1. IN the preceding chapter we stressed the importance of the two facts that men are approximate equals by nature and that they can join hands to co-operate. Thanks to the first the possibilities of individual men to exercise normative authority over other men are narrowly limited. The significance of the second fact we said was twofold. Thanks to it men can withstand and overrule any ‘tyranny’ which one man may attempt to exercise over his equals by commanding them. But thanks to it men can also subjugate their natural equals and exercise joint normative authority over them. Both consequences of men's capacity to join strength are simultaneously at work in society. The significance of both can be summed up in the words ‘society is stronger than the individual’.

Let us raise the question: What makes men join hands in cooperative efforts?

One thing which makes men co-operate is the practical necessity autonomously incumbent upon each of the members of a group of men of working together for the attainment of a common end. That the end is common shall mean that each of the men in the group wants a certain state of affairs p to exist. That working together is necessary shall mean that if any one of the men stays away from work the end will not be attained.

This case of co-operation under autonomous necessity is an extreme case. But it is an important prototype.

If ends are goods attainable through action an end which is common to several agents can be called a common good. Not everything however which is a common good i.e. a good for each member of a group of men is also a common end. That something is a good for a man means that it is something which is in itself welcome to him. But not everything a man welcomes if it befalls him is an end of his action. Hence though common ends are also common goods all common goods are not also common ends.

Co-operation is for some end. This end can be a common good of all the co-operants. We shall call this the first case. Or it is a common good of some of them. We call this the second case. Or it is a good for none of them. This is the third case.

In the first case there are two possibilities. Either the men have to work for the common good under autonomous practical necessitation or some can stay aside and let the rest work or relax and let the rest work harder.

The first of these two possibilities answers to the prototype case which we have already described.

The second possibility means that a group of men have a common end but that it is not necessary that they all work for it. We can in every such case distinguish a minimum number of necessary co-operants and a maximum number of possible bystanders. This division gives rise to a number of problems which we cannot discuss here. We give only one example: Three men have a common end of their action. The co-operation of two of them is necessary in order that the end shall be attained. But one man can stay aside and look on when the other two work. Assume that x and y work and that z stays aside. To make z participate in the work for the common good of the three men may then become another common interest of x and y. Since they are two and z is one they will probably be able to order z to co-operate with them. If z remonstrates they may jointly proceed to punish him. The cooperation of x and y in commanding z is co-operation in the sense of our prototype case.

We proceed to the second case. Then the end of co-operation is a common good of some but not of all of the co-operants. For example: Three men co-operate for a result p. Two of them are anxious that p shall be. For the third however p is not a good Perhaps he co-operates because he has been commanded by the other two and is anxious to escape punishment. If this is the case then making the third man work is another common interest of the first two men—beside the attainment of p. The co-operation of the two for this common end again is co-operation of the prototype case i.e. of the case when men are under autonomous necessitation to co-operate for a common good. If however the group of men in pursuit of the first common good is great then co-operation of a few of them may be sufficient for commanding some men from outside the group to labour for their end. Then the problem of co-operation for commanding outsiders becomes a case of the problem of co-operation in a group which may be divided into necessary co-operants and possible bystanders. This problem we already discussed.

Consider finally the third case. Here the end of co-operation is not an end of any man in the group of co-operants. If this is the case their co-operation serves as a means to some end of a being or of some beings outside the group. Use of this means may have to be secured by commanding the men to work. Then the problem of co-operation for a common end will arise for the men for whose ends the co-operants are assumed to work. This problem falls under our first case.

By these considerations I have wanted to indicate how that which I called a prototype case of co-operation for a common good holds a crucial position among a variety of cases of co-operation. I would maintain that all cases which are not in themselves of this prototype and in which co-operation has to be secured by commanding rest—in the last resort—upon autonomously necessitated co-operation for a common good.

There are however forms of co-operation which do not necessarily serve the attainment of a common good. We now turn to a study of some such cases.

