Introduction

Thank you, Madam Chairman, for the opportunity to speak to the Commission today on the subject of affirmative action in higher education.

My name is Roger Clegg, and I am vice president and general counsel of the Center for Equal Opportunity, a nonprofit research and educational organization that focuses on civil rights and bilingual education issues. I have also served as a deputy in the Justice Department’s civil rights division from 1987 to 1991, and have taught employment discrimination law at George Mason University School of Law.

I am glad that the Commission is having this briefing, and indeed I think it is long overdue. For many years, colleges and universities in the United States have used racial and ethnic preferences in the admissions process. My organization has collected strong evidence of such discrimination and published it in studies of public colleges and universities in California, Colorado, Michigan, North Carolina, Washington, and Virginia, as well as the service academies at West Point and Annapolis. I am submitting copies of those studies today to the Commission, and they are also available on our website, <www.ceousa.org>.

The purpose of the Commission on Civil Rights, according to the letter of invitation I received for this briefing, is to investigate discrimination. There is no longer any doubt that many, many schools discriminate on the basis of race and ethnicity in ways that cannot be squared with the decisions of the Supreme Court, let alone the plain language of Title VI of the Civil Rights Act of 1964, which bars any recipient of federal money -- and this includes most colleges and universities -- from "subject[ing] to discrimination" any "person in the United States … on the grounds of race, color, or national origin." That’s why I say this briefing is long overdue -- assuming, of course, that the Commission is still opposed to racial and ethnic discrimination.

This is the fundamental question before the Commission today: whether college admission programs should discriminate on the basis of race and ethnicity. In framing the issue in these terms, I do not mean to suggest that the answer has to be "No"; but I do think that we have to be honest in admitting that the question is whether to discriminate or not. That is the only kind of "affirmative action" that’s in dispute.

Be Honest

Frankly, those favoring such discrimination frequently resist being honest about what they want. If accused of advocating discrimination, they will respond, for instance, by saying that race or ethnicity is "only one factor among many that should be considered." But that hardly means that discrimination on these grounds won’t take place. To the contrary: even if these considerations are "only one factor," there will be some cases -- in fact, there are many such cases -- in which they make the difference between whether someone is admitted to a college or not. If that is not true, then why consider these factors at all, right? And in those cases where race or ethnicity has been the deciding factor, then someone has been discriminated against because of that immutable characteristic. If he or she had different melanin content or ancestors, he or she would have been treated differently.

Another favorite ploy of the apologists for this kind of discrimination is to respond that they "are opposed to unqualified applicants being admitted." But, again, this hardly means that there is no discrimination. The question is whether someone who is less qualified -- even if not unqualified -- is being admitted because of race or ethnicity. If so, then there has been discrimination.

Or the apologist will argue that they are not favoring discrimination because they "are not favoring quotas." And, of course, quotas then are defined in such a narrow way that no one could favor them. But, once more, the absence of such quotas hardly means that there is no discrimination. It is easy to have discrimination without there being an absolutely rigid numerical target; a general or flexible target can result in discrimination just as easily.

Defenders of preferences also will frequently point out that written tests are not perfect predictors of future performance at college and that other admission criteria frequently used -- like being the offspring of an alumnus or being a good tennis player -- are even less predictive. But none of this makes the consideration of race or ethnicity any less discriminatory. If schools are using selection devices that are defective for whatever reason, go ahead and criticize them, but don't think for a minute that such criticisms answer the objection to using race and ethnicity

factors that are, in any event, more irrational, more divisive, more stigmatizing, and more constitutionally problematic than any other criteria.

A good way to tell whether discrimination has occurred is to ask what the reaction would be if the shoe were on the other foot. This is a good test to keep in mind whenever someone proposes a justification for discriminating in favor of a particular group: ask yourself how persuasive it would be if it were offered as a justification for discriminating against that group. For instance, suppose that, just after Brown v. Board of Education, a college in Mississippi had said that, while it was not going to have an absolute ban on admitting blacks to a particular school, nonetheless race was going to be "a factor" of unspecified weight. Or suppose that this school said that it was not discriminating because only "qualified" whites were being admitted. Or suppose that the school denied it was discriminating because the low ceiling it set on blacks was just a general and flexible target, or that it was permissible to consider race because the school also used other, equally irrational criteria in its decisionmaking. I don’t think that any of us would hesitate to label those denials of discrimination as specious. Maybe the discrimination was not as categorical or as absolute as total segregation, but no one would doubt that discrimination was taking place.

