

    Eth. Nic. 1155 a 16–22.

  • 2.

    Sermon I (Gladstone's ed.), 38–9.

  • 3.

    Cf. pp. 112–13.

  • 4.

    Ch. 11.

  • 5.

    By Professor H. H. Price, in Mind, xl (1931), 353.

  • 6.
    Viz. (ignoring cases in which the effects of the two actions include an identical good, which may be cancelled out) when the effects of the two actions are to be as follows:
    First action Second action
    A S+C+D
    A B+C
    A B+D
    A C+D
    B A+C+D
    B A+C
    B A+D
    B C+D
    C A+B+D
    C A+D
    C B+D
    A+B C+D
    A+C B+D
  • 7.

    pp. 169–73.

  • 8.

    If the parts are, as I think, bad in two different senses of ‘bad’ (cf. pp. 271–9), there is, of course, no sum of their badnesses. But even if they are bad in the same sense, the whole has not a badness equal to the sum of the badnesses of the parts, since the fittingness of the punishment to the sin takes away from the badness of the whole.

  • 9.

    L. A. Reid, Creative Morality, 109 f.

  • 10.

    pp. 173–5.

  • 11.

    Ch. 7.

From the book: