I. Two Questions

Suppose you had the power to impose on a society one of two patterns. In the first, good and bad things are distributed randomly across the population. In the second, there are exactly the same good and bad things, but good people have the good things and bad people the bad ones. Let us call the first pattern “random,” and the second “ordered.” My first question is: Which pattern would you impose?¹

Sophisticated people will quibble. They will say that my first question is not well formed because it ignores complexities. They may say, for instance, that human beings could not have only good or bad things in their lives because good and bad things are unavoidably mixed. Success is good but it often depends on painful effort, which is bad. Courage is good but it can be shown only in the face of danger, which is bad. Deep love of someone is good but it gives a hostage to fortune, which is bad. All this is true and pedantic. Understand the first question as asking about the preferred distribution of the most favorable good-to-bad ratio.

Another quibble is that my first question assumes that there is agreement about what things are good or bad, but there is no such agreement. If you impose on a society an ordered pattern, you in fact impose your preferences on other people, and a decent person would not do that. Suppose, then, that you are forced to choose because if you do not, you will be sentenced to live in a society that conforms to the pattern you would not choose. Moreover, it is nonsense to say that there is no agreement about good and bad things. There certainly are disagreements, but there is also much agreement. Everybody agrees that in normal circumstances it is better to satisfy one's basic physiological needs than to frustrate them; better to be respected than humiliated; better to feel secure than terrified; better to have an interesting than a soul-destroying job; better to like than loathe oneself; and so on and on. Understand the first question, therefore, as being about the preferred pattern of distribution of those good and bad things about which there is agreement in a particular society. If you are tempted to invent extraordinary cases in which normal preferences do not hold, resist it because such cases are irrelevant. The first question is about the pattern preferred in normal circumstances, not in a lifeboat, or in the Garden of Eden, or on your deathbed. And remember also that your choice of pattern is for a society in which you will not live, so self-interest is not involved in your choice.

Yet another quibble might be to object to asking hypothetical questions. It might be said that such questions postulate unusual situations and the reasonable answer to them is that one does not know what one would do in unusual situations. I have a great deal of sympathy for this. Nevertheless, I persevere in asking the first question because I fail to see why conservatives, such as myself, should not be entitled to it when liberals endlessly discuss hypothetical questions about the original position, the veil of ignorance, ideal speech communities, the prisoner's dilemma, the consent of totally reasonable people, deserted islands with miraculously available resources awaiting distribution, compulsory insurance schemes, and the economic man, not to mention the state of nature or the social contract.

I submit, then, that the first question is legitimate and that virtually everyone would answer it by preferring the ordered to the random society, and thus the second question naturally arises of what is behind this general preference. I can now confess that the first question does not interest me very much. I asked it merely to permit me to ask the second question. And I want not merely to ask but also to answer it: A society with the ordered pattern (from now on “ordered society”) is preferable because people in it get what they deserve and do not get what they do not deserve. The trouble with the society in which the random pattern holds (from now on “random society”) is that people in it have no reason to believe that their actions will lead to their goals. Hard work, intelligent choice, and self-discipline, for instance, have exactly the same chance of success as sloth, stupidity, and self-indulgence. The conservative view I aim to defend is that the ordered society is just because its arrangements increase the likelihood that good actions will be successful and bad actions will not. To climb even further out on a limb, I also claim that reasonable people are, in fact, committed to the conservative view, if only they would take the trouble to think through what they believe.

II. Why Should People Get What They Deserve?

Consider a simple sequence leading to an action. I am hungry, want to eat, decide to go home for lunch, and I walk there. Implicit in the sequence is that I have a set of beliefs (my house is nearby, there is food in the fridge, there are no obstacles in the way, and so on) and I have the required capacities (I can walk, rely on my memory, estimate distance, and the like). The sequence that precedes action, then, has the following elements: motive, belief, capacity, goal, and decision. Each may be defective: motives may be irrational, beliefs false, capacities inadequate, goals unattainable, and decisions wrongheaded. Suppose, however, that they are not defective and I perform the action to which the conjunction of these elements leads. I expect, then, that my action will be successful, that it will achieve its goal. In simple sequences of this sort, normal people in normal circumstances have normally good reason to suppose that the elements are free of defect and no reason to doubt it. Their expectation that their actions will be successful is generally reasonable. The well-being of individuals requires that such reasonable expectations be generally met.

