Over the past three decades, the once-honorable aim of affirmative action—combating racial discrimination—has been replaced by its inverse. For most Americans, affirmative action now means not the combating of discrimination but rather its enforcement through a system of preferences. Whatever one’s judgment of the motives of those who engineered this historic reversal of aims, there is no doubt about the fact of it. Affirmative-action “plans” and “programs” are devices designed to give advantages and to impose handicaps according to racial, ethnic, and sexual categories.

Nowhere has affirmative action in this perverted sense had greater influence than in college and university admissions. Now, however, in an explosive decision of the United States Court of Appeals for the Fifth Circuit, Cheryl Hopwood v. State of Texas,¹ such preferences have at last been proscribed. If the Supreme Court upholds this decision on appeal, its impact upon the laws governing race-based programs in higher education will be very great, and this is a result long overdue.

_____________

Since 1978, the ruling Supreme Court precedent in this sphere has been Regents of the University of California v. Bakke.² In that case, Allan Bakke, a white man, was held to have been unlawfully disadvantaged in his application to the medical school of the University of California at Davis because of a racially discriminatory admissions system—a quota system—which favored nonwhites less qualified than he.

In Bakke, four members of the Court, led by Justice John Paul Stevens, found the admissions program at Davis an indubitable violation of Title VI of the 1964 Civil Rights Act.³ No constitutional issues need even be addressed, said they: the admissions system of the university violated the relevant statute and that was enough. Four other members of the Court, led by Justice William F. Brennan, argued that in view of the historical purpose of the equal-protection clause of the Fourteenth Amendment (“No State shall . . . deny to any person within its jurisdiction the equal protection of the laws”), which the university was claiming to advance through its quota system, constitutional issues were indeed unavoidable—and went on to find the Davis program acceptable.

The division between the two sets of Justices was as sharp as it was even, and between them no compromise was possible. Nor did Justice Lewis Powell, who cast the deciding vote, seek a compromise. In fact, he agreed completely with the Stevens group that the California system was unlawful and impermissible. But he also agreed with the Brennan group that to resolve the case, constitutional questions did need to be confronted—although he came out on the other side, concluding that the medical school’s admissions system violated not only the 1964 Civil Rights Act but the Constitution as well. And so Allan Bakke was ordered admitted, and Powell’s opinion became the judgment of the Court.

So far, so good. But Powell, fatefully, went a step further. He felt the need to adjust one portion of the earlier decision in this same case by the supreme court of California, which had ordered the university to eliminate all consideration of race in its admissions process. Believing that in this respect the California court had precluded too much, Justice Powell sought to draw a line close to theirs, but with more nuance. He observed that universities do well when they welcome a variety of viewpoints, and encourage robust intellectual exchange. And he posited that ethnic variety could in some cases advance that intellectual objective, and therefore that attention to race ought not to be completely proscribed.

Let us be clear. Never for a moment did Powell suggest that ethnic proportionality might be a legitimate goal in itself. On this point he could hardly have been more forceful: if the purpose of a university’s special-admissions program were “to assure within its student body some specified percentage [emphasis added] of a particular group merely because of its race or ethnic origin,” such a preferential purpose would be constitutionally invalid on its face. Only in pursuit of its proper intellectual business should a university be permitted to consider the race of applicants, along with other “diversity” factors, in making admissions decisions.

Thus it was that the concept of “diversity” first came to juridical prominence. In his worst dreams, Powell was not likely to have imagined the ways in which that concept would be exploited over the ensuing years. Because he was the only member of the majority giving an interpretation of the equal protection clause, and because his was the deciding vote, his opinion was widely taken to delineate the permissible constitutional use of race in university admissions. Although the particular affirmative-action program at the Davis medical school was condemned, affirmative action itself marched on victoriously, under the twin banners of Bakke and diversity. Today, diversity remains the shield of legitimacy for outright favoritism, the talisman with which institutionalized preference by race has been justified without (until Hopwood) apparent fear of constitutional attack.