2. Let it be assumed that x and y join hands to attain an end of x's which is not also an end of y's. x could not attain his end alone. He needs the assistance of y. Let us rule out the possibility that x can command y to serve him. Since men are approximate equals the assumption that the one cannot command the other is thoroughly realistic. Let us further rule out the possibility that y wants to help x because he wants to promote the good of x for its own sake. Since men even if they are not unlimitedly egoistic are not overwhelmingly altruistic either this assumption is thoroughly realistic too. Is there then any motive left which could make y co-operate?

There is at least one possible motive left: hope of reciprocal service in return. y could argue: if I do not help x to attain p now x will probably not help me to attain q on another occasion. If I help x now there is at least a good chance that he will help me then considering that he may need my assistance on some other occasion. Hence it is in my interest to help him.

This form of co-operation may be said to hinge logically upon three facts. The first is that men are approximate equals and cannot normally exercise commanding authority over one another. The second is that men far from always pursue the good of their neighbours as an ultimate end of their own. The third is that men are to some extent dependent for their welfare upon the assistance i.e. co-operation of others. Co-operation which is not for a common good can yet be to mutual advantage.

Co-operation which is to mutual advantage but not for the attainment of a common good is exchange of services.

Exchange of services be it observed need not take the form of co-operation. y does something alone say p which is wanted or welcomed by x. p is thus a good for x. x in return does something say q which is wanted by y. The exchange of services which has taken place can be called an exchange of goods.

Exchange of services and of goods plays an important role in the conceptual genealogy of the normative ideas known as agreement contract and promise. I shall not however discuss them here.

When exchange of services or of goods takes place people can be said to work in turns for the promotion of one another's good.

Hope of reciprocal service or of a good in return can thus be a motive for co-operation and for promoting another man's good generally. But it can also be a motive for respecting it.

A man may gain an advantage for himself by harming his neighbour i.e. by doing something which his neighbour thinks unwanted. A man cannot command another man of roughly equal strength to respect his good. But if the other man harms him he may be able to ‘pay back’ revenge himself and thus return evil by evil. Revenge may be regarded logically as a preform of punishment and threat of revenge as a preform—on the level of rough equality of strength among agents—of normative relationships. One could say that revenge and retaliation are what punishment ‘logically’ deteriorates into when approximate equals try to command one another.

To return evil by evil is also an exchange of services or of goods of its kind. One could call it an exchange of ‘negative’ services or goods. Such exchange is not to the mutual advantage but to the mutual disadvantage of the agents who engage in this sort of commerce. But the reciprocity of disadvantage is at the same time a ground for a reciprocity of advantage viz. the mutual advantage gained from not doing evil to one another. Thus respect for the neighbour's good can be to the mutual advantage of agents who are approximate equals by nature and therefore cannot effectively exercise normative authority over one another.

The principle called the Golden Rule is sometimes given the following formulation: Do to others what you want them to do to you and don't do to others what you do not want them to do to you.

When the principle is formulated in this way it is presupposed that one man regards as wanted and unwanted roughly the same things as those which any other man regards as wanted and unwanted respectively. This presupposition is largely fulfilled. But it is no logical necessity. There are exceptions to it. Something which I shun another man may like. Then it is not to be seen why the fact that I do not want it should be a motive for not acting in such a way that he gets this thing. Similarly it is not to be seen why I should do him something he does not want although I may want him to do this for me. But if we both want and shun similar things then we can exchange services to our mutual advantage by doing to one another what we welcome and not doing to one another what we shun.

If the Golden Rule is formulated in a way which is independent of the presupposition of similar wants it would run as follows: Do to others what they want you to do to them and don't do to others what they do not want you to do to them. The utilitarian defence of this principle would then be that its adoption by all agents is to their mutual advantage.

The rule consists of two parts. We could call them the positive and the negative parts. The positive part urges us to promote the negative part to respect the neighbour's good.

There is a noteworthy asymmetry between the parts. That the adoption of the negative part of the rule by all agents would be to their mutual advantage is a priori clear. Everybody would then gain the advantage of never being harmed by anybody else. That the adoption of the positive part of the rule by all agents must be to the mutual advantage of all is not in a similar way clear. Some men are more demanding on their neighbour's services than other men. If one man unlimitedly works for his neighbour there is simply no time for his neighbour to return the service. If promoting the good of one another shall be to mutual advantage there must be some ‘check’ on the demands which men have on their neighbours’ good services.

