Race Preference and the Universities

A Final Reckoning?

By Carl Cohen

Commentary Magazine, September 8, 2001

PREFERENCE BY race in university admissions is a widespread practice in the United States. Although the Supreme Court has condemned racial preferences in other contexts, it has addressed university admissions only once, in the 1978 case of Regents of the University of California v. Bakke.(1) Soon, however, the matter is likely to come before the Court again.

Two cases involving the University of Michigan–called, after their lead plaintiffs, Gratz and Grutter–have already been decided at the lowest judicial level. Both will be argued on appeal this autumn in the Sixth Circuit, and are then virtually certain to be appealed to the highest court. The outcome is bound to prove momentous. But at the present stage, the two cases are in sharp conflict, and this conflict opens a rich field of argument.

Gratz concerns admission to the undergraduate college of the University of Michigan, Grutter to the law school. In both cases, the respondents are the regents of the university; its president, Lee Bollinger; and other officers. Although the cases are distinct, the essential issue they present is the same: did the University of Michigan, which concedes giving preference in its admissions decisions to some racial minorities, violate the constitutional rights of applicants who were not preferred, like Jennifer Gratz and Barbara Grutter?

For their part, the plaintiffs do not contend they had a right to admission, but rather that the racial distortion of the admissions process denied them a fair review.(2) They also claim that the university broke the law, because Title VI of the Civil Rights Act of 1964 is explicit in forbidding discrimination by race or color or national origin by institutions (like the University of Michigan) receiving federal financial assistance. For its part, the university does not deny that it weighs the race of applicants in its admissions process, but contends that its uses of race are lawful and constitutional. Two acute and knowledgeable federal judges–Patrick J. Duggan in Gratz and Bernard Friedman in Grutter–have issued the irreconcilable opinions that I will explore below.

FIRST THE facts. From 1995 to 1998, the University of Michigan sorted all applicants for undergraduate admission into “cells,” using grids marked on the vertical axis with grade-point averages (GPA’s) and on the horizontal axis with test scores (SAT or ACT). Admissions officers were instructed to give standard ratings to applicants in each cell; but not everybody in the same cell received the same rating. In cell after cell, applicants with a given set of academic credentials were rejected if they were white but accepted if they were from a targeted minority–African American, Hispanic American, or Native American. Overall, admission rates for these minorities were much higher than admission rates for white and Asian applicants.

These and many other facts became public in 1996, when the university documents revealing them were reluctantly delivered up in response to a request under the Michigan Freedom of Information Act (FOIA).(3) For years, the university had explicitly denied giving racial preferences, but by the time the present cases were filed in 1997, the reality was no longer in dispute. Then, in 1998, the university changed the mechanics of its undergraduate-admissions system to eliminate the embarrassing grids while continuing to give preference in other ways. The newer system, still in force, is based on point totals: out of a possible 150, 100 normally suffice for admission, with minority applicants receiving twenty points automatically.(4)

In the undergraduate case, Gratz, both parties sought summary judgment, and in December of last year Judge Duggan granted it in two very different parts. The older, pre-1998 system he found to be a violation of the equal-protection clause of the 14th Amendment; because Jennifer Gratz and the class of plaintiffs she represented had applied for admission during the years that older system was in force, they won their suit. But with respect to the current preferential system, Judge Duggan granted summary judgment to the university. This system, he concluded, crosses the “thin line” into permissibility because, unlike the older system with its “practice of ‘protecting’ or ‘reserving’ seats for underrepresented minority members,” the newer one “does not utilize rigid quotas or seek to admit a predetermined number of minority students.”

I will return to Judge Duggan’s defense of the new system, but here let me observe in passing that his distinction between it and the old system is highly problematic. Again and again, the university itself asserted flatly that the new system “changed only the mechanics, not the substance” of admissions practices–which makes perfect sense, since the objective was to achieve the same results. As the Sixth Circuit court of appeals noted in another case, “quotas and preferences are easily transformed from one into the other.” Any quota can be satisfied (the court observed) by adding to the score of each minority applicant some number of points calculated to yield the desired outcome.