It would be comforting if simple sequences were adequate for the business of living, but they are not. Complexities unavoidably arise because it is often difficult to tell whether an element is defective. Motives conflict and we constantly have to make choices about which of them we should act on. We routinely have to evaluate the reliability of our beliefs on the basis of imperfect knowledge and insufficient evidence. Our capacities are always limited, and we have to estimate whether they are adequate for the achievement of difficult goals. We usually have several incompatible goals, and we must decide about their respective importance. But what seems important at one time may become less so at another. The world changes, we change, and we have to make guesses about how these changes might affect our goals. The choices we make depend on understanding our circumstances and what we want, all things considered, out of life, and both are unclear at the time we have to choose, and it is even more unclear how they might be in the future. Coping with complexities requires, therefore, in addition to getting the facts right, also judging the significance of such facts as we have available. Good judgment about such matters is difficult.

Suppose I act on the basis of a complex sequence and my judgment is good. Might I, then, reasonably expect that my action will be successful? No, because even if I act as reasonably as anyone in my position could, I may nevertheless fail because others justifiably prevent me from succeeding. Most of the time my success depends on the cooperation of others, and they may have more important concerns; or they may also want what I want and may be better at getting it; or what I want may run counter to their interests or to the interests of an institution, cause, or collectivity they wish to protect. More is needed, therefore, before I can reasonably expect to get what I want, and the same, of course, is true of everyone else. We all have to take into account that we live together in a society and depend on the cooperation of others. The terms of cooperation, therefore, have to be set. In the vast majority of complex sequences, we can get what we want only on the terms our society sets. These terms, of course, may be more or less adequate to the purpose of maintaining the optimum conditions in which as many of us as possible can go about getting what we want.

Can people in general reasonably expect that their actions will be successful if they are based on good judgment and conform to the terms set by their society? The answer is still no, for two reasons. One is that the prevailing terms of cooperation may be defective. I shall shortly say more about this, but let us assume for the moment that the terms are adequate because they set optimal conditions. The other reason why the expectation of success may be premature is that not even adequate terms of cooperation can eliminate competition. People may fail to achieve their goals, although they have good judgment and conform to the terms, because they lose out in a competition with others whose judgment is also good and who also conform to the terms. Only one person can win the race, get the job, be the first to make the discovery, and only a few can make the best-seller list, get elected, add to the canon, or make a lasting contribution to science. For each of those who succeed, there are many who try and fail.²

Putting all this together, the following requirements of successful action emerge: having good judgment; conforming to adequate terms of cooperation set by one's society; and, if goals are competitive, prevailing in the competition. The expectation of success is reasonable if these requirements are met. This finally allows me to make clear the point of the preceding discussion. What I mean by saying that people should have what they deserve is that they should succeed if their expectation of success is reasonable. And what I mean by saying that people should not have what they do not deserve is that they should not succeed if their expectation of success is unreasonable. The “should” and the “should not” above express requirements of practical reason.

Practical reason is the use of reason for making human life better. One way of doing that is to make it more likely that good people have good, rather than bad, things in life and that bad people have bad, rather than good, things in life. This is what happens in an ordered society. The reason for making a society ordered is to increase the likelihood that good judgment, conformity to adequate terms of cooperation, and prevailing in competition will lead to better lives. These are means derived from the history of human trials and errors for the betterment of the human condition. The conservative view I am defending is that justice prevails if these requirements of practical reason are met. This is at once the aim and the justification of justice.

People living together in a society should, therefore, do what they jointly can to make justice prevail, but what joint efforts can achieve is limited. Individuals alone can make their own judgments good, and they alone can make themselves succeed in a competition if the prevailing terms are adequate. What can be done by joint efforts, however, is to make adequate terms prevail, which is our next topic.

III. Terms of Cooperation

There are countless terms of cooperation, and it would be futile and tedious to try to list all of them. I shall proceed instead by calling attention to three general types that a society can hardly do without, and then go on to consider what these terms imply. There are, of course, others. These terms are conventional. They embody the habits, customs, practices, and the more or less clearly formulated rules that have emerged in the course of a particular society's history. The terms change, but usually slowly. There goes with them a continuum of evaluations that range between the extremes of good and bad, and allow many intermediate judgments in between. The resulting evaluations praise or condemn whatever appropriately falls in the domain of a general type of term. They imply responsibilities that people are praised for discharging or condemned for violating, or from which, under specifiable conditions, they are exempted or excused. They imply also the reasonable expectations of others that these responsibilities will be met. And they serve as the basis on which good and bad things are distributed in a particular society. In short, the terms of cooperation are conventions regulating what people should or should not have or do in particular areas of life.