The Real Question

So, we are dealing with discrimination -- the real question is, is the discrimination worth it? That is, are the costs of discrimination outweighed by the benefits? (I’m not talking only, or even principally, about economic costs here; I mean to include all social costs.)

I am not going to argue that there are no conceivable benefits to granting preferences. But, too often, the apologists will state the justifications and then, not only will they not answer the criticisms of those justifications, they will not even consider the fact that, even if those justifications have some merit, we must consider the costs of preferences before anyone can conclude that they are a good idea.

Again, let’s put the shoe on the other foot. Suppose an apologist for pre-Brown segregation had said, well, segregation makes sense because it improves school spirit. Now, there are two ways to attack that argument. First, you can argue that it doesn’t improve school spirit, or at least not as much as the apologist says. But, second, you would argue that maybe it does, maybe it doesn’t, but even if it does you don’t stop the inquiry there. There are costs to segregation that make it a bad idea even if it means school spirit won’t be improved. The chance that maybe school spirit will be improved just isn’t worth the likely costs.

The costs and benefits of discrimination must each be weighed, and they must be weighed against one another at the same time on the same scale. By that, I mean that you should not adopt a credulous, "Well, there might be a benefit," mindset for considering arguments in favor of discrimination and a highly skeptical, "But there might be some exceptions to that general rule, maybe," mindset for considering the arguments against discrimination.

So, let us consider, first, the benefits that preferences might have, and then weigh them against their costs.

Three Benefits

The claimed benefits for the use of preferences fall into three categories: prophylactic, remedial, and diversity. The prophylactic justification is that we must affirmatively discriminate in favor of a group’s members lest we fall into discriminating against them. The remedial justification is that discrimination now in favor of members of a group can help make up for discrimination against members of that group in the past. And the diversity rationale is that there are benefits to having certain groups represented in some substantial way at the school.

The justifications for preferences almost always fall into one of these three categories -- and not only for higher education, but for employment, government contracting, voting, and other areas as well. But sometimes a justification is more plausible in one context than in another.

Prophylactic Benefits

Consider, for example, the prophylactic argument. If we were dealing with a recalcitrant employer who had over and over tried to circumvent a court order that he stop discriminating against blacks for a relatively unskilled, entry-level position, then the prophylactic argument in favor of the judge ordering him to set and meet certain hiring goals has some power. That employer really is likely to discriminate against blacks unless such goals are in place. In this situation, the aim is to create the same nondiscriminatory result that would have come about anyway had the employer not been a bigot.

But the prophylactic rationale has very little plausibility In American higher education today. Do we really need to set goals and timetables to keep college admission officers from discriminating against blacks and Hispanics?¹ That seems very unlikely. The admission officers at the University of Texas law school, for instance, cheerfully ran an illegally segregated admission system for years to ensure that blacks and Mexican Americans would be given heavily preferential treatment.²

Remedial Benefits

There is also an obvious pitfall with the second, remedial rationale: discrimination in favor of today’s individuals in group X does nothing to help the different individuals in group X who suffered discrimination in the past. The justification, then, must argue one or both of the following: (a) the very individuals who suffered discrimination against them are the ones who now will be receiving discrimination in their favor, or (b) the discrimination suffered in the past has had discriminatory results still being felt by those in group X.

As to the first justification, bear in mind that, in the context of college admissions, we are dealing mostly with 18-year-olds, born around 1981 or 1982. They probably have not participated much in the work force; if they have, the laws prohibiting discrimination against them on the basis of race or ethnicity have been in effect since long before they were born. Nor have they suffered discrimination in education. Public schools are no longer segregated by race or ethnicity, nor are most private schools.

I can hear the objections. Yes, there are exceptions to the statements in the preceding paragraph. Some teenagers may have suffered employment discrimination because of their race or ethnicity; some public schools may receive less funding because of the color of the children who go there. But the point is that an 18-year-old today is unlikely to have suffered the kind of systematic discrimination against him or her that would justify systematic discrimination in his or her favor. Something as dramatic as a preference has to be based on likelihoods, not bare possibilities.

The Present Effects of Past Racial and Ethnic Discrimination

So, under the remedial rationale, we are left to consider preferences for race and ethnicity because of the historical effects of discrimination being felt by the descendants of those who suffered the discrimination firsthand. Whether such preferences make sense will hinge on how good a "fit" there is between the class of people who have a particular color or ethnicity and the class of people who are suffering because of past discrimination against their ancestors.