_____________

Justice Powell was a man of moderation, and he was honest. But in allowing a consideration of race as, in some circumstances, a proxy for intellectual diversity, he made a very great blunder. It is wrong and unfair to suppose that being black or being white, or being any other color (or sex, or ethnic group), entails the possession of any given attitudes or judgments. In a society in which so many have suffered so painfully from stereotypes, the last thing needed was official stereotyping by the Supreme Court.

And Powell’s blunder had practical consequences that could not be prevented. He himself saw that universities, under the guise of seeking intellectual diversity, might continue to employ outright racial preference in admissions deviously. Yet he believed that if the rules were clear, honorable educators would not cheat. “A court would not assume,” Powell wrote, “that a university, professing to employ a nondiscriminatory admissions policy, would operate it as a cover for the functional equivalent of a quota system. In short, good faith would be presumed.”

But good faith was not the chief concern of those to whom this warning was addressed. Having been advised that they might weigh race along with other factors to achieve diversity, universities found irresistible the pressure to use race in ways going very far beyond the limits that Powell had drawn.

Thus, although the Constitution forbids the use of race to achieve targeted percentages, such uses are nearly universal now. Although the Supreme Court, in Bakke, explicitly disallowed wholesale compensation for “societal discrimination,” such compensatory awards are now commonplace. University admissions systems are, in fact, thoroughly pervaded by racial preference. The preferences given are not merely at the margins; they are not occasional or secondary; they are not simply “plus factors” introduced when the records of applicants are otherwise nearly indistinguishable. Rather, the preferences given are very substantial; they involve the identification and primary sorting of all applicants by ethnic group; and they are carefully designed to achieve predetermined numerical objectives.

In clear if tacit acknowledgment that such programs are, in fact, legally and morally suspect, they are for the most part kept hidden. In most universities, discretionary authority, granted by the administration to admissions officers, obscures much of what goes on. Documents in which preferences are discussed or revealed are typically marked “confidential, Internal Use Only”—and can usually be obtained, if at all, only with great difficulty.⁴

Published descriptions of these admissions practices invariably invoke language that carefully mirrors the phrasing of Powell’s passages on diversity. At the University of Michigan, for example, administrators insist that “we consider race along with a range of other factors.” This is technically true, but it is also a deception. While other factors are considered for some applicants, only race serves as the threshold consideration in terms of which all applicants are first reviewed, and by which offers of admission are portioned out.

Admissions officers also routinely stress that the attainments and characteristics of applicants as individuals are carefully weighed: leadership qualities, social concern as exhibited by community activities, the capacity to overcome adversity, and other nonquantifiable virtues and achievements. They are right to consider such merits; neither undergraduate nor professional-school admissions ought to be determined exclusively by test scores or grade-point averages (GPA’s). Although intellectual qualities should be of the first importance, applicants do sometimes display nonintellectual credits that are so unusual as to justify what would otherwise be anomalous action. Everyone understands this.

But such nonquantifiable considerations cannot account for a pattern of racially distorted outcomes. After all, applicants with special talents, needs, or achievements will exist in every group. Altruism, evidence of handicaps overcome, dedication to one’s community, and the like are not found disproportionately among members of any one race or nationality. While such non-numerical factors may fairly count in individual cases, they cannot explain systematic bias by race.

_____________

Yet there is systematic bias by race, and this can be demonstrated beyond doubt. Anyone looking for proof is well advised to examine the data assembled in the Autumn 1995 issue of the Journal of Blacks in Higher Education. The editors of that journal had requested admissions figures from 25 highly esteemed universities and another 25 highly esteemed liberal-arts colleges. Each was asked: what percentage of all students applying to your institution is accepted? And then, what percentage of black applicants is accepted?

Among the universities, 10 of the 25—including Harvard, Princeton, Yale, Stanford, Duke, and Columbia—refused to supply the data requested. Of the 15 that did respond, 12 revealed that acceptance rates for blacks are (in the words of the Journal) “significantly higher than acceptance rates for whites.” At the University of Virginia, for example, 54.2 percent of black applicants were accepted in 1995, as opposed to an overall acceptance rate of 36.6 percent. At Rice, the acceptance rate for blacks in the same year was 51.7 percent, while that for whites was less than half that figure. This pattern is nearly universal.