One of these terms concerns relationships. Lovers, competitors, friends, colleagues, parents and children, teachers and students, judges and defendants, physicians and patients, merchants and customers are connected by conventional ties. These ties create responsibilities and expectations about how people related in these ways should treat one another. The first four of these relationships are symmetrical. Lovers, competitors, friends, and colleagues typically have the same reciprocal responsibilities and expectations. The next five relationships are asymmetrical because one party has authority over the other, or provides a service the other needs, or sells something the other wants to buy.

A second term of cooperation has to do with agreements, such as contracts, promises, loans, memberships in organizations, employment, political or legal representation of others, or holding a license to drive a car, sell a product, or build a house. Some of these agreements are formal. The responsibilities and expectations are governed by written rules and are legally binding. Others are informal. What is owed or expected rests on a tacit understanding of the parties. The force behind such informal agreements is a shared sense of trust and mutual good will.

A third term is connected with actions that affect the security of others. A society cannot endure unless it protects the security of those living in it, but different societies have different conventions about how far the protection of security should go, what violations are permissible, excusable, or prohibited, how people outside of the society should be treated, what counts as cruelty, negligence, accidental injury, and, of course, how security should be understood. Take, for instance, homicide. All societies must have conventions about it; but conventions regarding when homicide is murder and when it is excusable or justifiable vary greatly, as do conventions regarding abortion, suicide, revenge, euthanasia, capital punishment, infanticide, and the like.

Part of the importance of these terms of cooperation, and of many others I have not discussed, is that they establish what people in a particular society deserve. They deserve to have what the terms of their relationships, agreements, and actions establish, and they do not deserve to have what would violate those terms. Thus, the terms of cooperation provide reasons for the responsibilities and the expectations that people living together in a society have. Children deserve a decent upbringing from their parents because that is part of how parenthood is understood by us. Incompetent physicians deserve to lose their licenses because a physician is assumed by us to be able to treat illness well. Murderers deserve to be punished because that is how we go about protecting security. These terms are not merely conventions that happen to hold in our society. Our well-being requires that children be brought up well, that illness be treated, and that life be protected. It is conventional how we do these necessary things, but it is not conventional that a human society must do them in one way or another. If they were not done in a society, the well-being of those living in it would be badly served and the society would disintegrate. When we say, therefore, that people deserve this or that, what we are saying is that this is our particular way of doing what is necessary for our well-being.³

Claims about what people do or do not deserve may be mistaken of course. I shall consider two ways in which this may happen. One is that the prevailing terms of cooperation are misapplied; the other is that they are faulty. I discuss misapplication now and faulty terms in the next section. Any term may be misapplied because those who apply it get the relevant facts wrong. If a person is mistaken for a murderer, he does not deserve to be punished. If a student passes a course by cheating, he does not deserve credit for it. If a pretended friend is an enemy, he does not deserve my loyalty. Even if the facts are as assumed, the terms may be misapplied by getting wrong the proportion of what is deserved. A novel may deserve good reviews, but not the National Book Award. A thief may deserve a term of imprisonment, but not to have his arm cut off. A physician may deserve payment for services rendered, but not the deed to my house. Mistakes of fact and proportion lead to people's getting good or bad things they do not deserve at all or to their getting more or less than they actually deserve.

A less obviously detectable mistake is to be wrong about the goodness or badness of particular relationships, agreements, or actions. Because what is taken to be good or bad is in fact not, people are rewarded or punished when they do not deserve it. Those who suppose that telling a lie is always wrong, that contracts between employers and employees are always exploitative, or that suicide is murder are wrong because they suppose that what may be morally excusable is always bad. Similarly mistaken are the suppositions that frugality is a sign of virtue, that flogging children is good for their character, and that belief in God makes people virtuous.

Terms of cooperation are associated with a wide array of relationships, agreements, or actions which are judged by a society to be good or bad. These terms are not invalidated if some few of the judgments conventionally made within them turn out to be mistaken about what is good or bad. The mistakes call for the correction of some conventional beliefs and practices, but they leave the large remaining part of that term intact. It may happen, of course, that not merely a few but most applications of the term are mistaken. Then the term must be regarded as faulty in its entirety, and it should be abandoned. This is just what happened, or should have happened, to magic, astrology, and human sacrifice.

