Texas 10% Plan: Diversity or Discrimination? (Continued)

A little while back I discussed the Texas 10% plan and tentatively argued such plans were constitutional. This argument provoked some interesting comments, including one by Terry Pell of the redoubtable Center for Individual Rights, which in turn led me to post additional comments, and I also discussed that issue here. You may want to review those now. (There will be no quiz, but you are expected to be familiar with the material….)

More recently, two powerful if brief arguments that the Texas 10% and similar plans are unconstitutional have appeared, and I believe this question is important enough (and likely to become even more important) that it is worth revisiting. I respect these arguments and their authors and find myself in the unusual situation (as my friends, the six other readers of this blog, and even my former friends will readily agree) of disagreeing with a strong version of color-blindness, something I almost never do. I do so almost expecting to be persuaded that I’m wrong about this. But then, that’s what blogs at their best are good for.

Few people have written more or better about discrimination than Roger Clegg, vice president and general counsel of the Center for Equal Opportunity, whose letter criticizing the Texas plan appeared in the Washington Post a little over a week ago. According to Clegg, the plan is unconstitutional “because it was adopted to circumvent a federal court decision striking down the use of racial preferences and rewrites admissions criteria to achieve a particular racial mix.”

But this begs the question. The plan “circumvents” the federal court ruling (Hopwood) only if its means or ends are unconstitutional. Are they? Clegg continues:

This sort of affirmative action is unfair to the better-qualified students who are not selected because they have the “wrong” skin color. Yet there are many other costs: It sets a terrible legal, political and moral precedent to allow racial discrimination; creates resentment and stigmatizes the so-called beneficiaries in the eyes of their classmates, teachers and themselves; fosters a victim mind-set; removes the incentive for academic excellence and encourages separatism among racial and ethnic minorities; compromises the academic mission of the university and creates pressure to discriminate in grading and graduation; breeds hypocrisy within the school and encourages a scofflaw attitude among college officials; mismatches students and institutions, guaranteeing failure for many of the former; papers over the real social problem of why so many African Americans are academically uncompetitive; and requires unsavory activities such as deciding which racial and ethnic minorities will be favored and which ones will not, and how much blood is needed to establish group membership.

Clegg does not say, however, at least not here, why the top 10% plan is in fact “affirmative action,” and especially why those students not accepted under it “are not selected because they have the wrong skin color.” On the contrary, they — and the “they” includes students of all races, ethnicities, religions — were not accepted because they were not in the top 10% of their classes.

Similarly, he asserts that it sets a precedent of allowing racial discrimination without showing how the plan discriminates by race. The remainder of Clegg’s paragraph makes a very strong case for why the top 10% plan is a bad idea, but not how or why it is racially discriminatory. I shall return to what I think his deeper argument really is.

In the same vein, a second argument comes from Brian Fitzpatrick, who has the twin distinction of being a friend of Stuart Buck‘s and a former law clerk to Justice Scalia. He, too, argues that the Texas Ten Percent Plan “is as illegal as the explicit racial preferences it replaced.”

Fitzpatrick’s argument explicitly emphasizes motivation, as I believe Clegg’s does implicitly.

Everyone knows why Texas decided to make class rank the only admissions criterion for what turns out to be half of each year’s freshman class at the University of Texas at Austin. The Texas legislature was quite explicit about its purposes. It enacted the plan not because it thought this approach would yield the smartest students, nor because it thought this was the most administratively feasible way to admit students. It enacted the plan because it wanted to bring the number of blacks and Hispanics admitted back to the level it had achieved during the good old days of its illegal discrimination. Little was left to chance; the legislature relied on studies showing that the plan would increase black and Hispanic representation at the University of Texas at Austin to almost exactly the same level it had been at before.

This raises a legal question that has been largely ignored: If it is illegal to admit applicants based on race in order to boost black and Hispanic representation, is it also illegal to admit applicants based on some other criterion, where the reason that criterion was selected over others was to boost black and Hispanic representation? The answer to this question must be yes. Otherwise, anti-discrimination law would be all form and no substance. For example, if a law against age discrimination only prohibited an employer from firing employees over a certain age, then it would not stop the employer from seeking to achieve the same result by firing those with gray hair. If anti-discrimination laws prohibited only form and not substance, they would be very hollow protections indeed.

…. Deciding to admit students on the basis of class rank alone because it will increase the representation of blacks and Hispanics in universities is no different than deciding to send students on one side of town to a certain high school because it will decrease the representation of blacks at other high schools, or deciding to disenfranchise citizens who commit certain crimes because it will decrease the representation of blacks in the electorate.

But is the top 10% plan really “no different” from the above examples? I’m not persuaded. In those examples, the formally neutral policies — geographical zoning by race for the purpose of promoting school segregation; moral turpitude laws applied only to blacks — not only were intended to impose a burden based on race; they did impose a burden based on race. Firing people with gray hair is even less persuasive because there is no rational basis for it whatsoever.

