Is a government required or permitted to redistribute the gains and losses that differences in biological endowments generate? In particular, does the fact that individuals possess different biological endowments lead to unfair advantages within a market economy? These are questions on which some people are apt to have strong intuitions and ready arguments. Egalitarians may say yes and argue that as unearned, undeserved advantages and disadvantages, biological endowments are never fair, and that the market simply exacerbates these inequities. Libertarians may say no, holding that the possession of such endowments deprives no one of an entitlement and that any system but a market would deprive agents of the rights to their endowments. Biological endowments may well lead to advantages or disadvantages on their view, but not to unfair ones.

I do not have strong intuitions about answers to these questions, in part because I believe that they are questions of great difficulty. To begin, alternative answers rest on substantial assumptions in moral philosophy that seem insufficiently grounded. Moreover, the questions involve several problematical assumptions about the nature of biological endowments. Finally, I find the questions to be academic, in the pejorative sense of this term. For aside from a number of highly debilitating endowments, the overall moral significance of differences between people seems so small, so I interdependent and so hard to measure, that these differences really will 1 not enter into practical redistributive calculations, even if it is theoretically i permissible that they do so.

Before turning to a detailed discussion of biological endowments and their moral significance, I sketch my doubts about the fundamental moral theories that dictate either the impermissibility or the obligation to compensate for different biological endowments.

Liberalism has often been viewed as a continuing dialogue about the relative priorities between liberty and equality. When the version of equality under discussion requires equalization of outcomes, it is easy to see how the two ideals might conflict. But when the version of equality requires only equalization of opportunities, the conflict has been treated as greatly muted since the principle of equality seems so meager in its implications. However, when one looks carefully at various versions of equal opportunity and various versions of liberty, the conflict between them is, in fact, both dramatic and inescapable. Each version of the conflict poses hard choices which defy any systematic pattern granting priority to one of these basic values over the other. In this essay, I will flesh out and argue for this picture of fundamental conflict, and then turn to some more general issues about the kinds of answers we should expect to the basic questions of liberal theory.

“Does the existence of unequal social and economic starting points in life nullify capitalism's claims to justice?” Notice is hereby given that this essay's answer to this question is an unequivocal “maybe.” For it is a banal but true claim that everything depends upon what is meant by capitalism, justice and life's starting point. And it is a less banal but no less true claim that their meanings are opaque or controversial or both. In what follows I shall devote little attention to the question of what justice is and shall simply presume that it is best characterized by historical entitlement theory. The last part of the essay discusses the notion of life's starting point and vacillates over what it would mean for one such point to be equal to another. Hence, the bulk of my argument is taken up with exploring what capitalism must be like to conform to historical entitlement theory. And my conclusion will be that, insofar as capitalism respects every person's right of self-ownership, its claims to justice require that no one be denied one type of equal payment – a payment which might be rendered at each life's starting point.

Rights to compensation are much invoked and much disputed in recent liberal debates. The disputes are generally about supposed fundamental (natural, human or moral) rights to compensation, whose recognition and legal enactment would transform some lives. For example, special treatment in education or employment are claimed as compensation for past denials of equal opportunity; special consideration for Third World countries in aid and trade terms is claimed as compensation for the injustices of the colonial past.

We can make ready sense of the idea of legal rights to compensation. Legal rights to compensation guarantee (some) recompense for damage suffered. The damage for which compensation is given may or may not be produced by wrongdoing; it may also be negligent or accidental or due to natural causes. In law, compensation is not always contingent upon the victim having suffered injury as well as harm, nor upon saddling a wrong-doer with the costs of compensation. Insurance policies standardly cover damage due to accident and neglect. Legal rights to compensation may provide for payments to victims of violent crime, to those whose property is requisitioned or damaged, to victims of libel or malpractice, and even to victims of natural disasters such as floods or earthquakes. Legal rights to compensation are a standard way of dealing with the predicament of those who become victims, whether of others' (criminal) action, of their own negligence, or of natural catastrophe.