I shall not here discuss at length how this check is provided. I only suggest that it may be provided by the negative part of the rule. This would mean that a man must never demand of his neighbour greater services than those which the neighbour can render without detriment to his own good. By demanding more he may become responsible for harm to the neighbour and thus sin against the negative part of the Golden Rule.

These last considerations support the view that the negative part of the principle holds a more fundamental position than the positive part. I shall henceforth in this chapter discuss only the negative part i.e. the rule which urges us to respect our neighbour's good.

The big problem before us is this: How can respect for our neighbour's good become duty? In order to see how this is possible we must try to link considerations relating to mutual advantage with considerations relating to co-operation for a common good.

3. Consider a number of men ‘in a state of nature’. They occasionally do harm to one another; not often merely for the sake of harming more often perhaps as an act of revenge for some evil they have suffered from others usually however because pursuit of some end of theirs is causally incompatible with respect of the neighbour's good. Since some of them do harm some of them also suffer harm. Some of them may never do and often suffer others again often do and never suffer; but even those who escape harm from their neighbours may suffer under the inconveniences of having to take various self-protective measures. The more evil a man does the more reason he has to fear evil himself.

Suppose the men come to agree that each of them has to gain more from never being harmed by anybody else than from sometimes harming somebody else. I shall call this a basic inequality of goods.

The men who agree or subscribe to the basic inequality I shall speak of as constituting a community. Observance of the rule never to harm anybody (in the community) I shall speak of as the adoption of a practice. We can call it ‘the practice of not-harming’.

A member of the community who adopts the practice can be said to make a ‘personal sacrifice’. He gives up any such advantage which he may gain from doing harm to others. It is not certain that by giving up this advantage he will gain the greater advantage of never being harmed by anybody. But it is certain that if every member of the community relinquishes the smaller advantage of sometimes harming somebody else then they will all gain the greater advantage of never being harmed by anybody (in the community). Universal adoption of the practice is thus a necessary and sufficient condition for universal attainment of the greater good of the basic inequality. For being to the mutual advantage of all its members universal adoption of the practice is also a common good of the community.

Escaping harm from others I shall call each member's share in the community's greater good.

Universal adoption of the practice of not-harming can be called the price which the community has to pay for the greater good. Individual adoption of the practice i.e. not doing harm to others can be called each member's due or due share of the community's price for the greater good.

In order that the greater good of the community shall exist it is (logically) necessary that all members pay their due. But in order that an individual member shall get his share in the greater good of the community it is not (logically) necessary that he pays his due. To this end it is in fact not necessary that anybody pays. The only thing which is necessary is that nobody does harm to him. If however everybody else pays his due it is (logically) necessary that he will get his share. For that all the rest pay their due means that they adopt the practice of never harming anybody else in the community. If they do this the one remaining man can be sure of getting his share in the greater good independently of whether he pays his due or not. He can therefore try to add to the greater advantage of never being harmed by anybody also the smaller advantage of sometimes harming somebody—try to get the best of both worlds so to speak. If he adds to the greater advantage of getting his share also the smaller advantage of skipping his due he will necessarily deprive some of his neighbours of their share in the greater good. This I propose to call parasitic action.

The problem before us is to show how in spite of the possibility of parasitic action adoption of the practice which is conducive to the greater good of the community nevertheless may become a practical necessity.

4. Assume that a man harms one of his neighbours for the sake of some advantage i.e. end of his. The man who has suffered harm may wish to revenge himself. We need not here discuss whether revenge is or is not in origin associated with ends in acting or with primitive ideas of retributive justice. Revenge can in any case be said to have a natural self-protective purpose.

