The Foundations of Morality

CHAPTER 14 The Problem of Self-Sacrifice

1. “Individual” and “Society”

Let us summarize the discussion in the two preceding chapters to see where it has brought us.

We have seen that there tends to be a coincidence between the actions or rules of action that best promote the interests of the individual in the long run and the rules of action that best promote the interests of society as a whole in the long run. We have seen that this coincidence tends to be greater the longer the period we take into consideration. We have seen, moreover, that it is difficult to distinguish “egoistic” actions from “altruistic” or “mutualistic” ones, because an enlightened and farsighted selfishness might often dictate precisely the same course as an enlightened and farsighted benevolence.

Now each of us sees himself sometimes as the individual and sometimes as a member of society. In the former role he is apt to emphasize the necessity of liberty and in the latter the necessity of law and order. A as a member of society is concerned that neither B nor C do anything to injure him. He insists that laws be passed to prevent this; and injuries that cannot satisfactorily be prevented by law he seeks to prevent by condemnation or disapproval. But he soon realizes that he cannot consistently or successfully use devices of condemnation or praise to influence the actions of others without accepting them for like actions by himself. Both to seem consistent to others and to be consistent in his own eyes (for the “rational” man tends to accept consistency as an end in itself) he feels an obligation to accept for himself the moral rules he seeks to impose on others. (This is part of the explanation of the origin and growth of conscience.)

And the moral rules that we seek, for egoistic reasons, to impose on others, do not stop at inducing them not to inflict positive injury on us. If we found ourselves on board a ship sinking at sea we would think it the moral duty of those on any vessels near by to answer our SOS signals, and to come to our rescue, even at considerable risk to themselves.

I do not mean to imply by this that all moral rules arise out of egoistic considerations. There are people who are spontaneously so moved by the suffering of others or a danger to others that they do not need to imagine themselves in the same predicament in order to think it their duty to come to the rescue of others. They will do so out of their spontaneous desire. Nearly all of us, in fact, do take spontaneous satisfaction in the happiness of others—at least of some others. What I am concerned to point out is that even if we were to assume, with Hobbes, that people are guided only by egoistic motives, we would probably arrive at the conclusion that they would be driven, in the end, to impose virtually the same outward code of morals on each other as if they were guided by altruistic motives as well. And because it is to the interest of each individual to live in a society characterized not only by peace and order and justice, but by social cooperation and mutual affection and aid, it is in the interest of each individual himself to help to create or preserve such a society through his own code and his own example.

We must repeat once more, then, that the antithesis between the interests of the Individual and the interests of Society is false. Normally and usually the actions that best promote the happiness and well-being of the individual best promote the happiness and well-being of the whole society. There is normally, to repeat, a coincidence between the long-run interests of the individual and the long-run interests of society. But we must frankly face the fact that there is not a complete identity. There will be times when the interests of the individual, even his interests in the long run, appear in his own eyes to conflict with those of society. What, then, is his duty? By what rule should he be guided? What should the moral code prescribe?

2. Duty vs. Risk

But surely it cannot be wrong to do anything to benefit oneself simply because an incidental consequence may be to hurt or injure the interests of another. Should one reject the offer of a better job than one already has, simply because the present occupant, or another candidate, may then lose that particular job and may not be able to get another as good? Should a scientist refuse to publish a truthful criticism of another scientist’s work because the result of that criticism may be to increase the first scientist’s reputation at the cost of destroying the reputation of the scientist criticized? Evidently the rule proposed by Adam Smith would have to be carefully qualified to forbid injury to others only through coercion, violence, malice, misrepresentation, or fraud—i.e., the class of actions forbidden must be only those that tend to injure the long-run interests of society as a whole, and the class of actions prescribed must be only those that tend to benefit the long-run interests of society as a whole.

But what should be the rule when the risk to the would-be rescuer rises? Here the problem becomes difficult, and the answer may depend not only on the degree of the risk, but on the relationship (whether, e.g., that of parent or of stranger) of the potential rescuer to the person or persons to be rescued. (It may also depend on a numerical relation. For example, whether the situation is [1] one in which one person, say a sapper, or soldier whose job it is to get rid of enemy mines, may be asked to risk his life to save a hundred or a thousand, or [2] one in which a hundred or a thousand may be asked to risk their lives to save only one, say a king or a president who is being held as a hostage.)