Who are the “victims” of the Top 10% plans? Everyone and anyone — regardless of race, creed, or color — who is not in the top of his graduating class. Whites (in poor rural schools) as well as blacks and Hispanics benefit from it; blacks and Hispanics (who are not in the top 10% of their classes) as well as whites do not benefit from it. If the real issue is one of numbers, then the 10% critics are in the position of making a “disparate impact” claim, something they have been persuasively critical of in other areas. (See Roger Clegg, The Bad Law of ‘Disparate Impact,’ THE PUBLIC INTEREST, Winter 2000)

One is left, then, with what I believe is the essence of Fitzpatrick’s, and I believe Clegg’s, argument: that the constitutional flaw in the top 10% plan is its motivation and intent to increase the numbers of minorities in Texas colleges, ostensibly as a means of enhancing “diversity.”

Critics of racial preferences nearly always argue that they (we) are not opposed to diversity, only to racially discriminatory means (such as different, lower standards for minorities) adopted to achieve it. In his letter, for example, Clegg writes that it

is wrong [to suppose] that the issue is whether racial diversity on college campuses is desirable. No one would suggest the contrary; the issue is whether it is so desirable that it justifies racial discrimination. The answer is no, because the costs are too high.

Precisely. But if you believe that the intent alone to promote racial and ethnic diversity — even by means that, unlike racial preferences, do not award benefits or burdens based on race — is sufficient to invalidate a policy, then it is somewhat hollow to say that you don’t object to diversity. You might not object to it per se; you just object to any policy whose purpose is to promote it.

Similarly, we critics of preferences often say that we support “affirmative action” in its original incarnation. “Originally,” as Clegg has written (link requires subscription), affirmative action

meant simply taking positive steps — literally, affirmative action — to ensure that discrimination didn’t occur. Nothing is illegal about that — or about colleges’ casting the widest possible nets, recruiting extensively and eschewing old-boy networks and irrational qualifications.

But since the intent of casting wider nets is obviously to attract more minorities, it’s not clear why this would not also be invalidated by the intent argument that is said to bar the top 10% plans. Do we really want to prevent a university from recruiting more heavily in minority and rural schools from which it has received few students than in rich suburban schools who produce an abundance of applicants? I don’t think so.

In short, I’m not persuaded by the legal argument against the top 10% plans. Those plans are not only facially and formally neutral; they also have no racially targeted or even identifiable victims. No harm, no foul. They have academic victims — many students not accepted under the plans who have more academic merit than many students who are accepted, but that is not a constitutional violation. Nothing in the Constitution, for example, would prevent a state university system from choosing to admit all its students by means of a statewide lottery of all graduates. Many academically talented students would lose out under such a scheme. There are many good, even compelling, arguments against it, but I don’t think racial discrimination is one of them.

But put the legal argument aside. I am persuaded that it would be an enormous political mistake to go after these plans, and not only or even primarily because George W. Bush signed the Texas law. First, we are far from having gotten rid of overt preferences. Second, we would be seen as rigid and overreaching, in part because, Third, it could easily lead to absurd results: we would be in a position of objecting to Texas’s plans because we know its motive was to promote racial/ethnic diversity, but if, say, Kansas were to adopt such a plan, claiming that it would promote greater representation from all parts of the state, we would have no objection to it. Although it is true that intent is often the difference between legal and illegal activity, I don’t think we want to be in the position of arguing that the intent to promote diversity is itself illegitimate.

Here’s a proposal: if a policy is racially neutral on its face, does not distribute burdens or benefits to members of identifiable racial/ethnic groups because of their membership in that group, and has a rational basis, common sense justification such as increasing diversity, let’s not attack it simply because it is known in advance that minorities will disproportionately be its beneficiaries. If that knowledge, or even that intent, were enough to invalidate a policy, after all, we might have to close down all the prisons and not end welfare “as we know it” but altogether.

Say What? (43)

  1. Roger Sweeny November 17, 2002 at 11:07 am | | Reply

    Question (or query, as the law professors say): What if a state legislature said, “Academic achievement in high school is disproportionately concentrated among whites and asians. If that were the basis of admission to our state university, we would have more of them than we want, and we would have fewer blacks and hispanics than we want. We want all ethnicities to be represented he same way they are in the general population. Therefore, admissions will be by a lottery. Everyone applying for admission will be given a number. We will put all the numbers into a big drum and pick out as many numbers as we have space in the entering class.”

    Do you think that would be undesirable–or unconstitutional–discrimination?

  2. John Rosenberg November 17, 2002 at 11:25 am | | Reply

    Well, as I tried to suggest, I believe such a policy would be highly undesirable, at least at “flagship” institutions, but constitutional. Although I think they are educationally questionable, I see nothing unconstitutional about preferences for athletes, legacies, vegans, oboe players, … or students in the top X% of their classes.