None of this shows that there are fundamental rights to compensation. Legal rights to compensation may lack moral grounding: and if they are morally grounded, they may rest not on fundamental rights to compensation, but on background positions in which rights are derivative or inessential.

As a preliminary to the justification of equal opportunity, we require a few words on the concept. An opportunity is a chance to attain some goal or obtain some benefit. More precisely, it is the lack of some obstacle or obstacles to the attainment of some goal(s) or benefit(s). Opportunities are equal in some specified or understood sense when persons face roughly the same obstacles or obstacles of roughly the same difficulty of some specified or understood sort. In different contexts we might have different sorts of benefits or obstacles in mind. But in the current social context, and in the context of this discussion, we refer to educational and occupational opportunities, chances to attain the benefits of higher education and of socially and economically desirable positions, benefits assumed to be desired by many or most individuals, other things being equal. And we generally divide obstacles into two broad classes: those imposed by the social system or by other persons in the society, for example, the hardships of life in the lower economic classes or barriers from prejudices based on race, sex, or ethnic background; and those imposed by natural disabilities, for example, low intelligence or lack of talents.

The initial question is whether a moral society is obligated to create equality in opportunities in the senses just defined. I shall assume here initially that there is some such obligation on the part of society or the state, although I shall specify its nature and limits more precisely below. With the exception of certain libertarians, almost everyone, liberal and conservative alike, agrees in this assumption.

The United States has never been culturally or religiously homogeneous, but its diversity has greatly increased over the last century. Although the U.S. was first a multicultural nation through conquest and enslavement, its present diversity is due equally to immigration. In this paper I try to explain the difference it makes for one area of thought and policy – equal opportunity – if we incorporate cultural and religious pluralism into our national self-image. Formulating and implementing a policy of equal opportunity is more difficult in diverse, pluralistic countries than it is in homogeneous ones. My focus is cultural and religious diversity in the United States, but my conclusions will apply to many other countries – including ones whose pluralism is found more in religion than in culture.

We are often told that we are morally obligated to produce equal opportunity for all. Therefore, it seems we should examine what power we have to produce that desirable state. For it would be nonsense to say we are required to provide what is beyond our power to provide. When we examine this question, we find our power limited by two sets of constraints. One set comprises formal constraints upon the idea itself of equal opportunity. We cannot do the logically impossible. The other set comprises limits upon our ability to produce the directed socio-economic change, getting known outputs for known inputs. I illustrate the formal constraints by outlining the work of Douglas Rae. The constraints upon our abilities I illustrate with evidence from sociology and politics. At the end, we shall discover that our power to make opportunities equal is sharply though not unbearably limited. A critical but unbaised survey will reveal that in the past fifty years we have gone remarkably far towards doing all that we are presently capable of doing to equalize opportunities. Perhaps we shall go even farther when we learn how.

The word ‘real’ in the title is opposed to ‘ideal’ or even ‘chimerical'. It may seem an interesting question what equality of opportunity should consist in were we able to produce directed socio-economic change at will. But we are not. Therefore, a more interesting and more important question is what equality of opportunity consists in given the very large number of constraints within which we must work to achieve it.

Philosophers have long distinguished various interpretations of the principle of equal opportunity and argued over their implications and justifications. But they have almost always tacitly assumed that the context was a national one. They have not, in particular, considered whether some interpretation of the principle could apply and be justified globally, that is, to all people without regard to their nationality or citizenship. Yet, such an investigation is clearly demanded. The leading moral theories seem to support a case for at least some interpretation of the equal opportunity principle, and it is not obvious that they can support it only domestically.

Consider, first, those moral theories which place great value on negative liberty, for example, libertarianism. Libertarianism supports a standard interpretation of the equal opportunity principle – “formal” equality of opportunity; formal equality of opportunity requires that legal restrictions j on the taking of opportunities be lifted, and such restrictions diminish negative liberty. But libertarianism would also seem to support a global. version of formal equality of opportunity, for example, that laws be rescinded which require that candidates for jobs in a country be citizens of that country, or which restrict emigration or immigration. Such laws also diminish negative liberty.