Let us suppose that people never joined hands with others to revenge a wrong unless they had been wronged themselves. Even then it is easy to see how threat of revenge may assume an accumulative deterrent effect on prospective evil-doers. Fear of revenge need not deter x from harming y and z individually. It may nevertheless deter him from harming them both. For if revenge on x becomes a common interest of both y and z they may co-operate and successfully revenge themselves. Even if x is stronger than each of them individually it is unlikely that he will be stronger than both of them jointly. This is a consequence of the fact that men are approximate equals by nature and that they can co-operate for common ends. Fear of revenge may therefore work as an effective deterrent at least against evil-doing on a larger scale i.e. involving several wronged agents. It may thus go at least some way towards making observance of the practice of not—harming a practical necessity.

I shall call evil caused by revenge for evil done the natural punishment of evil-doing. It can also be called the natural sanction attached to the practice of respecting one's neighbour's good.

The step from these logical preforms of law to the establishment of normative relationships is not difficult to imagine. Let there be three men x y and z who constitute a community by subscribing to our basic inequality of goods (see sect. 3). x and y are both anxious that z shall pay his due for otherwise they will not get their share in the greater good of the community. x and y can command z being jointly stronger. If commanding z turns out to be necessary for making him pay co-operation in commanding becomes autonomously incumbent on x and y. Thus x and y impose upon z the (heteronomous other-regarding) duty of paying.

Similarly x and z impose the same duty on y and y and z on x. Thus the very same interest which makes each one of the three co-operate with one of the other two against the third viz. anxiety to get one's share in a greater good which benefits them all becomes the power which forces each of them to pay their due. We can imagine the men as thoroughly selfish void of any sense of justice or morality whatsoever. They have not the slightest desire to pay their due for their share. But the greedier they are on their share the stronger will the normative pressure become under which they are themselves to pay their due. This is a fascinating mechanism. In cooperating for the common end of imposing heteronomous other-regarding duties on others men come to get these same duties heteronomously imposed upon themselves.

The considerations which we have conducted show how men's self-interested pursuit of a common good may engender a practical necessity of adopting a practice which is to the mutual advantage of them all.

5. Fear of revenge or of the punishing arm of heteronomous law is of course far from being the only motive which may make men adopt the practice of not doing harm to their neighbours. The circumstances under which there is an advantage to be gained from harming others are special circumstances. In the life of some men they do perhaps not often arise. Observance of the practice of not-harming because of lack of motives or opportunity of harming is of course not more to the moral credit of men than observance from a motive of self-interest.

The reasons why a man wants to respect a particular man's good may be some particular relationship in which the two stand to one another. Depending upon the nature of this relationship the reasons can be self-interested or altruistic. Perhaps he expects some good in return from that particular man. Or perhaps he just loves him and therefore is anxious to respect that other man's good ‘for its own sake’. (I am here thinking of that aspect of love which consists in an unselfish regard for the good of another being.)

Could unselfish regard for the good of all men. be a reason why a man never does evil to his neighbour? It must not be considered self-evident that since a man can love some or other of his neighbours he can love them all. Love of man is different in kind from love which is a relationship between particular men. One could call the second with Kant pathological love. ‘Pathological’ of course does not here mean ‘sickly’. It refers to the foundation of this kind of love in a peculiar sentiment which one human being feels for another. We cannot exclude the possibility that a man felt this sentiment for every other man whom he happens to meet on his life's journey. His love of men would yet be different from that love of man which independently of relationships among individuals makes a man respect his neighbour's good ‘for its own sake’ or treat his neighbour ‘as an end in itself’.

In order to get a firmer grasp of this ‘abstract’ love of man let us go back to our problem of the greater good of the community and the individual's share and due.

We have seen how wanting one's share can engender a practical necessity of paying one's due. But we also know that paying one's due is not as such a necessary means to getting one's share as an end. It becomes a necessity only thanks to the intervention of heteronomous norm-pressure one could say.