  3. David Pittelli November 17, 2002 at 11:56 pm | | Reply

    What about poll literacty tests for voting? Aren’t they unconstitutional because they are racist in intent and effect, even though there is a logical purpose to limiting the franchise to literate people who can make reasoned political judgements? If so, then don’t intent and effect simialrly make the “10% plan” unconstitutional, since its intent and effect is to discriminate in favor of minorities in the face of a legal decision that it is unconstitutional to do so?

  4. John Rosenberg November 18, 2002 at 6:36 am | | Reply

    This is a good question, and shows why it is not sufficient to look just at the face or form of a policy. Literacy tests looked neutral. But the crucial fact is that they were not: they were applied virtually exclusively to blacks. Keeping blacks from voting was their purpose, application, and efffect. I don’t believe that is true with 10% plans. The crucial question for me regarding the latter is, who is discriminated against because of race?

  5. Roger Sweeny November 18, 2002 at 10:25 am | | Reply

    Okay, let’s play with the question from the first comment. Enormous State University goes through its usual admissions process but just before the admissions letters are about to be mailed out, the state legislature mandates a 10% plan FOR THIS YEAR. The admissions committee goes back to work and immediately comes up with a new list. Letters are mailed out, etc.

    A Freedom of Information Act request makes public the list of students admitted and not admitted in both cases. We now know exactly who was “hurt” by the change in policy. It turns out that the students that were admitted the first time and not admitted the second time are 92% white, 7% asian, 0.5% black, and 0.5% hispanic.

    Undesirable? Unconstitutional?

  6. John Rosenberg November 18, 2002 at 1:48 pm | | Reply

    Roger, These are very good questions, and indicate why this is a difficult issue. So let me repeat that I’m not convinced I’m right here, or that you’re wrong. Certainly if the numbers in an actual case resembled the ones you pose, AND there were corresponding evidence that the legislators intended such a result, the 10% plan may well be thrown out. I suspect that in real life a) the numbers would not be as dramatic as yours, and/or b) there might not be supporting evidence of racial intent (my hypothetical Kansas example in my original post). Also again, let me emphasize that, in my opinion, to be illegitimate on racially discriminatory grounds (as opposed to merely stupid educational policy) a policy requires both racially targeted victims AND no good non-racial justification. If that were not the case, as I meant to suggest by my welfare and prison examples, then all sorts of things could be invalidated as racial discrimination, such as, for example, any regressive tax, which imposes a heavier burden on the poor who are disproportionately black. Etc. Finally, repeating myself shamelessly yet again, even if your legal argument could prevail we would be mistaken to pursue it. If we argued that the non-discrimination principle invalidates any policy whose intent is to promote racial diversity, what we would primarily achieve is discrediting the principle among many people who now accept it or can be persuaded to accept it. In the absence of (in my opinion) real victims of racial discrimination, it’s simply not worth it.

  7. John Rosenberg November 18, 2002 at 4:26 pm | | Reply

    A reader has sent the following comments to me by email:

    >>>

    I enjoy your Web site. However, I see several flaws in your arguments justifying 10% (or whatever percentage) plans. As these plans clearly are intended to promote racial/ethnic diversity, they not only benefit those who possess the favored racial or ethnic characteristics, but they necessarily burden those who do not possess these characteristics. What is true for straightforward racial preferences is true here: even if the intent of legislatures and schools is focused on the benefits to minorities rather than the burdens to non-minorities thus conferred, it is still intentional discrimination, and even benign intent doesn’t constitute a legal safe harbor.

    To the extent they are successful, how can it be said that such plans do not actually impose a burden on non-minority applicants who otherwise would have been admitted had such plans not been in effect? The fact that some whites benefit, depending on where they go to high school, is irrelevant. In the later stages of the Aderand litigation, for example, the Clinton Administration claimed that the contractor was no longer injured by preference programs because the latter had been altered to include white firms under some (rare) circumstances; however, this shouldn’t alter our opposition to the core discriminatory nature of such programs. Similarly, the fact that some blacks are burdened, again depending where they go to school shouldn’t protect percentage programs. Again taking set-aside preferences for minority businesses, the establishment of cutoffs based on annual dollar volume, even though it removes some minority firms from the potential pool of beneficiaries, doesn’t make much of a difference in how burdensome such preferences are to white-owed firms, or make the programs comport with equal protection requirements.

    There is a crucial difference between the disparate impact claims that Roger Clegg and many others criticize, and the disparate impact produced by percentage plans, which you say motivates criticism: proponents of disparate impact litigation focus on impact rather than intent, and believe that they need not prove intentional discrimination to prevail in court. In the case of the percentage plans, it isn’t the impact that is key, it is the motivation behind it. I doubt if any organization opposed to racial preferences objects to “disproportionately” high numbers of Asians admitted to elite campuses when this is the result of high grades and test scores. However, such organizations do oppose preferences based on race that benefit any group on that basis, including Asians.

