RACIAL DISCRIMINATION

The nation was founded with the enslavement of blacks as an established and ongoing institution, and though we were not particularly proud of the institution, we were prepared to live with it. The Constitution did not mention the word "slave," and contemplated the eventual closing of the slave trade (referred to simply as the "importation of persons"), but, through similar circumlocutions, also created obligations to return fugitive slaves, and included a proportion of the slaves within the population base to be used for the apportionment of representatives and taxes. In dred scott v. sandford (1857) the Supreme Court viewed slaves as property and declared that the right of slaveholders to take their slaves to the territories was protected by the due process clause of the Fifth Amendment.

The civil war brought slavery to an end and reversed the basic commitment of the Constitution toward blacks. The law sought equality rather than enslavement, and it was through the elaboration of this egalitarian commitment that the concept of racial discrimination emerged. Prohibiting racial discrimination became the principal strategy of the American legal system for achieving equality for blacks. The laws against racial discrimination typically protect all racial minorities, not just blacks, and yet, for purely historical reasons, the development of those laws would be unimaginable apart from the struggle of blacks for equality in America. That struggle has been the source both of the achievements of antidiscrimination law and of its recurrent dilemmas.

The three amendments adopted following the civil war constitute the groundwork of this branch of the law, although only one—the fifteenth amendment—a ctually speaks of racial discrimination. It provides that "the right… to vote shall not be denied or abridged … on account of race, color, or previous condition of servitude." The other Civil War amendments are not cast in terms of racial discrimination. The thirteenth amendment prohibits slavery and involuntary servitude, and the fourteenth amendment, in relevant aspect, prohibits states from denying "the equal protection of the laws." But the Supreme Court has interpreted both these amendments to prohibit racial discrimination. With respect to the Thirteenth Amendment, the Court reasoned in jones v. alfred h. mayer co. (1968) that racial discrimination is a badge or incident of slavery. (See badges of servitude.) Similarly, in interpreting the Fourteenth Amendment, the Court, as early as strauder v. west virginia (1880), declared racial discrimination to be the kind of unequal treatment that constitutes a denial of equal protection of the laws. Indeed, over the years, racial discrimination came to be seen as the paradigmatic denial of equal protection, and supplied the standard against which all other equal protection claims came to be measured, even when pressed by nonracial groups such as the poor or women. They too had to show that they were discriminated against on the basis of some impermissible criterion such as their wealth or sex. The promise of equal protection was thus transformed into a promise not to discriminate.

It was, moreover, through the enforcement of the Fourteenth Amendment that the prohibition against racial discrimination achieved its greatest prominence. Antidiscrimination was the instrument that finally put to an end the system of white supremacy that emerged in the late nineteenth and early twentieth centuries and that worked by separating whites and blacks—Jim Crow. The discrimination appeared on the very face of Jim Crow laws and a principle that condemned racial discrimination easily brought those laws within the sweep of the Fourteenth Amendment. All that was needed was an understanding of how the separatism of Jim Crow worked to the disadvantage of blacks; that was the burden of brown v. board of education (1954) and the cases that followed. As the principle controlling the interpretation of the Fourteenth Amendment, antidiscrimination was a limitation only upon the actions of states, but once the step entailed in Brown was taken, the federal government was, in bolling v. sharpe (1954), made subject to an identical prohibition by a construction of the due process clause of the Fifth Amendment. Racial discrimination was deemed as inconsistent with the constitutional guarantee of liberty as it was with equal protection.

Statutes, too, have been concerned with racial equality. In the years immediately following the Civil War, Congress passed a comprehensive program to protect the newly freed slaves, and defined the conduct it sought to prohibit in a variety of ways. In the civil rights act of 1866 Congress promised that blacks would enjoy the same rights as whites; in the force acts (1870, 1871) it guaranteed all citizens the rights and privileges arising from the Constitution or laws of the United States. In the decades following Brown v. Board of Education, however, when the antidiscrimination principle of the Fourteenth Amendment received its most strenuous affirmation and the nation embarked on its Second Reconstruction, Congress cast the substantive standard in terms of a single idiom—do not discriminate. (See civil rights act of 1964; civil rights act of 1968; voting rights act of 1965.)