Among the liberal-arts colleges approached by the Journal, the figures show even greater partiality. Amherst, for example, reported an overall acceptance rate of 19.2 percent in 1995, but its acceptance rate for black applicants was 51.1 percent. For Bowdoin, the figures were 30.4 percent and 70.2 percent. And so on. The director of admissions at Williams, declining to respond, said gingerly: “We think these figures should remain in-house because of the current anger over affirmative action”; one can easily imagine what the figures are.

What the Journal of Blacks in Higher Education showed for the institutions it surveyed is replicated very widely. Take, for example, two large state institutions, the Universities of Michigan (UM) and California (UC), both outstanding centers of liberal-arts and professional studies.

Universities commonly prepare, for internal use, a table or grid reporting both applications and offers of admission by category. At UM this grid is divided for undergraduate applications into 108 categories, or cells, delineated by “former-school GPA” on the vertical axis, and “best test score” (SAT or ACT) on the horizontal axis. In each cell appear the number of applicants in that category, and the number offered admission. Separate grids in identical form are prepared for all students and for “underrepresented minorities.”

In the UM profile of 1994 (the latest available), the minority-admission rate was higher than the nonminority-admission rate in almost every cell in which there were any minority applicants at all, and in many cells the rate was very much higher. Thus, for applicants with GPA’s between 2.80 and 2.99 (B-), and SAT scores between 1200 and 1290 (out of 1600), the nonminority-admission rate was 12 percent and the minority rate 100 percent. For applicants whose GPA was between 3.40 and 3.60 (B+), and whose SAT scores were 900-990, the nonminority rate was 13 percent and the minority rate was 98 percent. For applicants with GPA’s between 3.60 and 3.79 (A-), and SAT scores 800-890, the nonminority rate was 12 percent and the minority rate was 100 percent. And so on. Similar patterns are disclosed by the reports of the University of Michigan’s law and medical schools.

At the University of California, the admission grids of the law schools and medical schools, pried loose only by repeated Freedom-of-Information-Act requests, disclose preferences so gross that the Regents understandably concluded they could not hope to defend the system in court. Thus, the UC Berkeley School of Law offered admission to every black applicant with both a GPA of 3.5 or above and a Law School Aptitude Test (LSAT) percentile score of 90 or above; of all the white applicants with the same records, only 42 percent were offered admission. Again, in the GPA range 3.25-3.49, with LSAT percentile scores of 70-74.9, all black applicants were admitted, while not one of the 34 white applicants or 12 Asian applicants was offered a place. The story is essentially the same at Davis, at UCLA (where in one cell the acceptance rate for blacks was 61 times higher than the acceptance rate for whites), and throughout the UC system, both in law schools and in medical schools.

Here is the plain truth of the matter: if you are a B or a C student, and if you are white or Asian, you would be wasting your time and your money (there are substantial application fees) applying to the University of California law or medical schools. But if you have the same record and are black or Hispanic, your chances for admission to one of these schools are vastly better. And whatever your academic record may be, if you are black or Hispanic, you have a very significant advantage in the competition for prized admission slots.

One would expect the acceptance rate for a given group to be higher if its academic performance were stronger; but, as everyone knows, performance does not explain current admissions disparities. In fact, the performance of minority groups accepted at greatly higher rates is significantly weaker, not stronger, than that of others—and in many cases it is weaker than that of others who have been rejected.

All this, now proved beyond cavil, has been repeatedly and publicly denied by the University of California, as it is denied by the University of Michigan. Both universities proudly proclaim that they do not discriminate on the basis of “race, color, or national origin.” Both universities are lying. And they are hardly alone.

_____________

And so to the University of Texas. Like all good law schools, Texas had established presumptive score levels for admission and for denial of admission. But these presumptive levels varied dramatically by race.