    As for your statement “You might not object to [diversity] per se; you just object to any policy whose purpose is to promote it,” there is a world of difference between percentage programs on the one hand, and outreach programs that seek to inform minorities about opportunities at colleges, but do not establish intentionally differential admissions standards.

    Lastly, regarding the political implications of opposition to percentage programs, the same calculations you espouse have led to Republican (and obviously, Democratic) timidity about criticizing affirmative action, with unfortunate consequences for public policy. I am sure that just as preference programs in every other sphere are overwhelmingly unpopular, percentage plans based on racial considerations similarly are opposed by the majority of the electorate.

    any program to promote diversity will be illegitimate, which I believe is a big mistake.

    3. According to the writer’s argument, it would be unconstitutional for a university to drop the SAT requirement. (Not merely a bad idea, but unconstitutional race discrimination.) Whites in general do better on the SAT than blacks. If the SAT requirement is dropped, some whites will not be accepted who otherwise would have been, and some blacks will be accepted who otherwise wouldn’t have been. Dropping the SAT may (or may not) be bad educational policy, but I don’t think doing so would or should be illegal. (I’m sure the folks at ETS, however, would love for their tests to be constitutionally required.)

    4. The writer claims that “there is a world of difference between percentage programs on the one hand, and outreach programs that seek to inform minorities about opportunities at colleges, but do not establish intentionally differential admissions standards.” I just don’t see the difference. Outreach programs are aimed at recruiting minorities. Outreachers disproportionately encourage minorities to apply. True, outreach programs do not “establish intentionally differential admissions standards,” but then neither do the X% plans! What is “differential” about saying all students, without regard to race, must be in the top X% of their classes? That is a single, non-racial standard.

    5. The writer mentions that “set aside preferences for minority businesses,” and argues that they are not (or rather, should not be) saved by dollar cut-offs that exclude some minority firms. True enough. But these are set-asides for minority businesses. That is discrimination based on race. Set-asides for low-income businesses would not be discrimination based on race, even if minority firms disproportionately benefited from them.

    6. Re politics: it may be true that the reason I think the politics of pursuing this matter would be bad is that I think the legal argument is so unpersuasive. I know it’s not good form to use one’s self as evidence, but I can’t help thinking that if I — who is as committed to neutral color-blindness as anyone you’ll find — am not persuaded that these plans are so bad, can we reasonably expect others to be? I think a more likely result would be undermining support for the core principle of neutral color-blindness itself.

  8. Clayton E. Cramer November 18, 2002 at 5:31 pm | | Reply

    The intent of the 10% rule is to make sure that kids from the crummy schools–overwhelmingly black and Hispanic–get into the state university, even though they are going to be less prepared than kids from schools that have done a better job of preparing kids for college.

    This is clearly racially motivated, and even the proponents acknowledge that this is an end-run around the racial discrimination of affirmative action.

    Here’s the point that you need to remember: these aren’t abstractions. When the kid with a 3.5 GPA and a combined SAT score of 1250 gets told, “Sorry, we don’t have a place for you at Enormous State University” but a kid with a 3.0 GPA and an SAT score of 1050 gets in because he goes to a high school that is 60% black and 20% Hispanic, this is an unfairness.

    I remember, vividly, and with a bit of pain, graduating from high school. I had a 1390 SAT score, a 3.83 GPA (much of that in hard classes like accelerated physics, accelerated English, calculus, and so on). There was a kid from a wealthy family at my high school, who because she was a racial minority (back when East Indians were still considered victims) received a scholarship to attend Vassar that picked up not only tuition, but living expenses. I was from below the poverty line, but the wrong race (white), and the doors were far less open for me because of it.

    If we are going to just drop the pretense that merit matters, fine. It’s all a racial spoils system? I can live with that. But I want the pretense that merit has anything to do with college admissions to end. Merit has little to do with the process.

    Another point that isn’t being addressed is what happens to the kid from the crummy school who ends up at Enormous State University, and because of the lousy education he has received, can’t cut it? He drops out–while he might well have been able to make it ending community college for a couple of years first.

    The incentive to fix crummy schools also goes away with this scheme.

  9. John Rosenberg November 18, 2002 at 5:42 pm | | Reply

    I agree with everything Clayton Cramer says. I’m just not persuaded (nor, for that matter, is it clear that Cramer is persuaded) that universities are constitutionally required to reward merit. If they were, then not only would these Top X% plans be barred; so would preferences to legacies, people from strange or far-off places, athletes, etc. Again, I’m not sure any of those preferences are good ideas, but I don’t think they’re unconstitutional. Race preferences are (or should be).

  10. Clayton E. Cramer November 18, 2002 at 5:56 pm | | Reply

    Athletes: talk about the strongest argument against preferences imaginable!