During this period, Congress introduced new mechanisms to enforce the equal protection clause; for example, it authorized the attorney general to bring injunctive school desegregation suits, required federal administrative agencies to terminate financial assistance to segregated school systems, and provided for criminal prosecutions against those who forcibly interfered with desegregation. Congress also broadened the reach of federal antidiscrimination law beyond the scope of the Fourteenth Amendment by regulating, in the name of racial equality, activities of private agencies (for example, restaurants, employers, or landlords), which otherwise would not have been covered by that amendment because of its "state action" requirement. In each of these measures, Congress used the language of antidiscrimination. So did the President in promulgating executive order 11246 (1965), which regulates government contractors. Many state legislatures also intervened on behalf of racial equality during the Second Reconstruction, and these enactments were also couched in terms of prohibiting discrimination.

Sometimes Congress and the state legislatures exempted certain discriminatory practices from the laws they enacted. One instance is the federal open housing law, which exempts discrimination by small residences ("Mrs. Murphy's roominghouse"); another is the federal fair employment statute, which exempts from its coverage small businesses (at first businesses with fewer than twenty-five employees, later reduced to fifteen). Apparently Congress viewed the interest in associational liberty present in these settings as sufficiently strong to justify limited exemptions to the ban on racial discrimination. Yet, putting these exemptions and a handful of others to one side, it is fair to say that today, primarily as a result of the Second Reconstruction, the prohibition against racial discrimination is all-encompassing. It has both constitutional and statutory bases and is the subject of an executive order. It is a pervasive feature of both federal and state law and calls forth a broad array of civil and criminal remedies. It almost has the status of a moral imperative, like the norm against theft or killing. The issue that divides Americans today is thus not whether the law should prohibit racial discrimination but what, precisely, doing so entails.

The antidiscrimination norm, as already noted, was largely fashioned at a time when the nation was swept by the separate but equal doctrine of Jim Crow and when blacks were disadvantaged in a rather open and crude manner. In such a context, the principle of antidiscrimination invites a color blindness: When allocating a scarce opportunity, such as a job or a place in a professional school, the decision maker should not prefer a white candidate over a black one on the basis of the individual's color or race. Here antidiscrimination requires that individuals be judged independently of race. This much is settled. Interpretive problems arise, however, when the social context changes—when we have moved beyond Jim Crow and blacks have come to be disadvantaged primarily in ways that are hidden and systematically entrenched. Then we confront two issues. One arises from the exclusion of blacks on the basis of a seemingly innocent criterion such as performance on a standardized test; the other from the preference given to blacks to correct for longstanding unequal distributional patterns.

To clarify the first issue, it should be understood that the appearance of innocence might be misleading. Although a black scores higher than a white on a test, the employer might manipulate or falsify the scores so that the white is given the job. In this case, the apparent use of an innocent criterion is simply a mask for racial discrimination. The decision is still directly based on race and would be deemed unlawful. The most straightforward remedy would be to set aside the decision and allow an honest application of the test.

There are, moreover, situations when a test is honestly administered and yet the very decision to use the test in the first place is based on an illegitimate concern, namely, a desire to exclude blacks. A highly sophisticated verbal aptitude test might be used, for example, to select employees for manual work because the employer, wanting to maintain a predominantly white work force, assumes that fewer whites than blacks will be screened out by the test. Here again, the "real" criterion of selection is race; a court would disallow the use of the irrelevant test, and require the employer to choose a criterion that serves a legitimate end. In both of these cases—the dishonest application of legitimate criteria and the honest application of illegitimate criteria—the appearance of color blindness is a sham and a court could use the simple, colorblind form of the antidiscrimination norm to void the results.

The more troublesome variant of the first issue arises when (1) the facially innocent criterion is adopted in order to serve a legitimate interest; (2) the criterion in fact furthers that interest; and (3) the application of the criterion disadvantages the racial minority in much the same way as would the use of race as the criterion of selection. The job may in fact require sophisticated verbal skills and the test that measures these skills may screen out more blacks than whites. The test is job-related but has a disparate adverse impact on blacks. The question then is whether an employment decision based on the test violates the antidiscrimination prohibition. This is a question of considerable difficulty because while the law, strictly speaking, prohibits distinctions based on race, this particular decision is based on a criterion other than race.