Another kind of mistake, much closer to home, is to suppose that there is no special reason needed for the distribution of good and bad things. One way of making this mistake is to suppose that everyone is equally deserving of the good things in life and that the bad things are burdens that we all have to bear equally. To appreciate the force of this familiar but nevertheless mistaken claim, it should be remembered that the distribution of good and bad things we are discussing includes only those about which there is agreement in a society. Opinion is divided on whether owning a cat is good or bad. But there can be no reasonable disagreement about it being good to satisfy one's basic physiological needs. The mistaken claim, then, is that everyone in a society is equally deserving of those good things that everyone there regards as good and everyone is obliged to carry a fair share of the generally agreed upon onerous burdens. If asked why good and bad things are equally deserved, the usual answer is that everyone has the same basic moral worth, or that everyone is entitled to the same basic respect, or that everyone has the same basic freedom and welfare rights.⁴

This is one of the passionately held shibboleths of our egalitarian age. It comes packaged in a self-righteous, moralistic rhetoric that maligns doubt as indecent.⁵

The familiar indignant reply is that they believe no such things. What they believe is that there is a presumption in favor of everyone being equally deserving. The presumption can be defeated, and it is defeated in the case of terrorists, felons, and other malefactors. But this presumption flies in the face of facts whose denial is absurd. Human beings differ in their characters, circumstances, talents and weaknesses, capacities and incapacities, and virtues and vices; in their moral standing, political views, religious convictions, aesthetic preferences, and personal projects; in how reasonable or unreasonable they are, how well or badly they develop their native endowments, how much they benefit or harm their society, and how hardworking or disciplined they were in the past and are likely to be in the future; and they differ also in the relationships they have, the agreements they have made, and the effects of their actions on the security of others. In view of these and other differences, surely the reasonable presumption is that there is an overwhelming likelihood that different people will deserve different things.

The implausibility of the egalitarian presumption is compounded if it is asked who should distribute what people deserve. Should parents regard their own and other people's children as equally deserving of the benefits they can bestow on them? Should married couples treat each other and strangers as equally deserving? If, by way of avoiding such absurdities, it is said that the egalitarian presumption is meant to guide the government, not individuals, then no less damaging absurdities follow. A government betrays its most elementary responsibility if it proceeds on the presumption that citizens and foreigners, law-abiding and criminal citizens, defenders and subverters of the society are equally deserving. The questions grow in number and urgency when it is asked, as it must be, what reasons might be good enough to count against the egalitarian presumption. If differences in morality, reasonability, law-abidingness, and citizenship count, then virtually nothing remains of the presumption since the great and undeniable differences among people in these respects constantly defeat the presumption. And if these differences are said not to count, then what would justify the government in ignoring them in the distribution of good and bad things at its disposal?

The conservative view is that the time has come to consign this egalitarian shibboleth to the dust heap of history, where it will join other passionately held follies of other ages, such as the divine rights of kings, classless society, possession by the devil, planned economy, damnation outside the church, and an idyllic prehistoric society from which civilization has caused us to fall farther and farther. What should take the place of this absurd belief about what people deserve is the recognition that different people deserve different things and any claim that people deserve this or that must be backed by reasons. I have been arguing in this section that the reasons must be derived from the terms of cooperation that prevail in a society. The reasons are good if they are free of mistakes in fact, proportion, the identification of good and bad things, and in missing the central importance of individual differences. It must be recognized, however, that even if good reasons of this kind are given in support of claims about what individuals deserve, the claims may still fail because the terms of cooperation that prevail in a society and from which these reasons are derived may be faulty.

IV. The Justification of Terms of Cooperation

An often heard objection to conservatism is that it is committed to the mindless perpetuation of the conventions that happen to prevail in a particular society. There may have been conservatives who actually held this commitment, but, if they did, they should not have. What conservatives should be committed to is the perpetuation of conventions that have endured for a considerable length of time, measured in decades or more rather than months, because people have voluntarily adhered to them—and have been right to do so because the conventions have made their lives better.⁶

Terms of cooperation are conventions, and their justification depends on whether they have stood the test of time. There are deplorable, coercive, unfair, exploitative, ignorant, and stupid conventions, and they fail the test of time. Since the conservative view is that justice consists in people having what they deserve and not having what they do not deserve, and since terms of cooperation that fail the test of time fail to distribute good and bad things according to what people deserve, conservatives are committed to abandoning or reforming unjust terms of cooperation and to perpetuating just ones if, and only if, they have stood the test of time.