    Race preferences are unconstitutional, and this “top 10% plan” is every bit the same as the grandfather clauses used at the beginning of the 20th century to dscourage blacks and recent white immigrants from voting. They didn’t discriminate based on race; they discriminated based on whether their grandfather had voted in the county before the Civil War–but everyone knew what both the intent and effect of those laws was.

  11. John Rosenberg November 18, 2002 at 6:06 pm | | Reply

    Grandfather Clauses (which limited voting registration to citizens whose grandfathers had been elibible to vote) applied only to blacks. The benefits of Top X% plans are not limited (in intent or effect) to minorities, and the disadvantages are not limited (in intent or effect) to whites.

  12. AMac November 18, 2002 at 6:09 pm | | Reply

    There is an additional aspect to John’s 10% plan that hasn’t been mentioned. Affirmative action and related plans require that individuals be identified by group. We take this for granted, e.g., notice how often the classifications “white”, “Asian”, “Hispanic”, and “black” are used in the preceding post and comments (not to mention in daily life).

    As a practicing biologist, this is something I find troublesome. Race and ethnicity have biological meanings, but nobody knows exactly what they are–yet we speak as if these meanings are obvious! To the contrary, these groupings are social constructs.

    Consider the person with four black grandparents. Black, obviously. Three? Two? One? Should society sanction the “one drop of blood” eligibility rule for blacks? Asians? Hispanics? How about whites? Should self-identification matter? The intricacies of these sorts of rules for group membership obviously get worse, not better, on reflection.

    In my opinion, we should pause before enacting new social programs that distribute benefits on the basis of racial or ethnic group identity. At the least, we should bear in mind that these groupings are largely subjective. If they are revised to be more objective (i.e. reflective of some biological reality), then they will become correspondingly more complex and difficult-to-use.

    At a minimum, top-10%-of-grads get around this particular b’rer patch.

  13. Jesse November 18, 2002 at 6:16 pm | | Reply

    What if the status quo admissions rule was to rank students by an index that was 0.5*SAT score + 0.5*high school GPA, and the legislature passed a law mandating that the index be changed to 0.25*SAT score + 0.75*high school GPA? Would that be unconstitutional? If it were, have we left any scope at all for the legislature to set policy?

    Would it change things if there was, in the legislative record, a statement by the bill’s sponsor that he believed SAT scores were unfair to minority students, and that the new index would be a more fair representation of student merit?

    Finally, would it be unconstitutional for a school to offer admissions preferences to students from broken homes or otherwise disadvantaged backgrounds, reasoning that it takes more merit to obtain a 1300 SAT score if you attend an underfunded school and your parents are illiterate than if you attend Plano High School (in a wealthy Dallas suburb) and have been going to Kaplan courses since age 12?

    I have a hard time imagining constitutional arguments against these policies that I would find even remotely compelling. I have an even harder time imagining such arguments that wouldn’t put all power over college admissions rules (and, for that matter, governmental hiring rules) under the control of the judiciary.

    However, I would be very interested to hear such arguments: I do research in this area, and I’m eager not to be attacking a “straw man” version of the argument against percent plans. I believe there is a strong argument to be made that SAT scores only effectively measure what high school a student attended; if admissions rules based on high school attended are unconstitutional, it isn’t at all clear to me that the use of SAT scores in admissions can pass that bar.

  14. John Rosenberg November 18, 2002 at 6:18 pm | | Reply

    Good points. The fact that no one knows what race is (in this context) or how to count it is indeed a serious flaw in race preference programs. And you’re right that the Top X% Plans do not depend at all on identifying the beneficiaries by race, and so do not suffer from this flaw.

  15. Jeff Bishop November 18, 2002 at 6:51 pm | | Reply

    I agree that public colleges are not specifically required to admit students according to merit. In the lottery example, I agree it would be constitutional if the principal purpose and effect were constitutional, and any effects on racial makeup largely incidental. Roger’s example, however, stipulates that the motive was to boost minority admissions through the back door, an unconstitutional purpose. If the lottery was intended to accomplish something else (e.g., a race-neutral effort to end the immoral practice of taxing stupid people to subsidize smarter people’s education), then the program probably would be constitutional.

    The top X% programs, however, are more problematic than a lottery because we all know what their main purpose is: to accomplish covertly what the courts will no longer allow to be done overtly. If circumventing court rulings were not the central purpose of these programs, then one would expect that at least some universities might have thought of the idea before courts started striking overt racial preferences down.

    For that matter, I wouldn’t be surprised if suspicious timing also have played a role in the challenges to literacy tests. After all, if the real issue had been to ensure that only literate/informed people vote, then that issue should have arisen among illiterate whites long before blacks were allowed to vote in large numbers.