The University of Michigan did precisely that. As sworn depositions revealed, the number of points awarded for minority status in the new system was decided upon by determining statistically how many points would be needed to insure the same number of minority admissions as the old system! If the fatal flaw of the old system was that it protected a fixed number of seats for minorities, the new system is undeniably flawed in the same way.

Nor does the use of points in place of grids make a significant difference. Any preferential system, including the one now in force, can be exhibited on a grid, which is only an instrument for showing what is being done. The university, embarrassed by what its earlier grids had plainly revealed, made but a cosmetic change. If preference given for skin color is wrong, it is wrong whether highlighted on a grid or obscured by placing the racial factor low on a list.

WHICH BRINGS us to the law-school case, Grutter. There the university also admitted that race was commonly considered in admissions; what was chiefly in dispute was the weight given to race.

The Michigan law school is highly competitive; thousands apply each year for some 380 seats. Using undergraduate grade-point averages, and LSAT rather than SAT scores, the law school charted admissions by race from day to day. Each cell in these charts (also obtained through a FOIA request) disclosed the number of applicants and the number of acceptances in each category of academic achievement. Separate charts were prepared by the school for “Caucasian Americans” and for “African Americans” and other minorities. The admission rate for white applicants in many cells was 2 or 3 percent while–in those same cells–the admission rate for blacks was 100 percent. Statistical analysis showed that, for minority students, the odds of admission were hundreds of times greater than the odds for majority students with the same academic credentials.

None of this could be denied. But how the number of minority students was decided, and whether numerical targets were in fact used, were matters hotly contested at the trial ordered by Judge Bernard Friedman. Deans and admissions officers of the law school were examined and cross-examined, minutes of faculty meetings and other records were analyzed scrupulously. The university repeatedly sought to assure the court that although race was indeed considered, it was merely one minor factor among many. But, with awkward inconsistency, administrators were also anxious to persuade the court that, if they were not permitted to continue to use race in admitting students, the Michigan law school would be devastated. Thus was confronted what may be called (after the university’s president) Bollinger’s dilemma: if the elimination of racial preferences would indeed be devastating, race could not have been a minor factor; if it really were a minor factor, the elimination of preferences could not be devastating. The law school wanted it both ways.

University officials found cross-examination at this trial greatly discomfiting. They were obliged to duck and dodge the questions seeking to determine what the target percentage of minority students had been. Law-school officers expressed abhorrence of any “quota,” denied using any target numbers, and insisted repeatedly that they sought no more than a “critical mass” of minority students. But how did they know when they had achieved a critical mass? Was there really no target number at all? How did they know when they had too few? Responses were doggedly evasive, but the university could not escape disclosing, through its own documents and in answer to questioning, that it had indeed always found its critical mass at the point where minority admissions reached 10 to 12 percent of the entering class, or more. The regularity with which minority enrollments over recent years fell neatly into that range was quite remarkable.

What was really going on in the Michigan law school could not be hidden, and proved highly embarrassing for a great university. Judge Friedman, in his findings of fact, put the inescapable conclusion bluntly: the system of racial preferences at the Michigan law school, he said, was “practically indistinguishable from a quota.”(5)

THE CONFLICTING arguments in Gratz and Grutter are as interesting as they are important. District courts and circuit courts are bound by principles laid down in earlier cases by the Supreme Court, and on the uses of race in college admissions the Supreme Court has issued, as I remarked earlier, one landmark decision: Bakke. Just like Jennifer Gratz and Barbara Grutter, Allan Bakke, a splendid applicant twice rejected in the 1970’s by the medical school of the University of California at Davis, had argued that the review of his application was distorted by racially discriminatory practices. He won in California’s highest court, and again when the case was appealed to the Supreme Court. Bakke’s admission was ordered, and he completed medical school there.

Nevertheless, the Bakke decision has given rise–with reason–to great and continuing uncertainty. The case plainly forbids practices like those of the University of California in the 1970’s the 1970’s. But its principles, and how they are to be applied to admissions programs like those at Michigan today, are questions not easy to answer. What Bakke permits and does not permit remains highly controversial.