The fit is a poor one. There are, to mix a metaphor, too many false negatives and too many false positives. That is, there are many people who are descended from past sufferers of discrimination who are not eligible for preferences, and many who are not so descended who will be eligible. For instance, not only blacks and Hispanics but also Jews, Asians, American Indians, and Americans of Irish, Italian, Polish, and German origin have all been subjected to discrimination at one time or another in this country, yet probably no college would grant a preference to all these groups.³ Conversely, some blacks are descendants of recent immigrants and can hardly claim to have suffered even indirectly from slavery and the Jim Crow laws;⁴ a high percentage of Hispanics are immigrants or descended from recent immigrants.⁵ It is also unclear why the only relevant discrimination to consider is that which occurred in the United States. If a college wants to right the wrongs of past discrimination that it did not itself commit, then why does it matter where the discrimination occurred? This really would open the floodgates, since many, many immigrants -- especially the early WASPs, whose descendants everyone hates -- were fleeing religious or some other kind of persecution in their native countries.

Some have suggested that preferences ought to be limited to blacks -- a special case of particularly heinous discrimination (although no black was interned during World War II, like Japanese Americans, or shunted off to reservations, like American Indians).⁶ But, again, not all blacks have slave ancestors. And not every black was confronted with discrimination or the same kind of discrimination (among the most obvious variables are geography and occupation).

It is likewise dangerous to generalize among the many different subgroups who make up the group "Hispanic," a social-scientist construct. And many blacks and many Hispanics do not suffer a depressed socioeconomic status even though they or their ancestors may have suffered discrimination -- which makes it impossible to justify a preference for them on the assumption that they do.

It is also a non sequitur to assert that those blacks and Hispanics who did suffer a depressed socioeconomic status and who also may have suffered discrimination became or remained impoverished because of racism. Intervening and independent decisions probably had a greater impact. Illegitimacy, substance abuse, and poor work habits can mire an individual in poverty just as racism can. I understand the argument that racism leads to despair and that despair makes bad lifestyle choices more likely, but this is a tenuous sequence -- overcome by many -- and the additional connection between this sequence and an entitlement to racial preferences is more tenuous still.

There is also the problem of mixed ancestry. If a wealthy white doctor marries a nurse who belongs to a discriminated-against minority, are their children entitled to a full preference, or no preference, or a half preference? There is no way to avoid this question. Either you say that one drop of minority blood entitles someone to a full preference, or you draw the line Nuremberg-fashion somewhere else,⁷ or you award weighted preferences. All are unattractive choices.

Finally, it would be desirable if those now penalized by the use of preferences also happened to be the beneficiaries of past discrimination against those now receiving the preferences. There is, however, little if any correlation in this regard. Recent immigrants and children of recent immigrants, descendants of working-class northerners, Midwest farmers’ sons, Hasidic Jews, and so on -- none of them is likely to have benefited in more than a very indirect way from discrimination against blacks or Hispanics, yet each of them is placed at a competitive disadvantage with them by admission preferences.

Diversity

The final justification for the use of preferences is, to quote Justice Lewis Powell’s opinion in Regents of the University of California v. Bakke, "the educational benefits that flow from an ethnically diverse student body."⁸

Since Justice Powell’s opinion is relied on so heavily by the advocates for diversity -- legally, there is nothing else they can rely on -- it is worth quoting at the outset a paragraph from the opinion that makes clear what the "diversity" argument cannot justify, at least as far as Justice Powell was concerned:

My own suspicion is that this is a fair description of what most colleges are doing. They have decided that they want to have a particular mix of students, and they have asked their lawyers to come up with a rationale, any rationale, that will allow them to achieve that mix. Diversity is one such rationale, and it has been one of the few successful ones as a matter of law, but it remains a pretext.

Justice Powell also rejected the argument that preferences could be justified on the grounds that minority graduates (in the case of Bakke, minority doctors) would be more likely to serve minority communities.¹⁰ Eight years later, in Wygant v. Jackson Board of Education,¹¹ he also rejected the so-called role model argument -- that preferential treatment was justified because minority children needed role models (in that case, teachers) to look up to. As he wrote in Wygant, "Carried to its logical extreme, the idea that black students are better off with black teachers could lead to the very system the Court rejected in Brown v. Board of Education ...."¹²

The "Educational Benefits" of Diversity

But let’s assume then that colleges really believe that there is something educationally beneficial about diversity. The first problem to note is that this belief is a two-edged sword. If diversity can justify preferences in favor of certain groups that are "underrepresented," then surely it can justify negative weights on those that are "overrepresented." There is simply no way to justify the former without justifying the latter. Recall the unhappy history of the Ivy League’s ceilings on overrepresented Jews.