  16. Plum Crazy November 18, 2002 at 6:54 pm | | Reply

    Promoting Diversity at Universities

    There is an excellent discussion of the constitutionality of the Top X% Plan enacted in Texas over at Discriminations. The

  17. Anonymous November 18, 2002 at 7:02 pm | | Reply

    Imagine the following: A state legislature passes a 10% plan for the main campus of its state university. The following year, the entering freshman class has twice as many blacks as the previous one and twice as many hispanics. Whites and asians are correspondingly down.

    It is well-known in the state that the plan was worked out at a late night meeting of the legislative leaders. What becomes known after a tv station shows it is that the entire meeting was taped.

    The leaders are tired and have been drinking. One of them expresses his frustration at how few blacks and hispanics there are at the state university. There is general mumbling that there should be more. Someone comments, though, that to have more blacks and hispanics, there will have to be fewer whites and asians. “It’s a matter of simple arithmetic.” “Yeah,” says someone sarcastically, “too many whites.” Someone else repeats it and after a while everyone is drunkenly chanting, “Too many whites. Too many whites.”

    Eventually, they calm down and come up with the 10% plan.

    Does we now have something unconstitional? Is it unconstitutional but desirable anyway?

  18. Roger Sweeny November 18, 2002 at 7:03 pm | | Reply

    Oops, left off my name. The above is mine.

  19. TJVM November 18, 2002 at 7:29 pm | | Reply

    Clayton Cramer’s story illustrates a point that has always rankled me about affirmative action policies: they don’t, in practice, seem to be focused on the truly disadvantaged. Proponents of affirmative action make heavy use of ghetto imagery: kids living in poverty, going to crumbling schools, walking home through crime-ridden neighborhoods, and so forth. But when you look at college campuses, at least the better ones, you don’t see many kids like that. What you see, at least in my experience, are a lot of minority kids who come from middle- or upper-class homes and who went to the same types of schools as their non-minority classmates.

    In that respect, I think the “top 10%” approach is an improvement over explicit race preferences. In order to benefit from that policy, you actually have to go to a crummy school and thus be personally disadvantaged–not just belong to a disadvantaged group. I’m not sure if this is fair, but I think it’s less unfair than giving preferential treatment to “disadvantaged” kids from good homes and good schools.

  20. Andrew Lazarus November 18, 2002 at 7:48 pm | | Reply

    I am one of the many people who will be following this link from The Volokh Conspiracy.

    I want first to salute the author for his hard work in distinguishing whether a plan is unconstitutional vs whether it is good educational policy.

    I feel the argument that the Top X% plan is unconstitutional is really pretty weak. It seems to boil down to a claim that once a State Legislature has its unconstitutional preference system stricken down, there’s nothing it can replace it with. After all, even if the Legislature merely mandated a hundreds of ads on WB and Univision TV, it’s clear that the intent is the same as for the preference system.

    I’ve also been converted to the idea that when X is sufficiently small (10% might be right, might be a little too high), the plan is sound educationally. There is indeed a cutoff below which students are likely incapable of successful work at Enormous State University, but if this cutoff is 90% at Suburbia High School and 10% (or even 3%!) at Ghetto High School, I’m not sure Enormous State will suffer overmuch. And there’s something for rewarding drive and initiative. What’s more, the Top X% system sets each student a goal that he/she can strive for, and that’s probably a good thing too.

  21. Andrew Lazarus November 18, 2002 at 7:52 pm | | Reply

    Side followup: it’s about five years before my time, but one of the problems with Southern literacy tests is that in practice much more difficult problems were set for black applicants.

    A great deal of American law arises from the unwillingness of the Southern states to deal justly with the “Negroes” (as we called them then). There’s something eerie about seeing a generation that truly seems to have forgotten the dynamics of that era.

  22. Phil Bennett November 18, 2002 at 9:28 pm | | Reply

    Doesn’t the constitutionality of the 10% really come down to how the Supreme Court decides the U. of Michigan diversity cases?

    This may be gross oversimplification, but if I recall, those cases ask the question of whether the interests of “diversity” are important enough to take race into consideration in admissions decisions.

    Here, the law is clearly intended to raise the number of minorities admitted to the school. Although the admission criterion itself does not take race into account, the policy underlying the criterion certainly does.

    So, I guess what I’m asking is this: Won’t we know whether the 10% plan is constitutional when the Supremes decide the Univ. of Michigan case?

  23. Steve November 18, 2002 at 9:32 pm | | Reply

    Will enrollment at County High School be “closed” to prevent the student from Suburban High School who is at 12% moving in and becoming 8% by virtue of his/her transcript?

  24. Andrew Lazarus November 18, 2002 at 10:49 pm | | Reply

    Will enrollment at County High School be “closed” to prevent the student from Suburban High School who is at 12% moving in and becoming 8% by virtue of his/her transcript?

    My experience is that principals of marginal schools love to have more students from higher socioeconomic strata, for what they bring to the PTA, bake sales, volunteer hours, etc. If admission to Bigname U is what it takes, hey great!