Suppose a man refuses to receive the smaller of two advantages if he can have it only at the expense of another man losing the greater advantage. He gives as a reason for his refusal that his receiving the advantage would hit his neighbour harder than it would benefit himself. Is such comparison of goods possible? We have touched upon the problem before. We then suggested that a balancing of goods against one another is possible only from the point of view of one and the same valuating being. Therefore if a man thinks that an action of his injures another man more than it benefits himself and on that ground abstains from this act then he respects the good of that other man as though it were his own. This peculiar respect for the good of another being as one's own is a motive which without the Intervention of heteronomous norm-pressure will put a man under an autonomous necessity to pay his due for his share in the greater good of a community of which he is a member.

It would be quite wrong to call this attitude to one's fellow humans self-interested. It would be misleading if not wrong to give to it the name of self-love. But it could be called a man's love of himself in his neighbour. This is the ‘abstract’ love of man which I was contrasting with the ‘pathological’ love of one man for another.

6. I have tried to show under what conditions there could exist a duty to abstain from evil-doing. We must now stop to reflect on the nature of the argument which we have been conducting.

That something is duty we have said means that it is a practical necessity with a view to the welfare of some being or beings. To ask under which conditions something or other is duty entails asking what the wants of men must be in order that something or other shall be a practical necessity.

It is not off-hand clear which wants of men would necessitate abstention from evil-doing. Mere wanting to escape harm from others cannot affect this—for one thing because men cannot as individuals command one another to reciprocal respect of one another's good. Wanting not to harm others would not make abstention from evil duty—wanting the very thing is not practical necessity of doing the very thing. These considerations show that our problem is not trivial.

The method which we employed for its solution was a general method. The derivation of the duty to abstain from evil was only an illustration of how the method works in a particular case. We shall now describe the method in abstraction from this particular application of it.

We started from the assumption of a community of men who agree about that which we called a basic inequality of goods. A certain good G₁ is for them (individually) greater than a certain other good G₂. This basic inequality is subject to the following three conditions:

• (i) If all men in the community relinquished the smaller good G₂ then all of them will get the greater good G₁.
• (ii) It is logically possible for a man to get both goods the greater and the smaller one for himself.
• (iii) It is logically necessary that if one man gets both goods for himself then some other man will lose the greater good.

The existence of the greater good for one man we called his share in the greater good of the community. Giving up the smaller good by one man we called his paying his due. In this terminology the first condition means that the greater good of the community will exist if each member pays his due. The second means that it is possible to get one's share without paying one's due. The third means that if a man gets his share without paying his due another man will lose his share.

To get both goods for oneself or which means the same to have one's share without paying one's due can be regarded as a basic form of injustice. The idea of justice has many roots and has accordingly assumed many conceptual forms. We shall not here discuss the varieties of justice. The only form of it with which we are concerned is embodied in the following principle: No man shall have his share in the greater good of a community of which he is a member without paying his due. (It is essential that ‘greater good’ ‘share’ and ‘due’ are defined in such a way that having one's share without paying one's due logically entails that some other member of the community loses his share in the good.)

This Principle of Justice I would regard as the cornerstone of the idea of morality.

The Principle of Justice has as many applications as there are communities of men for whom a basic inequality of goods satisfying our three conditions holds true. I shall call any such community a moral community of men and a duty to act in accordance with the Principle of Justice a moral duty. Thus e.g. in the community of men who have more to gain from never being harmed by anybody else (in the community) than from sometimes harming somebody else (in the community) a duty to abstain from evil-doing would be a moral duty. This restriction of a moral duty to a moral community we must discuss presently (see sect. 7).

The problem is now: How can action in accordance with the Principle of Justice become duty in a moral community?

This can happen in two ways. We can call them the outer and the inner way. The first imposes the moral duty as a heteronomous other-regarding duty on the members. The second imposes it as an autonomous other-regarding duty.