One crucial task conservatives have in justifying terms of cooperation, therefore, is to provide reasons for thinking that adherence to them contributes to human well-being. These reasons emerge from reflection on life in a good society, by which I understand a society that protects the conditions in which people can make good lives for themselves. Or, to put it in another way, if individual lives go badly in a good society, it is not because of adverse social conditions but because of personal failure or misfortune. Good societies are cohesive and enduring partly because their members largely agree about what they owe one another, what sorts of lives are good or bad, and what are many of the good and bad things that tend to make lives good or bad. Their agreement rests on shared values, and the prevailing conventions embody them. These values have endured and members of the society express in their terms what they want, and the conventions govern permissible and impermissible ways of trying to get it. Individuals make sense of their lives partly in the conventional framework of their society.

One of the benefits a good society provides is order. A good society makes it possible for its members to have reasonable expectations about each other's conduct. These expectations rest on the justified assumption that fellow members are most of the time guided by the same conventions. Thus, they know what counts as kind or cruel, routine or exceptional, appropriate or inappropriate in countless different contexts, such as marriage, child-raising, employment, competition, joking, electioneering, quarrelling, and so on. This knowledge comes from a shared moral education, which is the usually informal initiation into the prevailing conventions. As a result, members of the society can count on each other because there is a social bond between them. They recognize each other as having the same allegiances. Their individual identity is partly defined by their shared conventions, including the shared terms of cooperation.

If you doubt that this picture fits American society, consider this random list of our shared conventions to assuage your doubt. Parents are responsible for their children; breaking a promise requires excuse; politicians speaking in their public capacity should not lie; winners must not gloat over losers; people ought not brag about their talents; a life of idleness is wrong; following one's conscience is right; eating people is wrong; one must not torture animals; parents should not have sexual intercourse with their children; disagreements should not be settled by violence; people ought to be allowed to express unpopular views; it is wrong to spread malicious lies about rivals; permission must be asked before borrowing anything; handicapped people should not be ridiculed; white lies are permissible; we should not publicly rejoice in our enemy's misfortune; deliberate cruelty is wrong; one should be loyal to friends; confidential information should not be made public; courage, honesty, and fairness are good, and their opposites are bad.

This list is not exhaustive or representative. The items on it are more or less randomly selected, and many of them are uncodified. They are often held unconsciously, and conformity to them is usually habitual. People often go against these conventions, but when they do they can normally be shown, in a cool moment, the wrongness of their actions. Habitual violators of these conventions are regarded as immoral or abnormal.

No society is perfect, of course. None can dispense with the need for improvement, and that is true of American society as well. The prevailing conventions should always be open to change, and are always likely to be changing in one respect or another. Nonetheless, there must also be continuity, because the reasons for change depend on criticizing some prevailing convention in terms of other prevailing conventions, which are thought to be deeper or more important. Our society is presently in moral turmoil about abortion and euthanasia. A significant feature of these controversies is that both sides try to justify their position by appealing to conventions they can assume to be generally shared, even by their opponents. Supporters of abortion and euthanasia appeal to freedom rights; opponents appeal to the importance of protecting life.

At this point, the objection is likely to be made that the conservative attempt to provide reasons for terms of cooperation is circular, question-begging, and hence a failure. According to the conservative view, it may be objected, justice consists in people having what they deserve and not having what they do not deserve; what is or is not deserved is said to depend on the prevailing terms of cooperation; and these terms are explained by the conventions governing relationships, agreements, and actions that affect the security of others. But how could it be decided, on this view, whether the terms of cooperation are themselves just? On what grounds could the prevailing terms be criticized if they are conventional but unjust? This objection, however, is misplaced; the answer to it is obvious, and follows from what has already been said.

Terms of cooperation could be criticized on the grounds that they rest on some mistake of fact, proportion, identification of good and bad things, or the recognition of relevant individual differences, and that the mistake prevents people from having what they deserve or not having what they do not deserve. Master-slave relationships are prejudice-ridden because they wrongly suppose that slaves are inferior to masters. Marriage contracts in which the husband acquires control over the wife's property are unfair because they falsely deny the wife's capacity to make reasonable decisions. The punishment of theft by cutting off an arm is deplorable because of its disproportionality. Depriving people of the fruits of their work in order to benefit those who do not work although they could is wrong because it ignores the difference in what they deserve. In all these cases, the terms governing the distribution of benefits are unjust because they lead to people having more or less than what they deserve. The bases for these criticisms are the mistakes on which the terms of cooperation appealed to rest.

It must be acknowledged, of course, that there may be serious disagreements about whether or not a particular term of cooperation is mistake-free. These disagreements, however, concern the question of what is deserved. They arise because, and only because, the parties to them accept the conservative view that justice depends on people getting what they deserve and not getting what they do not deserve. The existence of such disagreements thus strengthens, rather than weakens, the conservative view.