    I’m still a little confused by the assumption above that nixing one “diversity” program outlaws all of them. I take it that you expect not just an unconstitutional method but an unconstitutional goal?!

  25. John Rosenberg November 18, 2002 at 11:57 pm | | Reply

    Since I’ve started repeating myself even more than usual, I’ll stop commenting on every comment. I do appreciate all these comments that have come in, even the ones I disagree with. This is a vexing question, and I at least have found this discussion helpful. Here a few additional responses.

    Jeff Bishop writes that the motive of boosting minority admissions is an unconstitutional purpose. That is the essence of what I’m disagreeing with. What is unconstitutional, I believe, is awarding a burden or benefit to someone because of his race. It is the combination of intent and effect. Having a scholarship program limited to blacks is, and should be unconstitutional. Having a scholarship for poor people is fine, even though it is both known and intended that blacks will disproportionately benefit. I think the top X% plans are like the latter, not the former.

    Roger Sweeney’s hypothetical example of the legislators in the back room may not miss actual the Texas mark by much!

    Phil Bennett suggests that the constitutionality of the Top X% plans comes down to how the Supremes decide the Michigan case(s). If the Supremes uphold the Michigan policies, then this is true. Those policies used clear racial preferences to achieve diversity. They held minorities to a lower standard than whites. In effect, it’s as though Michigan had said whites have to be in the top 5% to be admitted; blacks only in the top 15%. So, yes; if diversity is held to be so important that even overt race preferences are acceptable to achieve it, then clearly the Top X% plans will be acceptable as well. If, however, the Michigan policy is barred, it does not follow that the Top X% plans will be barred as well, for all the reasons already discussed.

    Stephen Quick asks if transfers will be barred to prevent what might be called “top 10% hopping.” It’s hard to see how that could be done. On the other hand, who’d a thunk that school districts would prevent transfers based specifically on race, as many of them have done and probably still do? An example (such as one I discussed here) is when a school district denies a request for transfer because the change would add to “racial imbalance” at either the school the student wanted to leave or to attend. These have generally been struck down lately.

    Again, thanks to all for your comments.

  26. Roger Sweeny November 19, 2002 at 10:13 am | | Reply

    Thomas Schelling, in his wonderful “Micromotives and Macrobehavior” points out an unpleasant truth about tennis clubs that rank their members (this is done so good players don’t get in matches with bad players; the match would probably be a blow-out, unsatisfying for both players.). Such a club will unravel if people can’t accept being ranked last. The last-ranked person leaves, so the next-to-last is now last. She leaves, creating a new last person. And on and on. Some things are simple logical necessities, point out Schelling.

    A simple logical necessity is also involved here. Anything which deliberately catches “more blacks and hispanics” also NECESSARILY catches “fewer whites and asians.” Many people pretend that this simple, logical necessity doesn’t exist.

    Another way to react is to say that it is irrelevant because it’s just a byproduct of another good thing, something that Justice Powell years ago called “diversity.” (though as a matter of stict logic, diversity would have to mean more than just more blacks and hispanics, else the argument is circular: Why does warm milk put you to sleep? Because it has dormitive properties.)

    I think Phil Bennett is on to something. If the Supreme Court accepts diversity as a reason for allowing discrimination in the Michigan cases, 10% plans will also be declared constitutional. If they don’t, the plans will have harder going. Though the Court could easily pretend that the logical necessity of the second paragraph doesn’t exist and the plans are just peachy. (Hey, they said for years that Louisiana and other states had public schools that were separate but “equal.”)

  27. Roger Sweeny November 19, 2002 at 10:32 am | | Reply

    Hypothetical: The University of Mississippi decides to automatically offer admission to all direct descendents of Ol’ Miss graduates. Various reasons are given.

    Does anyone doubt that the courts would look beyond the surface here and say, “Almost all graduates are white, as are their descendants. Many of them would not get into Ol’ Miss otherwise. This is obviously a plan to increase the number of whites at the University of Mississippi. It is unconstitutional.”?

  28. Roger Sweeny November 19, 2002 at 10:47 am | | Reply

    Jesse,

    “What if the status quo admissions rule was to rank students by an index that was 0.5*SAT score + 0.5*high school GPA” and the state legislature passed a law requiring only SAT scores to be considered? It is well-known that this will increase the number of whites and asians at the university and decrease the number of blacks and hispanics.

    I’m sure there are some judges who would strike this down on its face (How many? Maybe John knows.) If the legislation were passed amid speeches saying there were too many dark faces at the university, just about every judge would strike it down.

  29. Brendan November 19, 2002 at 12:20 pm | | Reply

    Does anyone know what the actual, net effect of “Top X%” plans in any given state has been? Are there in fact high schools that are 100% minority such that the “Top X%” of all students in the high school are guaranteed to be minority?

    Assume that Ghetto High School is 90% black/hispanic and 10% white/asian. What happens if, over time, the “Top X%” of students at Ghetto High School end up being 9% of the white/asian kids and 1% of the black/hispanic kids?