One critical rule of law in this arena is clear and well settled: racial classifications, in view of their cruel and invidious history, are always suspect. Their use–although not absolutely excluded–must be very solidly justified; any government program that treats different races differently will be subject to a very high standard of review, called the standard of “strict scrutiny.” This standard in turn has two prongs: any use of racial classifications by an arm of government must be shown both (a) to serve a compelling state interest, and (b) to be “narrowly tailored” to serve that compelling interest.

Why so high a standard should be imposed is not hard to see. In light of the ways our government has used race in generations past, no government agency can be trusted to distribute benefits to one racial group at the expense of another without reasons that are demonstrably overriding in force. The University of Michigan is a public university, an arm of the state of Michigan. It therefore must sustain the heavy burden of identifying the state interest that justifies its admitted use of racial classifications. When it favors applicants of some races and disfavors applicants of other races, what compelling need is served?

Ordinary citizens often suppose that the justification for preferences is the desire to give redress for the cruel damage done in the past. It is time, the argument runs, to “level the playing field” by compensating for the “societal discrimination” that has long oppressed racial minorities in this country. We may call this the “compensatory defense” of preferences, and it was put forcefully (if unsuccessfully) by the University of California in Bakke.

But there is another defense of racial preferences, built upon the conviction that such a system enriches the educational experience of all. This is the so-called “diversity defense.” Although most people would consider it not nearly so strong as the compensatory defense, it came to be relied upon after Bakke largely because, in that case, Justice Lewis Powell had found diversity to be sufficiently important in university education that for some applicants in some circumstances it could be a compelling state interest supporting preferences.

As a strategic issue, the University of Michigan had to decide which of these two arguments would have the best chance of success. Its decision was to put all of its eggs in the diversity basket.(6) It adopted this strategy because it wanted to win, and because it knew with virtual certainty that it could not win by relying on a compensatory defense.

The reasons for this are not widely understood. In Bakke, four of the nine Justices, led by William F. Brennan, would have accepted compensation for past injury as the needed state interest (under a standard less strict than what is now universally agreed to be necessary). Four other Justices, led by John Paul Stevens, thought the California preferences an obvious violation of the Civil Rights Act of 1964, and certainly not defensible on compensatory grounds. (The statutory prohibition, wrote Stevens, was “crystal clear: race cannot be the basis of excluding anyone from participation in a federally funded program.”) So Powell stood alone as the ninth and deciding voice.

POWELL WAS a man of moderation. In his Bakke opinion, he condemned systematic racial preferences vigorously and repeatedly.(7) But, in deciding between closely matched applicants, he thought it would not be unreasonable for an admissions officer to weigh the contributions that particular applicants might make to the richness of the learning environment. The only way to permit such discretion, however, would be by softening the absolute prohibition of racial considerations that had been ordered by California’s highest court.

The language with which Powell sought to effect the needed softening has been the source of untold confusion. Because, he wrote, a diverse student body contributes to the “robust debate” protected by the First Amendment, diversity might be a compelling interest–a constitutional interest–in some cases, thus warranting what Powell referred to as a “plus factor” for certain individual applicants. These few words about the possible role of “diversity” became, and remain, the bulwark of those seeking to protect racial preferences in admissions.

How widely Powell’s view was shared by other Supreme Court Justices in Bakke is an interesting question, but before getting to it I want to emphasize something else: namely, Powell’s categorical rejection of the compensatory defense that had been strongly urged by the University of California. Powell’s reasons for rejecting it have been reaffirmed by courts over the years. Their ineluctable force which is what compelled the University of Michigan to build its argument on the alleged need for “diversity”–is rooted in the certain fact that no university is competent to make the findings and frame the remedies that just compensation might require. In Powell’s words:

“We have never approved a classification that aids persons perceived as members of relatively victimized groups at the expense of other innocent individuals in the absence of judicial, legislative, or administrative findings of constitutional or statutory violations… Without such findings of constitutional or statutory violations, it cannot be said that the government has any greater interest in helping one individual than in refraining from harming another. Thus, the government has no compelling justification for inflicting such harm.