The conservative view, therefore, should not be understood as advocating that what people deserve depends merely on the prevailing terms of cooperation. The conservative view is that what people deserve depends on those prevailing terms that are free of mistakes in fact, proportion, identification, and the recognition of individual differences. That it is reasonable to maintain terms of cooperation that are mistake-free and have passed the test of time—those that endure, are voluntarily adhered to, and contribute to the well-being of members of a society—may be the conservative view, but it is also the view that reasonable people are likely to share.

V. Is the Conservative View Practical?

It has been said by more than one critic that the distribution of good and bad things on the basis of what people deserve is impractical.⁷

This objection rests on two assumptions that have led to misunderstanding the conservative view: one is that what people deserve depends on their personal qualities; the other is that the distribution of deserved good and bad things is the responsibility of the government. Both assumptions are mistaken, but neither is totally mistaken. Personal qualities may provide reasons for saying that people do or do not deserve something, but they are merely one possible kind of reason among many. The government does have responsibility for distributing some deserved good and bad things, but there are many others distributed by civic institutions that function independently of the government.

The reasons I have given for the distribution of what people deserve were derived from terms of cooperation governing relationships, agreements, and actions affecting security. Personal qualities—virtues and vices, for instance—may be reasonably added to the three I have considered. But even if it were true that there are great practical difficulties in ascertaining what personal qualities people have, this is not true of their relationships, agreements, and actions. The latter are open to public observation, and there is no greater practical difficulty in identifying them than there is in identifying countless other features people have.

It adds to the implausibility of this objection that it is often possible to identify personal qualities, even though they have a subjective dimension. What, according to this objection, runs into insuperable practical difficulty is routinely done by families, teachers, coaches, selection committees, employers, and countless other people who are charged with making judgments about the personal qualities of some people. These judgments, of course, can be good or bad. If they are good, the alleged practical difficulties have been overcome. If they are bad, it need not be because practical difficulties stand in the way.

This last rejoinder, however, will be regarded as irrelevant because of the second mistaken assumption on which the objection rests. If the distribution of deserved good and bad things were the responsibility of the government, then the success of individuals in identifying personal qualities would be irrelevant. The question then is whether governments, not individuals, can succeed in formulating policies about the distribution of good and bad things on the basis of personal qualities. The answer to this question is twofold. First, as we have seen, the distribution of deserved good and bad things is not based exclusively on personal qualities; it is also based on the relationships, agreements, and actions of the recipients. Second, the distribution of deserved good and bad things is not the exclusive prerogative of the government; the institutions of a civil society have a very large role in it.

These institutions are families, schools, universities, corporations, athletic competitions, small businesses, orchestras, museums, quiz shows, honor societies, foundations, committees that award prizes, parole boards, neighborhood groups, arbitration panels, and many other more or less formal associations of people that stand between the private concerns of individuals and the political responsibilities of the government. They distribute money, honors, status, prestige; they distinguish between excellence, mediocrity, and deficiency; they set standards and evaluate performance by them; they assign rewards and mete out punishments; they have ways of rectifying violations of their procedures and standards; and they continually face and resolve disputes among their members about these matters. The institutions of a civil society, therefore, routinely identify personal qualities relevant to their concerns, and, of course, they do the same with the relevant relationships, agreements, and actions of their members.

Even if it were true, therefore, that practical difficulties prevent the government from knowing what the deserved distribution of good and bad things is, the same would not be true of the institutions of a civil society. But, of course, the government can also do what these institutions do. The government has no great practical difficulty in identifying some requirements of relationships, such as marriage, parenthood, or citizenship; of agreements, such as taking out a mortgage, enlisting in the army, buying a car, or being a patient in a hospital; and of actions affecting the security of others, such as driving a car, having a fight, or owning a gun. All these cases may constitute reasons for the distribution of good and bad things. There is, therefore, no good reason to suppose, as this objection does, that distribution based on what people deserve is vitiated by great practical difficulties.

An entirely different kind of objection to the practicality of the conservative view is that it is unrealistically optimistic. It supposes that if terms of cooperation are free of mistakes and have stood the test of time, then justice will prevail because there will be no obstacles left to people's having what they deserve and not having what they do not deserve. This supposition is false, it will be objected, because it ignores human wickedness that often leads people to act unjustly, the scarcity of resources that may make it impossible for people to have what they deserve, and the contingency of life that may result in undeserved misfortune.