The ethical problem here may be difficult to answer precisely because, for example, the degree of risk being run may be indeterminable unless the risk is actually undertaken. Many a man has been tortured by conscience all the rest of his life because he has suspected that cowardice or selfishness led him to overestimate a risk that he refused to take to save another.

If we turn for help to the answers given by traditional ethical systems and by “common-sense” ethics we find them to be in some cases not only clear but stern. There are conditions under which these traditional codes demand not only that a man risk his life for others but that he be willing, indeed, to sacrifice it. A soldier who deserts or runs away in battle, a captain who violates the rule that he should be the last to leave his ship, a doctor who refuses to enter a city where there is an epidemic or to attend a patient suffering from a contagious disease, a fireman (or father) who fails to try to rescue a child or an invalid from a fire, an armed policeman who stands idly by or runs away when an innocent citizen is being held up by a bandit at the point of a gun—all these are condemned by nearly every traditional or common-sense moral code.

And the reason for this condemnation is plain. A nation that cannot depend on the bravery and self-sacrifice of its armed forces is doomed to conquest or annihilation. The inhabitants of a city who could not depend on the willingness of their policemen to take risks would be overrun by criminals, and would not be safe in the streets. The welfare and survival of a whole community, in brief, may depend upon the willingness of certain individuals or groups to sacrifice themselves for the rest.

But the duty is not always clear. If an unarmed citizen happens to be near when another unarmed citizen is being held up at gunpoint, is it the duty of the former to try to take the gun away? If even a hundred other unarmed citizens are by when a bandit is robbing one of them at gunpoint, is it the duty of one of the bystanders to try to take the gun away? And which one? No doubt collectively they could succeed; but it is the first to try who takes the greatest risk.

The answer of common-sense ethics to this situation is far from clear. The people who read in the next day’s newspapers about a thug shooting a victim and getting away because a crowd of a hundred did nothing to stop him, may be righteously indignant, and contemptuous of those who were too cowardly to act. Some of those who were in the crowd will feel secretly ashamed of their inaction, or at least a little uneasy. But most of them will argue to themselves or others that it would have been an act of sheer foolhardiness for them to take the initiative in interfering.

3. Search for a General Rule

Can we find the answer to the problem of self-sacrifice in any general rule or principle?

I think we can reject without any further argument the contention of a few contemporary ethical writers that it is never the duty of an individual to sacrifice himself for others, or that it is even “immoral” for him to do so. The examples we have cited, and the reasons why such self-sacrifice may sometimes be necessary, are sufficient and clear.

With these two extremes out of the way, we can try to formulate an acceptable rule. Suppose we frame and examine the rule as follows:

Now it is at least reasonably clear that no one should sacrifice his own interests to another or others unless a greater good is accomplished by the sacrifice than is lost to the agent. This is clear even from the most impartial view. Any rule of action should tend to promote a net gain of good on the whole rather than a net loss.

4. The Concept of Costs

Here we may draw a parallel not only with what has already been said about the requirements of simple prudence, but with the whole conception of costs in human action. The only rational prudential reason why a man should give up a pleasure, a satisfaction, or a good is to gain a greater pleasure, satisfaction, or good. This greater good may, of course, be nothing more than the absence of the subsequent pain or suffering caused by excessive indulgence in the pleasure given up—as a man may give up excessive drinking or smoking or eating in order to feel better in the long run—to improve his health and prolong his life. Prudential sacrifices are usually sacrifices of immediate pleasures or satisfactions in order to enjoy greater future happiness or satisfactions.