The appellate court that heard the Hopwood case was nearly incredulous. Here is its report of the way the system worked:

[B]y March of 1992, because the presumptive denial score for whites was a TI [Texas Index, a composite of undergraduate GPA and LSAT scores] of 192 or lower, and the presumptive admit for minorities was 189 or higher, a minority candidate with a TI of 189 or above almost certainly would be admitted, even though his score was considerably below the level at which a white candidate almost certainly would be rejected. Out of the pool of resident applicants who fell within this range . . . , 100 percent of blacks and 90 percent of Mexican Americans, but only 6 percent of whites, were offered admission.⁵

Scrutinizing these racial preferences, the Fifth Circuit Court concluded that the admissions system at Texas plainly violated the equal-protection clause. All the defenses put forward by the state of Texas—and most emphatically the goal of achieving a diverse student body—were rejected unconditionally.

Universities around the country have found the Hopwood decision shocking. They have reviled the court, labeling its decision “ignorant,” “unequivocally mistaken,” and “cavalier.” But the reasoning of that decision was rigorous and sound. The court proceeded first by determining the standard by which the university must be judged, and then by applying that standard to the law school’s admission practices.

The first of these steps was easy: race-based programs are always suspect, always invidious. Therefore, by long-settled law, the standard by which such programs must be judged is that of “strict scrutiny”: to be justifiable, any racially preferential device must be shown to serve a compelling government interest, and then must be shown to have been narrowly tailored to satisfy that interest.⁶

Are there any interests so compelling as to justify governmental uses of racial classifications? Yes, there are. Persons found to have been unlawfully discriminated against because of race are surely entitled to remedy. And race, once having been proved to be an element in the wrong done, may also be an unavoidable element in ameliorating the harm. The burden upon those who would so use race is very heavy, as it ought to be—but it is a burden that may, in some special circumstances, be sustained.

Needless to say, the Texas law school sought to persuade the court that the admissions preferences it gave fell into this class. Texas’s arguments boiled down to three, all rejected by the court.

First, the injuries for which the preferences were, allegedly, a remedy had been inflicted in part by the public schools of the state of Texas, which (the law school lamented) have historically discriminated against blacks from the elementary grades through secondary schools.

Yet, as the court pointed out, if the law school were permitted to “remedy” the present effects of past discrimination in Texas public schools, nothing could preclude racial preferences by any state agency in hiring, or in government contracts, or in any other state activity affected by the educational attainments of applicants. Moreover, the Supreme Court itself has explicitly rejected the use of past “societal discrimination” to justify race-based remedial programs, on the ground that it is too amorphous a category. For there is no way of determining the present effects of earlier discrimination in primary and secondary schools, or their magnitude, and therefore no way of determining what might be an appropriate antidote for those alleged effects.

In any case, the preferences in the law school’s admissions were very plainly not designed to offer any such remedy. They were conferred equally on applicants who attended public and private schools, and equally upon minorities whose schooling had been in Texas and out of Texas. It was simply not true that earlier wrongful conduct by the state of Texas had been the impetus behind the preferences.

_____________

This brings us to the school’s second defense. In the words of the decision, the standard of strict scrutiny would be satisfied, a compelling need for this “remedy” shown, only if the state could identify the law school itself, and not some other agency or institution, “as the relevant alleged past discriminator,” and then only if it could prove that the “present effects of [its] past discrimination [were] of the type that justify the racial classifications at issue.” And so the school maintained that the preferences were indeed a remedy for its own earlier misconduct (as well as that of the Texas schools); after all, the law school had discriminated against blacks in the past, and even though such practices were eliminated decades ago, that earlier discrimination still had present effects.

To which the court responded that there was no way for Texas to measure, or even rationally to estimate, the present effects of racial discrimination imposed by the law school in earlier generations, and therefore no way to know whether or if current admissions preferences addressed those effects.

A third try amounted to public chest-beating. We have been so bad, said the law school, that our badness is everywhere known and despised. We have, it asserted, a “lingering reputation in the minority community, particularly with prospective students, as a ‘white’ school . . . and [there is] some perception that the law school is a hostile environment for minorities.”

But these effects, too, even if substantiated, could not possibly justify the remedial use of race in the admissions process. As a different federal circuit court put it in a 1994 case involving the University of Maryland,

mere knowledge of historical fact is not the kind of present effect that can justify any race-exclusive remedy. If it were otherwise, as long as there are people who have access to history books, there will be programs such as this.