One school of thought answers this question in the negative. This view stresses process, and interprets antidiscrimination in terms of the integrity of the selection process: A selection process based on race is corrupt and cannot be allowed. A selection process free of racial influence might redound to the benefit of the racial minority, since it would allow them to compete on equal footing with other groups and thus give them a chance to alter the distributional inequalities that occurred under a regime such as Jim Crow, where they were penalized because of their race. Any actual effect on their material status as a group, however, would represent just an agreeable by product, or a background assumption, not the purpose of antidiscrimination law. According to this school, the aims of antidiscrimination law are fulfilled when the process of selection is purified of all racial criteria or motivations.

Another viewpoint stresses results or effects, not process; it would find the use of the innocent criterion unlawful even if it serves legitimate ends. What is decisive, according to this school of thought, is the actual disadvantaging of blacks, not the way the disadvantage comes about. If the application of a criterion has a disproportionately adverse impact on the racial minority, in the sense that it excludes substantially more blacks than whites, the criterion should be treated as the functional equivalent of race.

At the heart of this interpretation of antidiscrimination is a concern for the social status of blacks. It is motivated by a desire to end all practices that would tend to perpetuate or aggravate their subordinate position. Admittedly, the costs of this program are real, for it is stipulated that the contested criterion serves some legitimate end; the test is job-related. But these costs are seen as a necessary price of justice. Only when the costs become extraordinarily large or achieve a special level of urgency, as when the criterion serves some "compelling" (and not just a "legitimate") interest, will the use of the criterion be allowed.

The theorist who so emphasizes effects rests his argument principally on the Fourteenth Amendment and ascribes to it the grandest and noblest of purposes—the elimination of caste structure. He insists that antidiscrimination, as the principle that controls the application of that amendment, be construed with this broad purpose in mind and if need be, that a new principle—the group-disadvantaging principle—be articulated in order to make this purpose even more explicit. He also insists that the various statutes that prohibit discrimination—the principal argumentative props of the process school—should be construed derivatively. These statutes, unlike the Fourteenth Amendment, may contain in so many words a specific ban on "discrimination based on race," but, so the effects theorist argues, these statutes should be seen as a legislative adoption of the prevailing constitutional principle. When that principle is interpreted to forbid the use of criteria that effectively disadvantage blacks, the statutes should be interpreted in a similar fashion.

The process school emphasizes not only the precise language in which the statutory norm is cast but also the traditional rule that conditions judicial intervention on a finding that the defendant is at fault. This fault exists when a white is given a job over a black even though the black scored higher on a test; the employer is said to be acting wrongfully because race is unrelated to any legitimate purpose and is a factor over which individuals have no control. But the requisite fault is said to be lacking when the selection is made on the basis of the individual's performance under some nonracial standard, such as a job related test. On the other hand, those who subscribe to an effects test emphasize the prospective nature of the remedy typically sought in these cases (an injunction to forbid the use of the criterion in the future) and deny the need for a finding of fault. Such a finding may be necessary to justify damages or the criminal sanction, because these remedies require the defendant to pay for what he did in the past, and presumably such a burden can be placed only on someone who acted wrongfully. But an injunction simply directs that the defendant do what is just and does not presuppose that the defendant has acted wrongfully. Alternatively, the effects theorist might contend that if fault is necessary, it can be found in the defendant's willingness to persist in the use of the contested criterion with full knowledge of its consequences for the racial minority. Such persistence connotes a certain moral indifference.

The disadvantaging that the effects test seeks to avoid is usually defined in terms of the status of a group (for example, the criterion has a greater adverse impact on blacks than on whites and thus tends to perpetuate their subordinate position). Some see this group orientation as alien to our jurisprudence, and thus find a further reason for turning away from an effects test. Borrowing the Court's language in shelley v. kraemer (1948), they insist that "[t]he rights created by the first section of the Fourteenth Amendment are, by its terms, guaranteed to the individual" and that "[t]he rights established are personal rights." But those who subscribe to the effects test see the well-being of individuals and of groups as inextricably linked: They believe that the status of an individual is determined in large part by the status of the group with which he is identified. Slavery itself was a group phenomenon, and any corrective strategy must be structured in group terms. Effects theorists also point to practices outside the racial context that display a concern for the welfare of groups such as religious minorities, women, the handicapped, labor, and consumers, and for that reason insist that a group orientation is thoroughly compatible with American legal principles.