Well, let’s press on. What might the "educational benefits" to diversity be? There are two basic possibilities. First, it might be that there is something desirable simply about being with people who have a different color or ethnic background, even if they do not otherwise differ. Or, second, it might be that because of the difference in color or ethnicity, people are different inside, too. For instance, they might be more likely to hold particular views or beliefs, or to reflect different cultures or ways of thinking, or to have had particular experiences.

The first possibility hinges on the ability to show students that race and ethnicity don’t matter. The idea is that a bigot or potential bigot, if he is forced to see that people of other colors are really not so different from himself, will conclude that bigotry is wrong. But this plan will work only if the bigot is surrounded by students who really are similar to him in ability. If the minorities admitted into the school are less qualified than the nonminorities, the bigot’s attitudes will be reinforced, not eroded. And, of course, the first possibility is inconsistent with the second -- that is, the argument that minorities are in fact different inside from nonminorities.

So let’s turn to this second possibility. This is what Justice Powell had in mind in Bakke. The critical passages in his opinion there read as follows:

•  
  • ... [T]he nation’s future depends upon leaders trained through wide exposure to the ideas and mores of students as diverse as this Nation of many peoples.¹³
  • ... [U]niversities must be accorded the right to select those students who will contribute the most to the robust exchange of ideas....¹⁴
  • ... [A] student with a particular background ... may bring [to the school] ... experiences, outlooks, and ideas that enrich the training of its student body and better equip its graduates to render with understanding their vital service to humanity.¹⁵

To decide how plausible this rationale is, we must decide: (a) what the desired inner differences are and (b) whether the best way to find them is by considering race or ethnicity.

On the first point, not every difference is important. Maybe some groups, as a whole, have a particular "outlook" or "experience" when it comes to food, but so what? The desired differences ought to relate to the intellectual discipline being taught. Thus, for instance, it is unclear to me how any cultural difference is relevant to graduate work in mathematics. On the other hand, having experienced discrimination firsthand may well give one insights for sociology and law. Are schools making these distinctions?

The Dangers of Proxies

The second point is the most critical: Given that we might want some particular inner qualities or experiences represented at the university, does it make sense to use race or ethnicity as a proxy for them? Instead, why not select directly for the quality or experience rather than assuming that everyone with a particular race or ethnicity has it and that others can’t?

In the passages quoted from Justice Powell’s opinion, he stressed the importance of having students with particular "ideas and mores" and, later, "experiences, outlooks, and ideas," who will "contribute the most to the robust exchange of ideas." Well, if we want students with particular ideas, mores, outlooks, and experiences, then why not select those students -- rather than assuming that, because a student has a particular color or ancestry, he or she will meet the admission officer’s conception of, for instance, a typical Hispanic woman? And why assume that only certain racial and ethnic groups -- and not, for example, religious groups -- have a special perspective?

This is also a good time to put the shoe on the other foot. How persuasive would most affirmative-action apologists find it if a college were to limit the enrollment of blacks and Hispanics by using race or ethnicity as a proxy? Not very, I’m sure. There's not much difference in this context between a proxy and a stereotype. Suppose a university were to deny admission to women into a military history program on the grounds that women were unlikely to appreciate matters of military camaraderie? Or to grant a preference to persons of European descent in a European Studies program? Or, on grounds of diversity, to limit the number of Hispanics admitted into its Spanish graduate program?

Let us go through the items on Justice Powell’s list and consider how persuasive the claim is that race or ethnicity are the best way to select students. Can we make blanket generalizations about the "ideas," "outlooks," and "mores" of people by looking at their color and ancestors? The answer is no. Indeed, that used to be the required answer among liberals.

What about "experience?" I would be reluctant to generalize much about the kind of life a person had lived simply by looking at him. Moreover, there is no monolithic "black" or "Hispanic" culture that is part of every black or Hispanic. It is certainly invalid to assume that everyone in some racial or ethnic group is economically disadvantaged -- invalid and unnecessary since, if you want to grant a preference for those who have suffered economic disadvantage, you can do so without introducing race into the equation.

At the end of the day, the diversity rationale boils down to an argument that the firsthand experience of discrimination is one that ought to be represented in the study of certain disciplines, like law and sociology. Certainly this is the strongest case for the use of race or ethnicity as a proxy. But the weakness in this claim is essentially the same as the weakness in the claim for "remedial benefits" based on these characteristics: we are dealing, after all, with teenagers, born around 1981 or 1982. We are being asked to believe that all, or nearly all, of these teenagers -- regardless of other particulars in their backgrounds -- have suffered discrimination to the extent and of the sort that will give them special insights lost to other 16-, 17-, and 18-year-olds. I am very skeptical of this claim.¹⁶

Is It Worth It?