Petitioner [the University of California] does not purport to have made, and is in no position to make, such findings. Its broad mission is education, not the formulation of any legislative policy or the adjudication of any particular claims of illegality… Isolated segments of our vast governmental structures are not competent to make those decisions…

Hence, the purpose of helping certain groups whom the faculty perceived as victims of “societal discrimination” does not justify a classification that imposes disadvantages upon persons like respondent [Allan Bakke] who bear no responsibility for whatever harm the beneficiaries of the special-admissions program are thought to have suffered. To hold otherwise would be to convert a remedy heretofore reserved for violations of legal rights into a privilege that all institutions throughout the nation could grant at their pleasure to whatever groups are perceived as victims of societal discrimination. That is a step we have never approved.”

Nor will any court ever approve such a privilege under our Constitution, for once the authority to make such determinations is given to middle-level bureaucrats working behind closed doors, as admissions officers at most universities do, it would be uncontrollable, and inevitably abused.(8) So the University of Michigan will never be permitted to justify giving preference to minorities on the ground that it, in its wisdom, has decided such preferences are an appropriate remedy for oppression it concludes has been long suffered. This being very clear, nothing remains for the university save the argument from “diversity.” There is no other viable path.

AN INTELLECTUALLY diverse environment is a value every state university is wise to promote; no one is likely to quarrel with that. But the claim that ethnic preferences improve learning and teaching is an empirical one, and is disputable. And even if the claim could somehow be proved true, is diversity so good that it rightly justifies discrimination by race?

Let me deal with the second issue first, and with its moral aspect before its legal one. Suppose you have been shown strong evidence that segregated classrooms improved learning and teaching. Suppose that the data in support of segregation were very impressive, far more impressive than the materials offered by the University of Michigan in support of the alleged benefits of diversity. Would we think that constituted a justification for deliberate racial segregation? To the contrary: we would say that segregation by race, imposed by the state, is wrong, and any advantages that might flow from it could not begin to justify a policy that is intrinsically immoral and unjust.

That is what racial preference is. As it happens, the praises of diversity sung by the university are a tissue of exaggerated claims; but even if they had substantial merit, in a decent community they would carry no weight. Racial discrimination imposed by the state is despicable, and no studies aiming to persuade us of its benefits can make it less so.

But moral judgments will not decide the two Michigan cases–at least until they reach the Supreme Court. The legal issue that must be resolved in deciding them is this: under the standard of strict scrutiny laid down by the Supreme Court, can diversity serve as the “compelling state interest” that may justify the use of racial classifications by a government agency? This is the key question presented by Gratz and Grutter. Let me take each case in turn.

In Gratz, Judge Duggan contended that the Supreme Court in Bakke held the educational benefit of diversity to be a compelling government interest justifying the use of race as a factor in admissions. If that is so, five Justices of the Court must have expressed agreement with that proposition. But of the six opinions rendered in that case, only one, Justice Powell’s, so much as mentioned diversity; the eight other Justices had nothing to say about it.

True, concedes Duggan; but we must look beneath the surface. Four Justices (Brennan, White, Marshall, and Blackmun) supported racial preferences in Bakke, and sought to protect the university’s authority to give them. When one grasps the spirit of their decision, says Duggan, one sees also that they would have supported the diversity defense if doing so would have saved the preferences. Note that Justice Brennan, in a later case, described Bakke as “recognizing” that “a ‘diverse student body’ contributing to ‘a robust exchange of ideas’ is a ‘constitutionally permissible goal’ on which a race-conscious university admissions program may be predicated.” And the Brennan four certainly agreed with Powell that there are circumstances under which race may be appropriately used. This gives us enough (Judge Duggan contends) to resolve the uncertainty in favor of the diversity principle.

As both Judge Duggan and the university have pointed out with satisfaction, such a conclusion is also confirmed by one critical passage in Powell’s opinion that was joined by Brennan, Marshall, White, and Blackmun. All five agreed in saying this:

“In enjoining [the University of California] from ever considering the race of any applicant, however, the [lower] courts … failed to recognize that the state has a substantial interest that legitimately may be served by a properly devised admissions program involving the competitive consideration of race and ethnic origin.”