In the late 1960s and early 1970s, the Supreme Court responded to these arguments and moved toward adopting an effects test in cases such as Gaston County v. United States (1969), griggs v. duke power co. (1971), and swann v. charlotte-mecklenburg board of education (1971). There was, however, an element of ambiguity or hesitation in the Court's response. The Court prohibited the use of seemingly innocent criteria that disadvantaged blacks, even when their use served some legitimate interests, but the Court did not justify its decisions solely in terms of the adverse effects of the criteria. In addition, the Court characterized the adverse effect as a vestige of an earlier use of race. For example, a literacy test was disallowed as a qualification for voting not simply because it disqualified more blacks than whites but also because it perpetuated the disadvantages previously imposed on blacks in segregated schools. This insistence on analyzing the disadvantage as a vestige of past discrimination may have reflected a commitment to the process test insofar as the Court treated the earlier procedural imperfection (the assignment to schools on the basis of race) as the legally cognizable wrong and the present practice (the literacy test) as merely a device that perpetuates that wrong. But at the same time, the concern with past discrimination surely reflected some commitment to the effects test, for it resulted in the invalidation of facially innocent criteria that in fact served legitimate ends. Disallowing today's literacy test would avoid perpetuating yesterday's discrimination in the educational system, but only by compromising an interest the Court had previously deemed legitimate, namely, that of having a literate electorate. In fact, an interpretation of antidiscrimination law to forbid practices that perpetuate past discrimination could become functionally coextensive with an interpretation that makes effects decisive if some global practice such as slavery is taken as the relevant past discrimination, if the victims of past discrimination are identified in group terms, and if the remedial burden is placed on parties who had no direct role in the earlier discrimination. All disparate effects can be seen as a vestige of the special and unfortunate history of blacks in America.

By the mid-1970s, however, it became clear that the Court was not inclined to broaden its concern with past discrimination so as to make it the functional equivalent of the effects test. In fact, the Court turned in the opposite direction—away from effects and toward process. As Justice potter stewart announced, "Reconstruction is over." The Court did not flatly repudiate its earlier decisions, but instead tried to limit them by confining the effects test to those antidiscrimination norms that were embodied in statutes. For constitutional claims of discrimination, the Court in cases such as washington v. davis (1976) and mobile v. bolden (1980) required a showing that the process was flawed, or more precisely, that the defendant "intended to discriminate." The plaintiff had to show that the defendant's decision was based on race, or that he chose the seemingly innocent criterion not to further legitimate ends but to exclude or disadvantage blacks. The Court continued to honor claims of past discrimination, but by and large insisted that those claims be advanced by individually identifiable victims of the earlier discrimination, that past acts of discrimination be defined with a great deal of specificity, and that the causal links of those acts to the present racially disparate effects be manifest. No global claims of past discrimination have been allowed.

There is a certain irony in this distinction between statutory and constitutional claims, and in the Supreme Court's decision to confine the effects test to the statutory domain, for the statutes are couched in terms less congenial to such a test. The statutes speak specifically in terms of decisions based on race, while the Fourteenth Amendment speaks of equal protection. (Antidiscrimination is but the judicially constructed principle that is to guide the application of that provision.) Arguably, the distinction between statute and Constitution might reflect the Court's desire to find some way of limiting the practical impact of the effects test, for under the Fourteenth Amendment an effects test would have the widest scope and present the greatest possibilities of judicial intervention. The Fourteenth Amendment extends to all state practices and, because of its universality (it protects every "person"), could be used to protect even those groups that are not defined in racial terms. Indeed, in Washington v. Davis the Court expressed the fear that under the effects test the Fourteenth Amendment might even invalidate a sales tax because of its disproportionately adverse impact on the poor (never for a moment pausing to consider whether suitable limiting principles could be developed for avoiding such a result). The Court's distinction between statutory and constitutional claims might also stem from a desire to devise a means for sharing with other political institutions responsibility for the sacrifice of legitimate interests entailed in the application of an effects test. When attached to the statute, the effects test and its disruptive impact become the responsibility of both Court and Congress, since Congress remains free to repeal the statute or otherwise disavow the test.