Still, I am not prepared to say that the claims for benefits from preferences are ludicrous, although obviously I believe each claim is riddled with holes. As I warned earlier, however, the inquiry cannot end with the conclusion that there might be something to the claim that the use of preferences has some benefits. The inquiry must proceed to ask, is it likely that those benefits outweigh the costs?

And what are those costs? Unlike the claimed benefits, the costs are clear and undeniable.

To begin with, you are discriminating on the basis of race or ethnicity. You are making generalizations about people on the basis of these immutable characteristics. You are then rewarding or punishing them because they happen to have a certain color or ancestors. This is a bad thing to do. You set a bad precedent and give up a lot when you give up the principle of colorblindness. It is an especially dangerous activity for a state or state-run institution.

Next, you create resentment among those who lose out because of your discrimination. And this resentment is unlikely to be limited in time or scope to the one instance in which it occurs. It is also unlikely to be limited to the immediate victim: his parents, friends, and family will be resentful, too. So will his or her spouse and children, if any.

You will stigmatize the so-called beneficiaries of the preferences -- both in the eyes of others and in their own eyes. Where preferences are used, people will assume that a person in the preferred group who was admitted was less qualified than other people who were admitted. And, of course, that is a fair assumption. The whole purpose of the preferences is, after all, to admit those who would otherwise have been rejected as less qualified. And this lesson is taken to heart by the beneficiaries of the preference, too, as Shelby Steele has discussed.¹⁷ They will then begin to doubt their own ability to succeed without preferences, or they will conclude that they do not lack the ability but that the system must lack fairness. Both lessons are unfortunate and untrue.

You will compromise the mission of the university. You will be making intellectual ability a secondary attribute. Your graduates will not be as good. You will be tempted to discriminate in your grading, retention, and graduation policies. The school's graduates will be less competent, with all the attendant social and economic costs of that.

And, finally, as a lawyer I must point out the obvious: You will be breaking the law. The court of appeals in Hopwood v. Texas has ruled, correctly, that the Fourteenth Amendment (which applies to all state schools) and Title VI of the Civil Rights Act of 1964¹⁸ (which applies to all schools, private and state, that receive federal money) forbid the use of preferences, too.¹⁹ Other courts have rejected the diversity rationale in other contexts. And the Supreme Court has ruled that section 1981 of title 42 of the U.S. Code bars racial discrimination even by private schools that do not receive federal money.²⁰ Finally, section 1983 of title 42 creates broad civil liability for such discrimination, and sections 241 and 242 of title 18 provide criminal punishment.

The Last Argument

The last argument made by those favoring discrimination is that, whether or not the preferences can be justified, we must use them because otherwise certain groups will be dramatically "underrepresented" at our colleges and universities. There are several responses to this argument.

The first is that this is precisely the rationale that Justice Powell rejected. This justification is nothing more than "discrimination for its own sake."²¹

The second response is that the claim is exaggerated. The students who otherwise got into top-tier schools might now go to second-tier schools -- where, incidentally, they will be surrounded by students whose qualifications they more closely match. Not a bad thing -- indeed, a good thing.²²

But what about those top-tier schools? Isn’t it the case that for them there may be, for instance, fewer blacks or Hispanics -- even if other criteria are used in addition to standardized test scores? Fewer, but not none. Besides, if a school is using the best criteria it can devise to select the best qualified students, then why should anyone object to the school’s demographic makeup? Suppose there are no blacks admitted to Harvard Medical School one year. So what? Is it rational for black people to conclude from this fact that America is racist or that they have no stake, or less of a stake, in our nation? Only a Harvard intellectual could believe that. Different professions favor different interests, talents, and inclinations, which may not be perfectly mirrored at every point in time among our nation's many demographic groups. Again, so what?

Conclusion

So, tally up the costs and benefits. Maybe there is something to the diversity rationale in some cases, and maybe there is something to the remedial and prophylactic rationales in some cases, too, although we have seen that all three rationales are riddled with holes. But let’s suppose that there is something left, in each case, greater than zero. You then are obliged to consider what’s on the other side of the scale: the fact that you are discriminating, the resentment, the stigmatization, the compromising of the mission of the university, and the illegality. Which way does the scale tip? To anyone who is intellectually honest, it is not a close question. Preferences should not be used.

The Commission should condemn this discrimination, just as it has condemned other discrimination in the past.