Self-interest is sufficient as a moving force behind heteronomous moral duties. This is so because men's anxiety to get their share in a greater good of the community can engender a practical necessity for them to pay their due. The logical mechanism works as follows:

Wanting one's share in the greater good of the community is a motive for joining hands with other members to enforce a duty to pay their due on such members who might otherwise be unwilling. It is not a motive why one should pay one's due oneself. But fear of revenge or of punishment for breaking the very same laws which are upheld by one's anxiety to make others pay is such a motive. We can imagine a community in which everyone is anxious to add the smaller of the two goods of the basic inequality to his share in the greater good but is prevented from doing so by his even greater anxiety to prevent anybody else from trying the same unjust trick lest be should become the victim. It should namely be observed that from subscribing to the basic inequality of goods it follows that one's anxiety to get one's share is greater than one's anxiety should one feel it at all to profit from injustice. This is the very ‘point’ of the inequality: it provides a check on a man's desire to treat others unjustly in the form of his even greater desire to be treated justly by others. We can in other words make us a picture of a society in which justice and morality are ‘kept going’—even perfectly—through mere self-interest. It is essential that such a society could exist. In it men would observe their moral duties—but not from what we would call moral motives.

I shall say that an agent acts from a moral motive when he observes his moral duty neither from a self-interested motive such as fear of revenge or punishment nor from an altruistic motive such as love or respect for the neighbour whose welfare might become affected through his action but from a will to secure for all the greater good which similar action on the part of his neighbours would secure for him. Action prompted by a moral motive is thus beyond both egoism and altruism. The moral will is in a characteristic sense a disinterested and impartial will to justice. Its impartiality however consists in treating your neighbour as though his welfare were yours and your welfare his. Therefore the moral will is also a love of your neighbour as yourself.

When action takes place from a moral motive observance of the Principle of Justice has become autonomous other-regarding duty.

The question may be raised: Are moral duties in the sense defined ever heteronomously imposed? Or is the possibility ‘purely theoretical’?

The answer is that the possibility is not ‘purely theoretical’. It is an important aspect of the working of the huge normative power which we call the state that it serves the interests also of moral communities of men. It would be a great misunderstanding to regard the laws of the state as essentially void of a moral content. The very example of a moral duty which we have been discussing viz. the duty to abstain from evil is probably the most important single concern of the laws of all states. What may nevertheless make us think of the law of the state as essentially different from ‘the moral law’ is probably not so much the fact that there are branches of law which are not immediately concerned with matters of public or private welfare as the fact that although laws of the state need not be void of a moral content the power of the state as an authority of norms is founded on self-interested co-operation of men for common ends and not on a disinterested will to observe the Principle of Justice.

Since self-interest can be the safeguard of morality why should a man ever act from a moral motive? To raise this question is to express astonishment at the fact that men should pursue other than self-interested ultimate ends. But even if one cannot give reasons why men should act morally from moral motives one can try to make a man respect the good of another as if it were his own by the use of arguments which look like an appeal to ends. One can ask questions like this ‘What right have you got to put yourself in a privileged position? If you get your share without paying your due then somebody else who is equally anxious to get his share will necessarily be without it. Don't you see that this is unfair.’ One could almost call this appeal to a man's sense of justice an appeal to a man's sense of symmetry. ‘If my wants are satisfied at the expense of another man's then why not his wants at my expense?’ This is like saying: ‘For symmetry's sake you must want to be just. And for justice's sake you must make yourself pay your due for your share in the good of the community.’

Such argumentation as this need not be without effect upon men. To resort to it is not cheating. But it is not an appeal to further ends for the sake of which we should aim at acting justly. It is making the nature of justice as an end clearer by making us see logical relationships between concepts.

That which is here called action from a moral motive has a certain resemblance to that which Kant calls action from a motive of duty (Handeln aus Pflicht as distinct from pftichtmässiges Hanaeln). The resembling features are those of disinterestedness and impartiality and the detachment of the moral will both from self-interested concern and from altruistic sentiment.

The moral will is a will to do to others something which we want others to do to us. It is not however a will to do this because we count upon or demand a return of service. It is a will to do this because we think it but fair that we too should contribute to the agreed good of all. One could say that the moral will is a will to observe the principle known as the Golden Rule from a motive which comes at least very close to that which is a Christian love of our neighbour. But one could also call it a will to treat our fellow humans as ends in themselves. Here is another resembling feature with the ethics of Kant.