In the mid-1970s, at the very moment the Court was struggling to identify the circumstances in which the use of a seemingly innocent criterion could be deemed a form of racial discrimination and was moving away from an effects test, it also had to confront the other major interpretive issue posed by antidiscrimination law, the issue of affirmative action. The Court had to decide whether the norm against racial discrimination prohibits giving preference to blacks.

For much of our history, it was assumed that race-based action would be hostile to blacks and that therefore colorblindness would work to the advantage of blacks or at least shield them from hostile action. During the Second Reconstruction, however, as the drive for racial equality grew stronger, an assertedly "benign" use of race became more common. Many believed that even the honest application of legitimate criteria would not significantly alter the unequal distributional patterns that were produced among the races first under slavery and then under Jim Crow, and that it would be necessary, at least for the immediate or foreseeable future, to give blacks a preference in order to improve their status relative to other groups.

These affirmative action programs typically included other minorities, such as Hispanics, as beneficiaries, but were primarily seen as addressed to blacks and did not extend to all disadvantaged groups, such as the poor or white ethnic minorities. They had a distinctive racial cast and were sometimes described as a form of "reverse discrimination." These programs were also typically structured so as to require the decision maker to achieve a certain number of blacks or other minorities within the institution, say, as employees or students. Often that number equaled the percentage of blacks or other minorities in the general population, and was variously described as a goal or quota, depending on which side of the issue one was on. A "goal" was said to establish the minimum rather than the maximum and to be more flexible than a "quota." But more significantly, the term "goal" did not have the odious connotations of the term "quota," which had been used in the past to describe numerical limits on the admission of minorities, limits that were designed to preserve rather than eradicate the caste structure.

For the most part, these affirmative action programs were not treated as a constitutional or statutory requirement. Some of those who subscribed to an effects test argued that the failure to institute preferential programs would constitute a practice that perpetuated the subordinate position of blacks and thus would be itself a form of racial discrimination. But this argument equated inaction with action, and either for that reason or because the effects test was having difficulties of its own, this argument never established a toehold in the law. Equally unsuccessful were the arguments that emphasized those antidiscrimination laws, such as the federal fair employment statute or the executive order governing government contractors, that not only prohibited discrimination but also commanded in so many terms "affirmative action"; the inclusion of these two words were deemed insufficient to alter or add to the basic obligations of the law. The issue posed by affirmative action programs was therefore one of permissibility, rather than obligation: Were these programs consistent with the prohibition against racial discrimination?

Sometimes the purported beneficiaries of the programs (or people speaking on their behalf) objected to them on the theory that the use of race was not wholly benign. Affirmative action was premised on the view that the racial minorities would not fare well under a colorblind policy, thus implying that these minorities are not as well equipped as whites to compete under traditional meritocratic criteria. They are being told, as they were under Jim Crow, that they are inferior—nothing "reverse" about this distinction. This complaint forced those who ran affirmative action programs to be secretive or discreet about what they were doing, but it did not bring those programs to an end or even present an especially formidable obstacle. The proponents of affirmative action explained that the race-based preference was premised on an assessment of the group's history in America, on the wrongs it suffered, not on a belief about innate ability, and as such could not justifiably be seen as giving rise to a slight. The use of race is benign, they insisted, because it improves the status of blacks and other racial minorities by giving them positions, jobs, or other concrete material advantages that they otherwise would not enjoy, at least not in the foreseeable future.