Costs thus conceived in “real” terms are sometimes distinguished by economists from money costs by the special name opportunity costs. This means, as the name implies, that we can do one thing only at the expense of foregoing something else. We can seize one opportunity only at the cost of foregoing what we consider the next best opportunity. Mises defines the concept in its broadest form:

This concept, unfortunately, is not very commonly under-stood or applied by writers on ethics. When we do apply it to the moral field, it is clear that every action we take must involve a choice of one value at the expense of other values. We cannot realize all values at once. We cannot realize more of one value without realizing less of another. We cannot give more time to learning one subject, or developing one skill, for example, without giving less time to learning some other subject or developing some other skill. We cannot achieve more of one good without achieving less of some other good. All good, all value, can be achieved only at the cost of foregoing some lesser good or value.

This higher net value gained is of course the test of decisions and actions that concern oneself alone. It is the justification of the prudential virtues. But it should also be the test of actions that affect others. A man’s duty cannot require that he give up any good of his own except for the greater good of another or others. In fact, it can reasonably be argued that it would be immoral for him to go beyond this—to sacrifice his own good to confer a lesser good on others. For the net effect of this would be to reduce the amount of good, to reduce the amount of happiness and well-being, in the universe.

Now what are we to say of the argument, by such moralists as Kant, and more recently by Grote, Hastings Rashdall, and G. E. Moore, that Self-Sacrifice, or Duty, or Virtue (usually spelled with a capital to impress the point) is itself an end, or even the end?

I must content myself here with saying that I consider self-sacrifice essentially a means—a means sometimes necessary for promoting the end of maximum happiness and well-being for the whole community. But its value is wholly instrumental or derivative (like the value, in economic life, of irksome labor, or a raw material or a capital good). To the extent that an overzealous or misdirected self-sacrifice tends to reduce the sum of human happiness and well-being, its value is lost or becomes negative. It is therefore a mere confusion of thought to consider Self-Sacrifice (or Duty or Virtue) an additional good or value independent of the ultimate purpose it serves.

What leads to the confusion is the difficulty, if not the impossibility, of conceiving of a society in which happiness and well-being were maximized but in which nobody ever sacrificed his short-run interests to the long-run interests of others, in which nobody ever did his duty, and in which nobody had any virtues. But the reason for the difficulty or impossibility of conceiving such a society is that is involves a self-contradiction in concept and in terms. For the same reason it would be an impossibility to conceive of an economic community in which the production of ultimate consumer goods and services was maximized without the use of labor, raw materials, factories, machines, or means of transport. What we mean by rational Self-Sacrifice and Duty and Virtue is performing acts that tend to promote the maximum of happiness and well-being for the whole community and refraining from acts that tend to reduce such happiness and well-being. If the effect of Self-Sacrifice were to reduce the sum of happiness and well-being it would not be rational to admire it, and if the effect of other alleged duties and virtues were to reduce the sum of human happiness and well-being, we would cease to call them duties and virtues.

But the questions raised here are so wide that we may later have to return to them for more extended consideration.

This may be a useful point for a semantic digression. In using the word “Self-Sacrifice,” and in contending that there are occasions, however rare, when it is necessary, I am probably courting resistance from some readers to whom Self-Sacrifice means the equivalent of self-abasement and self-immolation, of asceticism and martyrdom. Many of these readers would find this view more acceptable if I used some milder term, like Self-Subordination. But the difficulty with this milder term is that it refers to a milder thing. Self-Sacrifice, as I conceive the term, is a duty that most of us are called upon to exercise only on a few rare occasions of crisis; self-subordination is a duty that most of us are called upon to exercise almost daily. We subordinate our own ego or our own immediate interests to wider interests whenever we refrain from starting to eat until everybody at the table has been served; or whenever, as part of an audience, we hear a speaker out without heckling or rushing up to the platform ourselves; or whenever we restrain a cough, at some inconvenience to ourselves, during, say, the soft bars of a symphony. Every member of a family, and especially the parents and the older children, must habitually practice self-subordination if family life is to be possible. But this self-subordination is something that each individual implicitly recognizes as necessary to the harmonious social cooperation that is in turn necessary to promote his own long-run interests.

5.Obligations Have Limits

A similar view, more violently expressed, appears in a letter to Lady Gray from Sydney Smith in 1823:

For God’s sake, do not drag me into another war. I am worn down, and worn out, with crusading and defending Europe, and protecting mankind: I must think a little of myself.