There is a great deal to this objection. Wickedness, scarcity, and contingency undoubtedly stand in the way of justice, and this has an important effect on how the conservative view should be understood. This view is not the description of any existing society. It is a description of what a society would be like if it were just. It is perfectly compatible with this description to acknowledge that wickedness, scarcity, and contingency are likely to be permanent obstacles to a society's actually being just. This, however, does not make the conservative view impractical. On the contrary, it provides a special impetus for trying to make one's own society as just as possible in the face of these obstacles. In order to come closer to this highly desirable goal, it must be recognized that justice has two aspects.

One is the distribution of good and bad things, which is what I have been discussing up to now. The other is the rectification of injustice that is bound to occur in distribution as a result of wickedness, scarcity, and contingency. The standard to which just distribution and rectification should aim to conform is that the good and bad things in life should be proportional to the goodness or badness of their recipients.

The fact remains, however, that this standard is unattainable because the human condition stands in the way. Scarcity limits just distribution through the insufficiency of available resources. No matter how strong is the commitment to just distribution, if there is not enough money, food, medicine, prison space, police protection, or hospital care available, people cannot have what they deserve. Rectification is concerned with correcting unjust distribution. Its purpose is to make good and bad things proportional to what is deserved.

Disproportionality, however, is often an insurmountable obstacle to rectification. Some forms of injustice cannot be rectified. Nothing could compensate people who sacrificed their lives for a noble cause, who were blinded or disfigured in an accident for which they were not responsible, or who were forced to spend the best years of their lives in concentration camps on trumped-up charges. Nor is there a proportional punishment for mass murderers, torturers, or fanatics who destroy great works of art. Nothing could redress such imbalances because no punishment could be commensurate with what is deserved by these evildoers. Disproportionality, therefore, unavoidably limits efforts at rectification.

Wickedness seems to be another obstacle to justice. The same familiar vices recur throughout recorded human history. Cruelty, greed, destructiveness, selfishness, malevolence, envy, fanaticism, and racial, religious, and ethnic prejudice have motivated people in very different times, places, and circumstances. There is no reason to suppose that in this respect the future will be different from the past. No doubt, the forms in which these and other vices are and will be expressed are bound to change, but they will be expressed and they will inflict much injustice on innocent victims. There are, of course, also virtues, but they coexist with the vices in any given society and often in the same person. If vices are part of human nature, as virtues are, then they cannot be eliminated from the human repertoire. The best we can do is to limit their scope.

But even if, per impossibile, scarcity and wickedness could be overcome in some way, the contingency of life will remain and it will be responsible for much misfortune befalling undeserving victims. Lightning will strike, buildings will collapse, volcanoes will erupt, earthquakes will occur, cancer will strike, metal fatigue will make bridges collapse, viruses will mutate and invade the human immune system, cars, ships, and airplanes will crash, and so on and on.

As a result, the conservative view should not be understood as having the unattainable aim of making society just, but as having the much more modest aim of making society as little unjust as scarcity, wickedness, and contingency allow. The aim is to decrease imperfection, not to achieve perfection. That is a realistic aim that cannot be convicted of undue optimism.

VI. Implications

The conservative view of justice has implications for many actual or possible policies of the government. I will now venture a few remarks about its implications for taxation and prosperity, but I do so hesitantly because I can claim no expertise in policymaking. To begin with taxation, I take it that everyone agrees that tax laws need justification. Let us ask what such justification or criticism may be of tax laws affecting individuals. (I ignore businesses and other organizations.) Levying taxes on the legitimate income of individuals means that individuals must give up a portion of what they deserve to have. This requires justification, and the government bears the burden of providing it. Part of the justification is the tacit agreement that the government provides services that people need and want, even if they do not use them personally, and they must pay for them.

The government provides one kind of service that all reasonable citizens benefit from. In a large society, such as that of the United States, individual well-being requires the protection of security, the maintenance of infrastructure and a legal system, the negotiation of treaties, and so forth. Since everybody benefits from such services, everybody should pay for them. Taxation for this purpose seems to me justified. The government also provides services that people may reasonably choose not to use because they are willing to do without the benefits, or because they prefer the services of private firms. Social Security, unemployment insurance, Medicare, various publications, licenses, TV channels, and so on, according to the conservative view, should be paid for by taxing only those who choose to use them in the present or the future. A third kind of service the government provides is help for those who are in need because of poverty, illness, misfortune, or some other reason. From the mere fact that people are in need it does not follow that they deserve to be helped or that other people deserve to be deprived of a portion of their legitimate income in order to help them. The needy deserve help only if the creation of their need is not their fault and if others have an obligation to provide the help at considerable cost to themselves and over and above their other obligations. Egalitarians favoring existing welfare policies ignore the question of fault and take for granted that there is an obligation to help without explaining its basis.⁸