If five Justices together held unequivocally that a substantial government interest could be served by race-based admissions, as they clearly did, we may conclude, says Duggan, that the Supreme Court has indeed spoken definitively on this matter. This conclusion is given further support, he thinks, by the way the two rationales–compensation and diversity–are related to each other. When a fragmented court comes to a result that nevertheless gets majority support, there is an established method for determining the governing principle. In such cases, we ask which of the grounds proposed is the narrowest, the one that goes least far. In this case, asserts Duggan, that narrower justification is diversity. Hence, using this interpretive principle, we may conclude that, in spite of the fragmentation on the Bakke court, a single holding can be derived from its decision: namely, that diversity may be a compelling need.

To support these views, finally, Judge Duggan quotes a decision by the court of appeals of the Ninth Circuit (covering California and eight other Western states):

“True it is that Justice Brennan did not specifically say that “race” could be used to achieve student-body diversity in the absence of any societal discrimination, but … we can hardly doubt that he would have embraced that somewhat narrower principle if need be, for he thought that it was simply an allotrope of the principle he was propounding.”

Taken together, these considerations constitute the very best defense that can be made for the claim that the current admissions practices of the University of Michigan can be defended under Bakke by the compelling need for diversity in college classes.

ADDRESSING THE same issue three months later in Grutter, Judge Bernard Friedman reached the very opposite conclusion. Each of the considerations registered by Duggan was weighed, and rejected.

First, according to Friedman, we may not consider the diversity defense to be a holding of the Supreme Court simply on the ground that four Justices who did not in fact support it may be supposed to have supported it. We know precisely what those four Justices did: they deliberately refrained from joining Justice Powell in any of his remarks about diversity.

Justice Powell’s opinion, Friedman points out, had many parts. He introduced and explained his diversity rationale in Part IV-D, where he concluded that “the attainment of a diverse student body … clearly is a constitutionally permissible goal for an institution of higher education.” Parts V-A and V-B also concern diversity; in them, Powell showed that the particular system of racial preferences in question at Davis could not be justified by diversity, or by anything else. These three are the only portions of any of the six opinions in Bakke in which diversity is mentioned.

Now, Justice Brennan and the three others in his group did not join Powell either in Parts V-A or V-B, or in the diversity rationale given in IV-D; they very self-consciously joined him in a wholly different portion of his opinion–namely, Part V-C. We therefore cannot say, as Duggan does, that they would have agreed with Powell, for if they had, it would not have been inconsistent for them to join him in those other passages even though theirs was a defense of preferences on compensatory grounds. The fact that they did not do so very strongly suggests that they did not share his views–did not believe diversity constitutes a compelling state interest in the needed sense.

As for Brennan’s later remark, asserting in retrospect that some unnamed persons viewed diversity at that earlier time as a constitutionally permissible goal “on which a race-conscious admissions program may be predicated,” it has no force in determining what was actually done in the Bakke case. Brennan may have later changed his mind, but even so we may not infer that others had changed theirs. Nor could this retrospective account justify the inference that he and his three colleagues had earlier asserted or affirmed that diversity is a compelling state interest. Very plainly they did not.

How then are we to account for the paragraph–part V-C of Powell’s opinion–in which five Justices (Powell and the Brennan four) did join? The University of Michigan repeatedly cited that paragraph, suggesting that by it the defense of the diversity rationale is made secure. But what the five Justices agreed to in that single paragraph was no more than this: all of them believed (in Judge Friedman’s paraphrase) that “the state has a substantial interest that legitimately may be served by a properly devised admissions program involving the competitive consideration of race and ethnic origin.” The nature of that substantial interest was very carefully not stated, because it was not agreed upon. For Brennan and company, it was compensation for injuries earlier done; for Powell, who had forcefully rejected the compensatory argument, it was the diversity that might contribute to the robust exchange of ideas. The paragraph, short and very carefully crafted, gathered the needed five votes in order to allow only that the state may have some interest that can justify the use of race (thereby overturning the absolute restriction imposed by the California Supreme Court), while remaining in deep disagreement about what that interest is.