I am sorry for the Spaniards—I am sorry for the Greeks—I deplore the fate of the Jews; the people of the Sandwich Islands are groaning under the most detestable tyranny; Baghdad is oppressed: I do not like the present state of the Delta; Thibet is not comfortable.

Am I to fight for all these people? The world is bursting with sin and sorrow. Am I to be champion of the Decalogue, and to be eternally raising fleets and armies to make all men good and happy? We have just done saving Europe, and I am afraid the consequence will be, that we shall cut each other’s throats.

No war, dear Lady Gray—no eloquence; but apathy, selfishness, common sense, arithmetic; I beseech you, secure Lord Gray’s swords and pistols, as the housekeeper did Don Quixote’s armour. If there is another war, life will not be worth having. I will go to war with the king of Denmark, if he is impertinent to you, or does any injury to Howick; but for no other cause.

Several moral strands are twisted together in both of these arguments. In the quotation from Sydney Smith the question whether the people of other countries should be helped is entangled with the question whether war is a desirable way to help them. But the implication of his plea for “apathy, selfishness, common sense, arithmetic” is that it is folly to sacrifice one’s own comfort for millions of unknown foreigners. Adam Smith’s chief reason, however, for dismissing “this extreme sympathy with misfortunes which we know nothing about” as “altogether absurd and unreasonable” is that, though “all men, even those at the greatest distance, are no doubt entitled to our good wishes,” we are in a position in which “we can neither serve nor hurt” them.

It is precisely this argument which would today be questioned. Americans are not only being importuned by private charities, but compulsorily taxed by their own government, to give food and aid and dollars to millions all over the world whom they will never see. What is their real obligation in this field? And when can they consider it discharged?

Suppose we conclude that sacrifice is required whenever it will yield more happiness to those for whom it is made than it will cost in happiness to those who make the sacrifice? It could plausibly be argued that, when we give this an objective or material interpretation, it would require us to keep giving away our fortunes or income or food as long as we had any more of any of these than the most miserably housed or clothed or fed person alive. We should have to keep giving, in other words, down to the point of absolute world equality of income and living standards.

Such an equal distribution of income, housing, clothing, and food, quantitatively and qualitatively, would be, of course not only physically impossible, but inconceivable. The attempt to achieve it, even by “voluntary” means and through pure moral approval and disapproval, would so tremendously reduce the incentives to work and production at both ends of the economic scale as to lead toward universal impoverishment. It would enormously reduce, and not increase, the sum of human happiness and well-being. The attempt to achieve such an egalitarian altruism, the attempt to impose such practically limitless and bottomless responsibilities, would bring misery and tragedy to mankind far beyond any harm resulting from the most complete “selfishness.” (In fact, as Bishop Butler pointed out, and as many have recognized since, if everyone were constantly guided by a rational, enlightened, and far-sighted “egoism,” the world would be an immensely better place than it is).

But, some readers may say, I have been presenting an argument that does not really touch the rule we have been testing. By hypothesis, the sacrifices we are enjoined to make are only those that will yield more happiness in the long run to those for whom they are made than they will cost in less happiness (in the long run) to those who make them. Therefore we are asked to make only such sacrifices as will tend in the long run to increase the sum of happiness.

This is true. But even if we bypass here the crucial question whether it is possible to speak validly of a sum of happiness, or possible to compare the “increase” of one man’s happiness with the “decrease” of another’s, the preceding discussion will also show that it is very dangerous to give this principle any merely physical or short-term interpretation—or to base our duty, say, on any mathematical income comparisons. The less our active sympathies with the persons we are called upon to help, the more remote such persons are from our direct acquaintance and daily lives, the more reluctant we will be to make any sacrifice to help them, the less satisfaction we will take in any sacrifice—and, conversely, the less likely are those helped to appreciate the sacrifice on our part or to be permanently benefited by it.

The ethical problem here is complicated by the fact that certain acts of so-called “sacrifice” are not considered by those who make them to be sacrifices at all. Such are the sacrifices that a mother makes for her child. Certainly as long as the child is very young and truly helpless, most such sacrifices may directly and immediately, as well as in the long run, increase the happiness both of the one who makes the “sacrifice” and the one for whom it is made. Such sacrifices present an ethical problem of limitation only when they are carried to the point where they may either permanently impair the ability of the benefactor to continue his or her sacrifices or where they coddle or spoil or in some other way demoralize the child or other intended beneficiary.