7. Someone may wish to raise the following objection against the argument which we have been conducting: This argument makes the existence of moral duties depend upon the wants of men. It may not be futile or uninteresting to try to describe the peculiar nature of the wants behind such duties. What men want however is contingent. Is it not a most unsatisfactory position in moral philosophy to let questions whether something or other is moral duty depend upon such contingencies?

There are moreover two ways in which moral duties on our understanding of the term depend upon the wants of men. The first is that moral duties exist only within moral communities defined in terms of that which I have called a basic inequality of goods. The second is that even within the moral community the moral duty exists only to the extent that it is either autonomously imposed by the member upon himself or heteronomously imposed upon him by others.

The restriction of moral duties to members of a moral community may seem very narrowing. This impression however is deceptive. Let us go back to the basic inequality in the example which we discussed and consider how big the moral community might be which it determines.

The moral community under consideration consists of all those who have more to gain from never being harmed by anybody else than from sometimes harming somebody else.

Some men may stubbornly refuse to acknowledge that they have more to gain from never being harmed than from sometimes harming and yet be mistaken. But they may also be right. In order to see under which conditions they would be right we must inspect more closely the nature of the goods which are balanced against one another in the inequality.

It is conceivable that a man never suffers any harm from his neighbours all through his life. It is also conceivable that such a man has done harm to some other man for the sake of attaining some end and that he attained his wanted end and never regretted it. This man has then gained an advantage from harming his neighbour without having suffered a corresponding loss in the form of revenge or punishment. Since our assumption was that he never suffers harm from anybody else there is no such harm which he could balance against this advantage so that he might say to himself: ‘I should have rather been without that which I gained from harming my neighbours than have had this harm done to me.’

If these are the assumptions does it follow that the basic inequality does not hold good for our man? This does not follow. All that follows is that this man actually gained more from harming others than he lost from being harmed by others. But this possibility is not denied in the basic inequality. It says that a man has more to gain from not being harmed than from harming. Therefore in order to decide whether the basic inequality holds true for our man we must also consider how much he has gained from escaping harm. How shall this quantity be estimated?

The harm which on our assumption the man has escaped is the harm which would have resulted to him had his neighbours never hesitated to wrong him when this could have been to their advantage. Escaping this harm is the advantage which a man has to gain from never being harmed by anybody.

The advantage against which this has to be put in the scale is not the advantage which on our assumption the man actually gained from doing harm to somebody. It is an advantage greater than that viz. the advantage which would have resulted had he never hesitated to wrong his neighbour for the sake of anything which was coveted by him. This is what a man can be said to have to gain from sometimes harming others. Unless our assumption is that the man is very ruthless we shall have to admit that he probably did not gain the full measure of what he had to gain from doing evil.

In considering whether the basic inequality in our example is true for a man or not we have thus to balance against one another the following two—negative and positive goods: On the one hand the loss which a man can be expected to suffer if others take as much advantage from harming him as they can. On the other hand the gain which a man can be expected to harvest if be takes as much advantage as he can from harming others.

Having thus made clear the nature of the goods which are balanced against one another we go back to the question whether the basic inequality is true for all men or not. The question is in other words whether the moral community which the inequality defines will comprise the whole of mankind.

To this question is relevant that which we have said before about the approximate natural equality of men and about the possibility of overruling natural inequalities by co-operation.

If there existed a giant among men who could treat the rest of mankind as insignificant worms who can do nothing to harm him then this man would not belong to the moral community as determined by the basic inequality in our example. Men sometimes imagine that they are such giants—metaphorically speaking. They are usually soon taught that they are not. Also when the lesson is ineffective the truth may nevertheless be that they too have more to earn from escaping harm which could befall them than from harming others.

Yet it is hardly an a priori necessity that an individual man should be weaker than all other men jointly. (I say ‘hardly’ since we may wish to deny the name of ‘man’ to a being of superhuman powers.) Suppose a man invents some fearful weapon to which he alone has access and by means of which he can wipe out any number of men who withstand his wishes or encroach upon his privacy. He could be a kindly man and never want to do harm to anybody. Could he be just and moral? I shall not attempt to answer the question. As long as he keeps his secret weapon and is aware of the superiority which it confers upon him in relation to the rest of humanity he in any case does not belong to the moral community which is determined by the basic inequality in our example.