I need now to say something about prosperity. Prosperity is clearly a good thing, but it is overrated. Individual well-being does not depend on living in a prosperous society, and it depends on many social and individual goods that are not economic (so that prosperity cannot provide them). What individual well-being requires is to have sufficient resources for living in a particular way, not the amplitude of resources that prosperity implies. The individual goods needed for well-being vary from life to life and from society to society, but there are some social goods that are necessary regardless of how well-being is understood. Without security, freedom, justice, order, health, division of labor, cooperation, and so forth, well-being is impossible. It is, therefore, a bad mistake to suppose that the well-being of individuals in a society can be measured by the prevailing prosperity. The prosperous life of Al Capone and the prosperous society of Nazi Germany were bad, and lives and societies can be good even if only merely adequate resources are available.⁹

A further implication of the conservative view concerns social change. All societies are always changing because the conditions to which people have to respond change. A society cannot endure unless it adjusts itself to new circumstances. Having some control over change requires asking and answering the questions of what makes change desirable and to what should the change lead. The conservative answer is that it is desirable to change if the prevailing conventions are not mistake-free or no longer pass the test of time. And the change should be to a revised form of the same convention that does not have the shortcomings of its predecessor. The reason for change, then, is to remedy the specific shortcomings of a specific convention. The conservative view is that change should be incremental, piecemeal, specific, and as little as is sufficient to remedy the shortcoming that prompts it.

Part of the significance of this conservative attitude toward social change emerges by noticing what it does not involve. It does not involve change according to a theory, an overall design, a general plan, the approximation of a distant goal; it does not involve wholesale changes, or abandoning conventions that have no discernible shortcomings, or change simply to try something new. The conservative view is that unless there is a shortcoming, there is no reason to change, and when there is no reason to change, there is reason not to change. The justification of this view is simply to point out that because the prevailing conventions are mistake-free, pass the test of time, are tried and true, command the allegiance of people, and set terms for agreements, relationships, and actions affecting the security of others, they have advantages that theories, overall designs, distant goals, and new possibilities lack. It is never reasonable to abandon satisfactory actualities in the present for the sake of uncertain possibilities in the future.

It may happen, of course, that the bulk of the conventions prevailing in a society are seriously defective. In that case, conservatives of course would not be in favor of maintaining them. Societies may be rotten through and through, as we know well from history, and then they should go under. To admit this is damaging only for a view that favors maintaining conventions simply because they have endured. This, however, is not the conservative view. The conservative view is to favor only those conventions that are reasonably regarded as mistake-free and passing the test of time.

Yet another implication of the conservative view is that the content of justice is likely to differ from society to society, or indeed, with the passage of time, within one and the same society. A society aiming to be just is committed to its members' having what they deserve and not having what they do not deserve. But what is and is not deserved depends on the prevailing conventional terms of cooperation. Different societies are likely to have different conventions about relationships, agreements, and actions affecting the security of others, and the conventions are likely to change with time in all societies. Commitment to justice, therefore, is not commitment to rules or principles that can be set a priori or with finality. It is commitment to treating people as they deserve to be treated, while recognizing that what they deserve depends on variable and changing conventions that are mistake-free and pass the test of time. It is necessary for justice that there be such conventions, but it is not necessary that these conventions should have the same content in all societies.

The last implication of the conservative view I shall mention here is that justice is essentially inegalitarian. Justice has to do with the distribution of what people deserve, and what they deserve is different because they have different relationships and agreements, and because their actions affect the security of others differently. To be sure, conservatives could agree that justice requires treating equals equally and unequals unequally. The crucial question, however, is about the respect in which people are supposed to be equal or unequal.¹⁰

Egalitarian views advocate the redistribution of economic resources in order to decrease the difference between the rich and the poor. They thus advocate taking resources from people who have more and giving them to those who have less without asking whether the first deserve what they have and whether the second deserve to get what egalitarians want to give them. Proceeding this way may conceivably be justified, but its justification cannot be that justice requires it. If egalitarians had the courage of their convictions, they would simply provide such justification as they can instead of misappropriating justice and hanging it as a label on the unjust policy they advocate.