What about Duggan’s contention that, in considering the two alternative rationales for the result in Bakke, diversity, being the narrower, must therefore be taken to be what the Court held? This would be a clever argument if only its minor premise were true. But it cannot be said of these two rationales that diversity is the “narrower” and compensation the “broader,” for the two rationales share no common ground, being rather completely different in spirit. Compensation for those who have been injured is retrospective, seeking remedy for what was earlier done; achieving a diverse enrollment is prospective, seeking an improvement in what is to be done. Neither of these two principles is in truth narrower or wider than the other; they are entirely disparate.(9) Thus, as Judge Friedman sums the matter up, “there is no force at all to defendants’ contention that the Brennan group’s joinder in Part V-C of Justice Powell’s opinion may be taken as an endorsement of Justice Powell’s discussion of the diversity rationale.”

Finally, there is the matter of judicial authority. The only commanding authorities in the Michigan courtrooms where Gratz and Grutter were heard are the Supreme Court of the United States and the court of appeals of the Sixth Circuit (covering Michigan, Ohio, Kentucky, and Tennessee). Other circuit courts (whose opinions are not binding here) have indeed addressed the issue, but are also in conflict. Thus, as I noted earlier, a panel of the Ninth Circuit did agree with Duggan, claiming with unwarranted confidence that Brennan had “embraced” Powell’s views about diversity, which he supposedly considered “simply an allotrope” of his own compensatory principles. But William Brennan was a learned man who probably knew exactly what an allotrope is–in chemistry, one of the several forms of a single element, as charcoal and graphite and diamond are allotropes of carbon. To suggest that the diversity rationale, aimed at enriching learning, is an allotrope of the compensatory rationale, aimed at redressing injury, is an intellectual blunder Justice Brennan was too keen to have made.

Authority on Friedman’s side of the issue comes from the court of appeals of the Fifth Circuit (covering Texas, Louisiana, and Mississippi). In Hopwood v. University of Texas (1996), a case very similar to Grutter, a three-judge panel of that court held emphatically that the diversity rationale has absolutely no support other than that of a single Supreme Court Justice, Lewis Powell; has never represented the view of a majority of the Court in Bakke or in any other case; and has no value as a rule of law. When that opinion was appealed to the full Fifth Circuit it was again affirmed; subsequently the Supreme Court declined to review it.

IN SUM, the longstanding dispute about what principle Bakke provides–whether it does or does not support the view that diversity in a university’s entering class can serve as a compelling state interest–remains unresolved. That is why these Michigan cases are so intriguing. Preferential admissions can survive only if the need for diversity will suffice to defend them, and how the district and circuit courts will come down on this matter must depend on how they read the holding of the Supreme Court in Bakke. Judge Friedman, after meticulous consideration of the legal precedents, has come to the view most likely to prevail. He writes: “[This] court concludes that Justice Powell’s discussion of the diversity rationale is not among the governing standards to be gleaned from Bakke.”

When these cases reach the Supreme Court, however, what may be gleaned from Bakke may prove to be not so important after all. That Court, though always mindful of precedent, is not firmly bound by Bakke, however interpreted. It may reverse Bakke, or amend it in whatever way the nine Justices think our civil-rights laws and the Constitution demand.

Reliance upon the diversity rationale for racial preference faces still another hurdle. Suppose again, for the sake of argument, that diversity can, as Powell suggested, serve as the required “compelling need.” By itself, that would not be enough: the standard of strict scrutiny requires that when racial devices are employed, they must be shown to be “narrowly tailored” to address the need presented.

The demand for a “narrowly tailored” relationship is the jurisprudential way of expressing a deep moral truth: the remedy must fit the wrong, the instrument its object. If we aim to give redress, what we do must compensate in appropriate form and degree for the injury suffered, and must compensate those persons who suffered that injury and not some other set of persons who may happen to share their skin color. Similarly, when an instrument using suspect classifications is defended as essential for some state purpose, there must be a close fit between that response and the need to which it is a response.