Or consider those moral theories which place great value on efficiency, for example, utilitarianism. Utilitarianism probably supports formal equality of opportunity because legal restrictions on the taking of opportunity not only diminish negative liberty, but also often prevent talent and skill from going where it can best be used and thus reduce efficiency.

Discrimination against particular groups has existed throughout history and in all types of societies. Few would challenge the idea that inequality of income based on discrimination is unjust. The more problematic issues are the extent to which discrimination is in fact a significant source of inequality and whether such discrimination-based inequality is inherent in a capitalist system.

There is little doubt that discrimination can affect a group's income. But the link is by no means automatic or certain. Thus, the incomes of blacks, particularly in past decades, seem surely to have been lowered by discrimination. Yet other examples are less clear. Jewish and Japanese Americans, for instance, have had incomes substantially above those of white non-Jewish groups, despite evidence of discrimination against them.

Equal opportunity requires (at least) that persons be selected for desirable positions on the basis of their qualifications. To assess an applicant's qualifications, we must both predict how well he would perform if chosen, and compare his projected performance with that of his rivals. Since we lack direct access to future performance (and since only those, who are chosen will ultimately perform), all such predictions must be based on some past– or present-tense information about the applicants, together with some relevant supporting information. But is any and every way of predicting performance acceptable? Or are some methods of predicting improper even if they are more accurate than any available alternatives? And, if some methods of predicting are improper, which ones are these, and why?

No one who cares about equal opportunity can derive much comfort from the present occupational distribution of working women. In the various industrial societies of the West, women comprise between one quarter and one-half of the national labor force. However, they tend to clustered in employment sectors – especially clerical, sales, and service J occupations – which rank relatively low in remuneration, status, autonomy, and other perquisites. Meanwhile, the more prestigious and rewarding managerial and professional positions, as well as the major categories of blue-collar labor, remain largely a male preserve. In the same societies the average income earned by full-time female workers is one-half to two- J thirds that of their male counterparts. Although this disparity owes much to i other factors, including lower pay for work similar or even identical to that r standardly done by men, much of it can be explained only by the concentration of working women in traditional female job ghettos.

The headlines at the outset of 1987 told of Howard Beach, where a group of blacks had been chased, and one killed, because they had unwittingly entered a white enclave in New York City. And they told of Forsythe County, Georgia, where the mere presence of civil rights marchers, in a place from which blacks had been driven three-quarters of a century earlier, brought out depths of antagonism unknown since an earlier era of civil rights marches.

Behind both events – indeed, behind almost every question of race to arise in recent years – was the specter of affirmative action. Even as, during the late 1960s, some blamed urban riots on the federal government's failure to achieve equal opportunity between the races by equalizing life outcomes, so in 1987 white antagonisms were regarded in some quarters as a crude reflection of the Reagan administration's hostility towards affirmative action. The rhetoric and the policies of that administration, it was said, contributed to a sentiment that blacks already had their share – indeed, more than their share. Official word and deed contributed also, it was argued, to a resentment against blacks who, because of quotas, had been unfairly advantaged.

This essay is about the ethical propriety and practical efficacy of a range of policy undertakings which, in the last twenty years, has come to be referred to as “affirmative action.” These policies have been contentious and problematic, and a variety of arguments have been advanced in their support. Here I try to close a gap, as I see it, in this “literature of justification” which has grown up around the practice of preferential treatment. My principal argument along these lines is offered in the next section. I then consider how some forms of argument in support of preferential treatment, distinctly different from that offered here, not only fail to justify the practice but, even worse, work to undermine the basis for cooperation among different ethnic groups in the American democracy. Finally, I observe that as a practical matter the use of group preference can, under circumstances detailed in the sequel, produce results far different from the egalitarian objectives which most often motivate their adoption.

It may seem fatuous in the extreme to raise as a serious matter, in the contemporary United States, the question “Why should we care about group inequality?” Is not the historical and moral imperative of such concern self-evident? Must not those who value the pursuit of justice be intensely concerned about economic disparities among groups of persons? The most obvious answer to the title question would seem, then, to be: “We should care because such inequality is the external manifestation of the oppression of individuals on the basis of their group identity.”