6. Maxima and Minima

But the problem we are concerned with here is whether it is possible to frame a general rule to apply to the duty or limits of self-sacrifice—for the benefit of people, say, whom we may not know, or even for the benefit of people whom we may not like. One difficulty of such a general rule is that it cannot be simple. Our duty or non-duty may depend upon the relations, as I have previously hinted, in which we find ourselves with other people, relations which may sometimes be accidental. Thus if we are walking along a lonely road, even if we are on a temporary visit to a foreign country, and find a man who has been seriously injured by an automobile, or robbed, beaten, and left half dead, we cannot pass by “on the other side” and tell ourselves that the whole matter is none of our business, and besides we are late for an appointment. Our duty is to act as the Good Samaritan did. But this does not mean that our duty is to take all the world’s burdens on our own shoulders, or to keep constantly touring around trying to find people to save, regardless of how they got into their predicament or what the long-run effect of our rescue operations would be on them.

This means that we must carefully distinguish between the special case and the general rule, or even between any single instance considered in isolation and a general rule. If you give a dollar to a beggar, or even $1,000 to a chance pauper who “needs” the money more than you do, a mathematical comparison of the supposed marginal utility of the money to him with its supposed much smaller marginal utility to you (assuming such a comparison were possible) may seem to result in a net gain of happiness for the two of you considered together. But to erect this into a general rule, to impose it as a general obligation, would result in a net loss of happiness for the community considered as a whole.

In brief, a single act of indiscriminate charity (or discriminate only in the sense of moving toward equalization of income without any other criterion) may seem to increase the happiness of the recipient more than it reduces the happiness of the donor. But if such extensive and practically limitless charity were erected into a general moral rule imposed on us it would lead to a great diminution of happiness because it would encourage permanent mendicancy in increasing numbers of people, who would come to regard such help as a “right,” and would tend to discourage effort and industry on the part of those on whom this moral burden was imposed.

But if the problem cannot be solved with precision, it does not follow that it cannot be solved at least within certain upper and lower limits. The upper limit, as we have seen, is that no act of self-sacrifice is justified unless it secures for another a greater good than the good that is sacrificed. The lower limit is, of course, that one should refrain from any positive harm to one’s neighbors. In between is a twilight zone of obligation.

7. Self-interest vs. Morality?

The problem we are concerned with in this chapter may be stated in another form. In Chapter 7 we were tempted to define morality as “essentially, not the subordination of the ‘individual’ to ‘society’ but the subordination of immediate objectives to long-term ones.”

Each of us, in his own long-run interest, is constantly called upon to make temporary sacrifices. But does morality require us to make “genuine” sacrifices—that is, sacrifices on net balance, sacrifices from which we cannot hope to realize any fully compensating gain even in the long run?

An enlightening but paradoxical answer to this question has been offered by Kurt Baier. I quoted part of it in Chapter 7 (p. 51). Now I should like to quote it more at length and analyze it more fully, because it poses what is perhaps the central problem of ethics:

Moralities are systems of principles whose acceptance by everyone as overruling the dictates of self-interest is in the interest of everyone alike, though following the rules of a morality is not of course identical with following self-interest. If it were, there could be no conflict between a morality and self-interest and no point in having moral rules over-riding self-interest. . . .

I have already pointed out one weakness in this ingenious statement. Its air of paradox stems from the use of the word “self-interest” in two different senses. If we distinguish immediate or short-term interest from long-run interest, much of this paradox disappears. Thus the proper statement is: Moralities are systems of principles whose acceptance by everyone as overruling the apparent dictates of immediate self-interest is in the long-run interest of everyone alike.

It is self-contradictory to say that “It is in the interest of everyone alike that everyone should set aside his interest.” But it is not self-contradictory to say that it is in the long-run interest of everyone alike that everyone should set aside his mere momentary interests whenever their pursuit is incompatible with the long-run interests of others. It is self-contradictory to say that “it may be in one’s interest not to follow one’s interest at times.” But it is not self-contradictory to say that it may be in one’s long-run interest at times to forego some immediate interest.