Nor, more fundamentally, could the allegedly hostile reputation be substantiated at Texas. There, programs designed to attract minorities were first implemented in the 1960’s; minority recruiting has long been vigorous, and substantial amounts of money are spent to advance it. Most faculty, staff, and students have had absolutely nothing to do with any discrimination that the school practiced decades ago, and if there are hostile relations among the races on the campus today, they simply cannot be tied to past discrimination by the school.

The upshot: “[T]he law school has failed to show a compelling state interest in remedying the present effect of past discrimination sufficient to maintain the use of race in its admissions system.”⁷

_____________

For the law school, that left the claim of intellectual diversity as its only hope. But that, too, turned out to be no hope at all. Powell himself had made it crystal clear in 1978 that the goal of intellectual diversity cannot possibly justify systematic preferences aimed at ethnic balance. But the Hopwood court went further. Attending to recent and carefully formulated judgments of the Supreme Court, it found the entire diversity caper to be misguided.

For one thing, the court noted, the word “diversity” appears nowhere in the Bakke case except in Justice Powell’s discourse on the topic. Not one other Justice, then or since, has concurred with his views of diversity as a constitutional interest that might support some use of racial classification. Even the Justices who supported the university’s admissions program in Bakke suggested in their dissent that the quest for “an integrated student body” could not by itself serve as justification for preference. Powell’s opinion alone therefore cannot bind lower courts, and may not be supposed to lay down the constitutional rules governing the way race can be used.

For another thing, the Supreme Court has clearly and repeatedly explained what can and what cannot serve as a justification for race-based government action under the Constitution. Its decisions, which are binding, explicitly exclude the quest for diversity. As four Justices of the present Court (Sandra Day O’Connor, William H. Rehnquist, Antonin Scalia, and Anthony Kennedy) put it in a 1990 case involving broadcast media:

Modern equal protection has recognized only one [compelling state] interest: remedying the effects of racial discrimination. The interest in increasing the diversity of broadcast viewpoints is clearly not a compelling interest. It is simply too amorphous, too insubstantial, and too unrelated to any legitimate basis for employing racial classifications.⁸

Finally, the Texas court observed, the use of diversity to justify race-based programs does not promote, it actually undermines, the goals of the Fourteenth Amendment. It enhances racial stereotypes, fuels racial hostilities, fosters racial divisions. To strive for an entering class displaying different skin colors, said the court, “is no more rational on its own terms than would be choices based upon the physical size or blood type of applicants.” In fact, using race in this manner “exemplifies, encourages, and legitimizes the mode of thought and behavior that underlies most prejudice and bigotry in modern America.”

_____________

Such truths have been a long time coming from the American judiciary, and are all the more welcome for that. Does this mean, then, that the days of racial preference in university admissions may be coming to an end? Yes, if the Supreme Court accepts Hopwood on appeal and affirms it. (That the Supreme Court will indeed hear this case seems to be the presumption of the circuit court, which in late April granted a stay of its ruling to give Texas officials time to file their appeal.) Alternatively, the Court may find it more prudent to rely upon the accumulation of circuit-court opinions, buttressed by legislation, to resolve the matter without direct intervention.

Such legislation is, in any case, pending in many jurisdictions. In Congress, the Equal Opportunity Act of 1995, if adopted, would bar the federal government from giving any preferences by race, or obliging or encouraging others to do so. Similar statutes are in the works in Pennsylvania, Michigan, and now Arizona. In South Carolina and Michigan, efforts are also under way to enact constitutional amendments for-bidding preferences. And, of course, in California the Civil Rights Initiative will be on the ballot this fall and has a good chance of success.⁹

Some universities are beginning to clean house on their own initiative, or as a result of prodding by public officials. At the University of California, affirmative action has already been drastically remolded; in Colorado, the state-university system has changed the rules so as to ensure that race can be, at most, one of several criteria considered; and in Georgia, where the attorney general has formally recommended that all race-based admissions and financial-aid policies be abandoned, state colleges and universities are likely to follow suit.