That suitability–of the device to the need it claims to answer–is what the universities cannot provide for diversity. If, for example, the alleged need of the Michigan law school is a “critical mass” of minority students, a preferential system can meet that need in a narrowly tailored way only if it is known how many minority students constitute a critical mass–a number the University of Michigan claims cannot be ascertained. But if the number cannot be known, it is plainly impossible to say that any given system closely meets the need. Again, if the alleged need is for diversity of many different kinds and not of race alone (as Justice Powell said emphatically), admissions devices like those at Michigan that give automatic preference to three specific minorities cannot possibly be tailored to the claimed need. So, even if diversity were found to have satisfied the first prong of the strict-scrutiny standard, it probably could never satisfy the second.(10)

Nor is that the last of the problems confronted by this alleged justification of racial preferences. Suppose that the compelling need for diversity really were good law, and might even protect the wholesale incorporation of racial preferences in the Michigan style. Even so, all we would know is that diversity can serve as the needed compelling interest in some context; whether it does so serve would still remain to be shown.

In the two Michigan cases, the university assumed the authority of the diversity rationale and then went on to compile what it took to be the evidence showing that the work of the university depends critically upon ethnic diversity. As one might expect, Judge Duggan made much of this evidence. (Obviously, it would not have to be considered by Judge Friedman or by anyone else who denies that diversity could serve as a compelling need in the first place.) But the same evidence has also been given very careful scrutiny by outside observers.

Two teams–Robert Lerner and Althea Nagai of the Center for Equal Opportunity in Washington, D.C., and Thomas Wood and Malcolm Sherman of the National Association of Scholars–have produced full-length books on the question, and others have analyzed the data more concisely. The Wood/Sherman treatise–a lengthy and refined exploration of what is known in this sphere, including a re-examination of the database used by the University of Michigan–concludes that, as an empirical claim, the university’s argument simply fails; careful analysis does not support the conclusion that campus racial diversity is correlated with positive educational outcomes.(11) And the Lerner/Nagai treatise–an extraordinarily detailed and meticulous look at the methods and claimed results of the university’s chief witness, the psychologist Patricia Gurin–tears her study to shreds.(12)

Taken together, these two critiques of the empirical claims made by the University of Michigan are devastating. What they indicate is that even if the possible use of the diversity rationale could be established, its use will not succeed in the actual circumstances.

A BRIEF look at what lies ahead: the two Michigan cases will very probably be consolidated into one by the Sixth Circuit court of appeals in Cincinnati, but are not likely to be argued until later in the autumn. A decision by a three-judge panel of that court will probably be issued in early 2002. Whatever the decision of the panel, it will almost certainly be appealed to the full circuit court; although such appeals are rarely granted, the process may consume many months. And no matter what the final outcome in that process, an appeal by the losing side to the U.S. Supreme Court will assuredly follow.

That the Supreme Court will eventually take the case is very probable. For one thing, the Court sees itself as having the task of clarifying the law where decisions in the courts below it are in conflict. In this sphere, ever since Bakke, such conflicts have been many. Two opportunities to resolve these conflicts–a Ninth Circuit decision upholding preferences, and a Fifth Circuit decision condemning them–have been declined by the Court in recent months, probably because each had technical aspects that rendered it not wholly suitable for the purpose.

The Michigan cases, by contrast, provide an ideal vehicle for a restatement of the law. The conflict between Gratz and Grutter is fruitfully sharp, and the full record in the two cases provides a well-documented factual setting. The degree of preference given at Michigan, the ways it has been given, and the statistical outcomes in the admissions process are well established in both cases, especially that of the law school. The intent of the university to give racial preference is undisputed, and the university is ready to defend its practices openly. The facts and the arguments are all on the table. Even if the decision of the Sixth Circuit were agreeable to the Supreme Court, the Justices are likely to want their imprimatur on the outcome. The issue is ripe for final resolution.

Finally, we know that several members of the Court feel very strongly–it would not be an exaggeration to say passionately–that racial preferences violate the Constitution, and are offended by the widespread defiance of Title VI of the Civil Rights Act of 1964, which outlaws such preferences (by institutions receiving federal money) in the plainest language. Indeed, all five of the more conservative Justices–William Rehnquist, Anthony Kennedy, Sandra Day O’Connor, Antonin Scalia, and Clarence Thomas–have made very clear their opposition to the sort of “affirmative action” now practiced by Michigan and other public universities, and four of them have made explicit, too, their common belief that diversity cannot serve as a justification for preferences. In a 1990 broadcasting case that involved racial preferences given by the FCC, Justice O’Connor, writing for herself, Justices Kennedy and Scalia, and Chief Justice Rehnquist, put the matter bluntly:

“Modern equal protection has recognized only one [compelling state] interest: remedying the effects of [identified] 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.”