Emphasis on the distinction between long-run and short-run interests solves half the problems raised by Baier’s statement, but it does not solve them all. The rest exist because of possible conflict or incompatibility in the interests of different people. But is there therefore a contrast between the requirements of “enlightened self-interest” and the requirements of “morality”? The moral rules are precisely the rules of conduct designed to maximize the satisfactions, if not of everyone, then of the greatest number of persons possible. The enormous gain to everyone of adhering faithfully to these rules entirely outweighs the occasional sacrifices that this adherence involves. I am tempted to say that for 99 per cent of the people 99 per cent of the time, the actions called for by enlightened self-interest and by morality are identical.

I have said that Baier’s antithesis between “self-interest” and “morality” depends for its plausibility upon the use of the word “self-interest” in two different senses—upon his failure to distinguish between short-run and long-run interest. It is ambiguous in another important sense also—in his conception of self-interest and his conception (elsewhere in his book) of “egoism.” If we (implicitly or explicitly) define “egoism” and “self-interest” as “disregard of or indifference to the interests of others,” then Baier’s antithesis stands up. But this is because our use of words has begged the question. This is because we have implicitly defined the “egoist” as a cold calculating person who habitually regards his “self-interest” as conflicting with the interests of others. But such “egoists” are rare. Most people do not consciously pursue their self-interest but merely their interests. These interests do not necessarily exclude other persons. Most people feel spontaneous sympathy with others and take satisfaction in the happiness of others as well as of themselves. Most people recognize, however dimly, that their principal interest is to live in a moral and cooperative society.

Yet all this, it must be conceded, is only a partial answer to Baier’s formulation. It is not conclusive. There remains the rare case when the individual may be called upon to make a “genuine” sacrifice. This is the occasion when a soldier, a ship-captain, a policeman, a fireman, a doctor, or perhaps a mother, father, husband, or brother, may be called upon to risk or to lose life itself, or to be maimed for life, in the fulfillment of some clear responsibility. There is then no future “long run” that can compensate for the sacrifice. Then society, or the rules of morality, say in effect: This risk you must take, this sacrifice you must make, whether or not you consider it in your own enlightened self-interest, because it is in the long-run interest of all of us that each abide unfalteringly by the responsibilities that the established rules of morality may lay upon him.

This is the price that any of us may be called upon some day to pay for the untold benefit that each of us derives from the existence of a code of morals and its observance by all the rest.

Baier is wrong, in brief, in contrasting “morality” and “the pursuit of self-interest.” Moral rules are designed precisely to promote individual interest to the maximum extent. The true contrast is between the kind of self-interest that is incompatible with the interest of others and the kind of self-interest that is compatible with the interest of others. Just as the best traffic rules are those that promote the maximum flow of safe traffic for the most cars, so the best moral rules are those that promote the maximum self-interest for the most people. It would be a contradiction in terms to say that the maximum interest of all was promoted by everyone’s restricting the pursuit of his own interest. True, some must forego the pursuit of certain apparent or temporary advantages because these are of the kind that would thwart the achievement of the real interests not only of most others but even of himself. But the happiness of all cannot be maximized unless the happiness of each is maximized.

If we have a society consisting (let us say for simplicity) of only two people, A and B, then the rules of conduct they should adopt and adhere to are not those that are solely in A’s interest, nor solely in B’s interest, but most in the long-run interest of both. The rules that are most in the interest of both must be in the long run the rules that are most in the interest of each: This remains true when our hypothetical society is increased from A and B to everybody from A to Z.

This mutualism is the reconciliation of “self-interest” and “morality.” For one best promotes one’s own interest in the long run precisely by abiding by the rules that best promote the interest of everyone, and by cooperating with others to hold everyone else to those rules. If it is to everyone’s long-run interest to adhere to and uphold the moral rules, it must therefore be to mine.

We come back once more to the conclusion that the real interests of the individual and of society nearly always coincide, but are not (such is our human predicament) in every case identical.