But the sailing will not be smooth everywhere. In Bakke, Justice Powell expressed his confidence in the good faith of university officers, a confidence their conduct has shown to have been misplaced. The Hopwood court was less trusting. Mindful of likely efforts to evade the force of its stricture, it gave fair warning:

[I]f the [Texas] law school continues to operate a disguised or overt racial classification system in the future, its actors could be subject to actual and punitive damages.

Whatever happens at the University of Texas, in other places the push to go on doing covertly what the law forbids may continue until evasion is made too costly. As the president of the University of Michigan put it defiantly, “If certain avenues are shut off, we’ll try to find other ways to get the same result.”

_____________

Still, There can be no doubt that the Hopwood decision represents, to quote a headline in the Chronicle of Higher Education, “A Stunning Blow to Affirmative Action”—at least as affirmative action has come to be defined. So let us suppose that we will indeed live to see the elimination of racial preferences in university admissions. What then? In particular, what would be the impact of such a development on the numbers of minority students on campus?

The answer, painful though it may be, is clear. To the extent that minority enrollments have depended upon the special preferences given to favored groups, those numbers would fall, at least temporarily. (As Thomas Sowell has pointed out, however, there might well be ancillary gains, as students turned down by elite institutions turn to the second- and third-rank schools which they are better qualified to attend and which would benefit from their presence.) But in any case, racial justice does not entail racial proportionality, and never did. Public policy cannot be judged, and it certainly cannot be advanced, by measuring the degree to which its results approximate some predetermined numerical outcome; if we can bear that in mind, we may yet arrive at the blessed day when we will no longer count by race in any public sphere.

The elimination of preferences in universities would have another worthy consequence as well. Once the residual impact of preferences has dissipated, we may live to see an abatement in the nasty stereotypes of inferiority which those preferences inevitably engendered and just as inevitably reinforced. The cruel supposition that all blacks and other minorities occupy their places by dint of charity may ultimately pass away. Racial canards, unspoken but widely believed, may lose their claims to credibility.

Relations between the races on our campuses today are as bad as they have ever been. Only when racial preferences come to an end, with all the mischief they have done, will we begin to see a healing of the wounds we have so heedlessly inflicted on ourselves.

¹ March 18, 1996, No. 94-50569.

² 438 U.S. 265. A very similar 1974 case involving the University of Washington law school, DeFunis v. Odegaard, was held moot in the end, and made no law.

³ “No person in the United States shall, on the grounds of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving federal financial assistance.” Virtually all American universities receive some federal financial assistance and are therefore governed by this title.

⁴ Even those who govern the university may sometimes be kept in the dark. At the University of California, where exposure of such practices recently led the Regents to eliminate the most egregious examples of preference, one member of the Board, Ward Connerly, wrote to another: “I came to the conclusion that we are breaking the law. There is no other way to put it. We are breaking the law!”

⁵ Hopwood pp. 8-9.

⁶ Adarand Constructors v. Peña, a 1995 Supreme Court decision, reaffirmed this standard, and made it clear that it applies to programs at all levels, including the federal government.

⁷ Since no remedial interest was shown, and no other interest was accepted as compelling, the court held it unnecessary even to address the question of whether the program was indeed narrowly tailored, as the strict-scrutiny standard requires. But the federal district court, hearing the case at an earlier date, had addressed that matter as well and held that, even supposing the remedial interests genuine, the law school's admission program was certainly not narrowly tailored to address them. So the school flunked both tests.

⁸ Metro Broadcasting v. FCC. This passage was written as part of a dissenting opinion, but the same reasoning lay at the core of the majority opinion that overturned Metro Broadcasting in 1995.

⁹ The one-sentence operative paragraph of the California initiative reads in full: “The state shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting.” This wording differs from that of the Civil Rights Act of 1964 only in the addition of the phrase: “or grant preferential treatment to.” At issue in the California Civil Rights Initiative, therefore, is not “affirmative action” in its original intent—of which no mention appears—but only preferential treatment.