Four votes are needed for the Supreme Court to accept any case. That there will be five votes to accept these Michigan cases seems very probable; I reckon they rest in palms already sweating in anticipation. And we may also find principled opposition to the diversity rationale for preferences among the other four Justices as well. Justice Stevens, after all, was the author of the opinion in Bakke that found the University of California’s preferences to be a violation of Title VI. For Justices Stephen Breyer and David Souter, the principle of equal treatment under the law carries awesome authority. And Justice Ruth Bader Ginsburg was general counsel of the American Civil Liberties Union in the days when–as I can testify personally–the defense of equal treatment under the law was very high on that organization’s agenda. The position of any one of these four cannot be confidently predicted.

I conclude on another personal note: every catalogue of the University of Michigan, on whose faculty I have served proudly for 46 years, carries a formal notice that the university

“is committed to a policy of non-discrimination and equal opportunity for all persons regardless of race, sex, color, religion, creed, national origin, or ancestry … in employment, educational programs and activities, and admissions.”

Although I have been openly critical of its racially distorted admissions policies, I am deeply devoted to the University of Michigan, and I look forward to the day when that formal statement can be published without embarrassment. And what goes for Michigan goes for all the other universities whose long-successful efforts to hide racial preferences behind the appeal of the phrase “affirmative action” have finally been exposed.

Legislators have feared to deal with this topic, but most would find the outright support of preferential systems to be politically damaging. As for the public, it has made its overwhelming opposition to such systems plain at every opportunity offered by referendum. Now the courts, confronting university ethnic preferences at last, may decide they can be tolerated no longer. Let us pray that is so.

(1) The legal citation is 438 U.S. 265. An earlier case, DeFunis v. Odegaard, was ultimately held moot.

(2) Their academic credentials are such that, had either been a member of one of the preferred racial groups, it is a virtual certainty she would have been admitted.

(3) Thee request was made by me.

(4) By contrast, twelve points are given under the current system for a perfect SAT score and six points for a very poor SAT score.

(5) In a later response to a university request for a delay in the imposition of his judgment, Judge Friedman removed the qualifying adjective “practically.” And to the university’s contention that it would be irreparably harmed if forced to abandon racial preferences, the judge memorably replied: “Defendants are not irreparably harmed by an injunction that requires them to comply with the Constitution.”

(6) For an account, see “The ‘Diversity’ Defense” by Jason L. Riley in the April COMMENTARY, and the letters beginning on page 16 of this issue–ED.

(7) Powell wrote: “The guarantee of equal protection cannot mean one thing when applied to one individual and something else when applied to a person of another color. If both are not accorded the same protection then it is not equal.”

(8) In 1999, in an effort to overcome the “imbalance” created by the presence of too many women on campus, the University of Georgia gave a substantial advantage to male applicants. But a district-court judge was not impressed, and found such a device unlawful. If admissions officers were empowered to give such group preferences, what group might next be helped?

(9) Of the two principles it might even be argued that diversity is the broader, since it applies to everyone and not to minorities alone and since, unlike compensation, it can never be completed. But that, too, would be sophistical.

(10) If the university does fail to show a close fit between object and instrument, the appeals court may find it convenient to reject preferences on that ground alone, leaving for the Supreme Court the question of whether diversity can ever be the compelling need that justifies racial preferences.

(11) Thomas E. Wood and Malcolm J. Sherman, Race and Higher Education: Why Justice Powell’s Diversity Rationale for Racial Preferences in Higher Education Must Be Rejected, National Association of Scholars, available online at www.nas.org.

(12) Robert Lerner and Althea K. Nagai, A Critique of the Expert Report of Patricia Gurin in Gratz v. Bollinger, Center for Equal Opportunity, available online at www.ceousa.org.

CARL COHEN is professor of philosophy at the University of Michigan and the author of Naked Racial Preference: The Case Against Affirmative Action. His articles in COMMENTARY include “Race, Lies, and ‘Hopwood’ ” (June 1996).