Affirmative action programs have also been attacked by whites, especially when there are discernible differences in the applicants under standard nonracial criteria and when scarce goods, such as highly desired jobs or places in professional schools, are being allocated. In such circumstances favoring a black because of his race necessarily means disfavoring a white because of his race; a job given to one is necessarily denied another. The rejected white applicant cannot truly claim that he is stigmatized even in these circumstances; no one is suggesting he is inferior. His exclusion comes as the by-product or consequence of a program founded on other principles—not to hurt him or the members of his group, but to help the disadvantaged. On the other hand, the rejected white applicant does not rest his complaint solely on the fortuity of the general, racially unspecific language of the antidiscrimination norm, the fact that discrimination based onany race is prohibited. The white applicant can also claim that he is being treated unfairly, since he is being judged on the basis of a criterion over which he has no control and which is unrelated to any conception of merit. The rejected white applicant might not be stigmatized, but he can insist that he is being treated unfairly.

This claim of individual unfairness finds support in the process theory of antidiscrimination: If the purpose of antidiscrimination law is to preserve the integrity of a process, to insure that individuals are treated fairly and to prevent them from being judged on the basis of irrelevant criteria, then it would not seem to matter whether the color used in the process were white or black. In either instance, the selection process would be unfair. The program may be well-intentioned, but the intention is of little solace to the rejected white applicants who, as Justice lewis f. powell put it, are being forced "to bear the burdens of redressing grievances not of their making."

Some of the proponents of affirmative action deny that there is any unfairness to the rejected white applicant. They argue that the claim of unfairness presupposes a special moral status for certain nonracial or meritocratic standards of evaluation, such as grades or performance on a standardized test, and that the requisite moral status is in fact lacking. The white has no "right" to be judged on the meritocratic standard. The more widely shared view among the proponents of affirmative action, however, acknowledges the unfairness caused to individual whites by the preference for blacks but treats it as a necessary, yet regrettable cost of eliminating caste structure. As Justice harry a. blackmun put it, "In order to get beyond race, we must first take account of race. There is no other way." Those who take this position, like those who support an effects test, argue that the purpose of antidiscrimination law is to guard against those practices that would perpetuate or aggravate the subordinate position of blacks and other racial minorities and that it would be a perversion of history now to use that law to stop programs designed to improve the status of these groups.

The Supreme Court confronted the issue of affirmative action and weighed these arguments in two different settings. In one, affirmative action was undertaken at the behest of a court order. The theory underlying such orders is not that affirmative action is directly required by an antidiscrimination statute or by the Constitution but rather that it is needed to remedy a pattern or practice of discrimination. Affirmative action is part of the court's corrective plan. A court might, for example, require a company to grant a preference in the seniority system to blacks who were previously excluded from the company and thus unable to earn seniority rights equal to those of whites. The Supreme Court has accepted such remedial uses of race, although it has insisted that this kind of preference be limited to identifiable victims of past discrimination and that some regard be given to the interests of the innocent whites who might be adversely affected by the preferences. For example, blacks might be preferred for vacancies, but will not necessarily be allowed to force the layoff of whites.

The second setting consists of the so-called voluntary affirmative action programs, which are adopted not under orders from a court but out of a sense of moral duty or a belief that the eradication of caste structure is a desirable social policy. These voluntary affirmative action programs have proved more troublesome than the remedial ones, in part because they are not limited to individually identifiable victims of past discriminations (they are truly group oriented), but also because they are not preceded by a judicial finding that the institution has previously discriminated and they are not carried out under the close supervision of a court. The Supreme Court approved these affirmative action programs, but its approval has not been a blanket one. By the mid-1980s, it was established that under certain circumstances color consciousness is permissible, but the Court has been divided in its effort to define or limit these circumstances.

These divisions have been especially pronounced when the voluntary programs were used in higher education. In the first case, defunis v. odegaard (1974), involving admissions to a state law school, the Court heard arguments and then dismissed the case on grounds of mootness because the rejected white applicant had graduated by the time the Court came to decide the case—a disposition that underscored the difficulty of the issue and the internal divisions on the Court. A few years later, the Court took up the issue again, in regents of university of california v. bakke (1978), this time at the insistence of a rejected white applicant to a state medical school. In this case the Court reached the merits, but the divisions were even more apparent. No single opinion commanded a majority.