The fiction of the superman although logically possible is yet highly unrealistic someone may say. Is it highly unrealistic? I would ask. Substitute in our example a team of men for an individual—and the example is less unrealistic. Substitute a state—and we shall recognize it as thoroughly realistic. The basic inequality from which we derived the moral duty to respect the neighbour's good holds true tot practically all individual men. It is therefore even if not logically necessary yet practically certain that the moral community which it defines will comprise the whole commonwealth of men. This is a consequence of men's natural equality and powers of co-operation. On the level of relationships between states the same basic inequality is not undisputably true. Therefore it is not practically certain that the moral community which it defines will comprise the whole commonwealth of nations. For although states can co-operate they are not individually approximate equals. For this reason too the problems of justice and morality in state-relationships are to a great extent logically different from the problems of justice and morality in the relations of individual men.

I hope to have succeeded in showing that the restriction of the moral duty which we derived to a community of men is of no practical importance. For it is a practical certainty that no man can put himself outside the community as we have defined it.

But what about the imagined superman who is not in the community? Is it not his moral duty too never to do evil to his neighbour? Must it not be everybody's duty? I would answer No. The superman cannot be commanded to respect his neighbour i.e. fear of punishment is no motive for him to obey such a ‘command’. Moreover he cannot even feel that love of man which puts him under the autonomous law of respecting all other men. For this love ‘of thy neighbour as thyself’ is conditioned by the similarity of needs and wants and powers of men. If a being stronger than all men together shows benevolent concern for the welfare of humans this would be more like an act of mercy than an act of Justice.

8. In conclusion I wanted to raise the following question: In which sense if any can justice and observance of moral duty be said to possess a ‘utilitarian justification’?

The Principle of Justice as we formulated it says that no man must have his share in the common good of a community of which he is a member without paying his due. Assent to this principle autonomously imposes a moral duty on the members of the community to adopt the practice of paying. The same duty may also become heteronomously imposed.

Observance of a moral duty on this view is a kind of ‘public utility’. This means: To do one's moral duty is a necessary condition for the existence of a common good of a community of which the dutybound agent is a member. In this sense justice and the observance of moral duty trivially has a utilitarian foundation.

From this does not follow however that the man who acts as justice and moral duty demand will fare better than the man who acts unjustly and neglects his moral duties. If all men in the community act justly they will all get a profit from their good conduct which is necessarily—this follows from membership in the community—greater than any profit they could have gained from acting unjustly. But as we have seen it is also possible for a man to add to the profits of justice the profits of injustice—at the expense of another man's loss of the first profit. This is a fearful possibility and its occurrence cannot be ruled out by logical argument. Of course to try to draw the extra profits of injustice and retain the profits of justice is an extremely difficult game and It is probably right to say that few are successful at playing it. It is even logically impossible that all men could thus profit from both justice and injustice. (This can be deduced from the conditions which the basic inequalities have to satisfy.) But some can theoretically.

The possibility of adding to the blessings of the reign of justice the profits of unjust action in short: the possibility of that which we have called parasitic action constitutes an important sense in which justice and morality can be said to be essentially void of a utilitarian justification.

Justice and morality are of public utility in a sense in which for reasons of logic injustice and immorality could never be. But injustice and immorality can be of private utility. If any attempt to increase one's well-being by immoral means would necessarily be doomed to be unprofitable then moral action would be what it is not viz. a practical necessity under considerations pertaining to the agent's own welfare alone. Moral action would be autonomous self-regarding duty.

There have been many attempts to ‘reduce’ the moral imperative to self-regarding duty. This attempted reduction is I think one of the roots of the idea that there will be a last judgment when the unjust shall suffer and the just triumph.

I fear I cannot share this optimism—if it be called by that name—in the ultimate triumph of justice. Injustice may as a matter of fact become revenged. Or it may become punished. But from the standpoint of a secular morality it must remain a contingent matter whether the unjust man will prosper or perish.