Four Justices thought the preferential program in Bakke unlawful. They stressed an antidiscrimination statute, which prohibited, in so many terms, discrimination based on race. These Justices reasoned that a preference for blacks is as much a discrimination based on race as one for whites. No discrimination means no discrimination. Another Justice thought preferential programs could be justified as a means of diversifying the student body, but he objected to the manner in which the particular program before the Court had been implemented. He would allow race to be considered in the admissions process, but would not permit separate tracks for applicants according to race. The remaining four Justices joined in an opinion that would sustain the program as it was in fact implemented, but two of these Justices also wrote separate opinions.

These deep-seated divisions did not resolve themselves substantially in the years following Bakke. One voluntary program received a slightly more resolute acceptance by the Court, however, in fullilove v. klutznick (1980). This program was established by Congress and required a preference for minority-owned businesses in awarding contracts for federally funded public works projects. Although, once again, no single opinion commanded a majority of the Court, the vote of the Justices shifted from 1–4–4 to 6–3, and Chief Justice warren e. burger, who had objected (without qualification) to the preferential program in Bakke, voted to uphold this one. He also wrote one of three opinions that supported the constitutionality of the program. The Chief Justice studiously avoided choosing among "the formulas of analysis" articulated in Bakke; that is, he refused to say whether the affirmative action program had to meet the "compelling" interest standard or whether it was sufficient if the corrective ends of the program were deemed "important" or just "legitimate" and the means substantially related to those ends. He simply said, whatever the standard, this program meets it. He did, however, specifically and repeatedly mention one factor that might be the key to the change in his position and the Court's attitude in general: "Here we pass, not on a choice made by a single judge or a school board, but on a considered decision of the Congress and the President."

With this emphasis on the role played by the coordinate branches of government in the affirmative action program, the Chief Justice returned to an idea that emerged in the analysis of the Court's treatment of facially innocent criteria, and that might well explain the Court's determination to confine the effects test to statutes: The Court is more prepared to accept the costs and dislocations that are entailed in the eradication of caste structure when it can share the responsibility for this project with the other branches of government. The Court does not want to go it alone. This suggests that the fate of equality will depend not only on the substantive commitments of the Justices, on their determination to bring the subordination of blacks and other racial minorities to an end, but also on their views about the role of the Court. The content of antidiscrimination law will in good measure depend on the willingness of the Justices to use their power to lead the nation, or if that impulse is lacking, on the willingness of the other branches of government to participate aggressively in the reconstruction of a society disfigured by one century of slavery and another of Jim Crow.

Owen M. Fiss
(1986)

Bibliography

Bell, Derrick 1980 Race, Racism and American Law. Boston: Little, Brown.

Brest, Paul 1976 In Defense of the Antidiscrimination Principle. Harvard Law Review 90:1–54.

Cohen, Marshall; Nagel, Thomas; and Scanlon, Thomas 1977 Equality and Preferential Treatment. Princeton, N.J.: Princeton University Press.

Eisenberg, Theodore 1981 Civil Rights Legislation. Charlottesville, Va.: Michie Co.

Fishkin, James 1983 Justice, Equal Opportunity and the Family. New Haven, Conn.: Yale University Press.

Fiss, Owen 1971 A Theory of Fair Employment Laws. University of Chicago Law Review 38:235–314.

——1976 Groups and the Equal Protection Clause. Philosophy and Public Affairs 5:107–177.

Freeman, Alan 1978 Legitimizing Racial Discrimination through Antidiscrimination Law: A Critical Review of Supreme Court Doctrine. Minnesota Law Review 62:1049–1119.

Garet, Ronald 1983 Communality and Existence: The Rights of Groups. Southern California Law Review 56:1001–1075.

Gunther, Gerald 1972 In Search of Evolving Doctrine on a Changing Court: A Model for Newer Equal Protection. Harvard Law Review 86:1–48.

Karst, Kenneth and Horowitz, Harold 1974 Affirmative Action and Equal Protection. Virginia Law Review 60:955–974.

Rae, Douglas 1981 Equalities. Cambridge, Mass.: Harvard University Press.

Tussman, Joseph and Ten Broek, Jacobus 1949 The Equal Protection of the Laws. California Law Review 37:341–381.