    For the record, I agree with Mr. Rosenberg that these plans represent bad policy, but are probably not unconstitutional.

  30. Roger Sweeny November 19, 2002 at 1:57 pm | | Reply

    Pure situation 1: Top 10% plans have been discussed for years and gradually get more popular. A consensus develops that those who have worked hard enough to get in the top 10% deserve college admission even if the school is bad (after all, they didn’t pick the school). Besides, it will encourage people in bad schools, etc. The state legislature passes a law requiring the main campus of State U to automatically admit all those who finish in the top 10% of their high school class.

    Pure situation 2: Top 10% plans are occassionally mentioned but no one pays much attention or really takes them seriously. This doesn’t change over time. However, a series of court decisions say that a number of programs that boost black and hispanic enrollment are unconstitutional discrimination. State U has several programs like this. They do a study guesstimating what losing the plans will do. Black and hispanic enrollment goes down. Another study guesstimates that instituting a top 10% plan will actually increase black and hispanic enrollment. State U announces that it is discontinuing the programs that the courts have frowned on. And though the school has never considered a top 10% plan ever before, it presents this study to the state legislature and asks them to authorize one “to keep up our minority numbers.” The legislature does.

    I don’t think many people would charge “unconstitutional racial discrimination” in the first case. But a lot sure would in the second (even though almost none of them would be politicians or educational officials). Is the present situation more like the first pure case or the second?

  31. Celeste November 19, 2002 at 1:59 pm | | Reply

    I don’t think the 10% plan is bad policy at all. All it says is – if you make the top 10% of your graduating class, you are guaranteed admission to UT schools. Given the disparities between public schools even within the same district, this is a simple, objective criteria for acceptance. You work hard enough to make the top 10%, you deserve the chance to try and make it at college. I don’t see why everyone is getting their panties all in a bundle over it. It’s not like ONLY students in the top 10% of their class are admitted to UT schools, nor is the 10% rule limited to minority heavy schools. It applies to all of them, even almost entirely white schools in rural areas who don’t have the equipment or the same level of coursework available to them that their suburban counterparts might.

    It is certainly fairer than ‘weighting’ admissions, or quotas, since the criteria you have to meet are straightforward, and have nothing to do with your race, sex, or religion, just with how well you stand in your school. Such a method of admitting students is no sillier than national merit scholarships, and far less silly than automatically admitting children of alumni.

  32. Jesse November 19, 2002 at 2:59 pm | | Reply

    Roger Sweeny asks “What if . . . the state legislature passed a law requiring only SAT scores to be considered?” He suggests that that would certainly be overturned as unconstitutional. I don’t agree with that intuition: I would be, frankly, shocked if such a law were overturned. (Rather, I wouldn’t be shocked that _some_ judge ruled against it, but I would be if his or her opinion were allowed to stand.) Employment discrimination law is probably relevant here: Judges will let stand even employment tests with disparate impact as long as they can be shown to have predictive power for job performance. The College Board and Educational Testing Service have developed an industry of producing these validation tests, and there is plenty of evidence that SAT scores are correlated with college grades. You may be right that evidence that legislators were motivated by the thought that “there were too many dark faces at the university” would change things. But that brings me to a more general problem I have with your line of thought.

    You read the percent plans as directly intended to accomplish affirmative action by other means. I don’t read the legislative intent this way. Rather, I think there have always been legitimate concerns about whether SAT scores and other admissions measures were fair across racial and class lines. No one really had to worry about that as long as affirmative action was allowed: Any deficiencies in the test could be repaired by using it differently for each racial group. But once that is not allowed, it is much more important that the criteria work across as well as within races. And as I suggested above (and as Celeste expressed better than I), there’s a strong case to be made that percent plans do a better job of identifying merit than do more conventional measures. Is it so implausible to you that this was the intent of the legislature? Based on my conversations with University of California officials, it seems much more likely that this is what they intended than that they were trying to replace affirmative action by other means.

    Moving to Brendan’s question: I know most about the University of California, where the “4% plan” has had a negligible effect on admissions. By my estimates, there are about 50 kids in California each year who are UC-eligible under the 4% plan but not under the regular GPA+SAT eligibility rules. Where the 4% plan seems to have had an effect is in outreach: The UC sends letters to students in the top 4% of their classes after their junior years, encouraging them to apply and assuring them that they are automatically eligible. Since many high schools do not have good college counseling, many students who get these letters do not realize that they would be eligible under the regular rules, and applications by 4-percenters seem to respond substantially to the letters. So in practice, the California rule (Texas may well be different, BTW) really amounts to an outreach program.

  33. Sheldon November 19, 2002 at 3:13 pm | | Reply

    Steve asks “Will enrollment at County High School be “closed” to prevent the student from Suburban High School who is at 12% moving in and becoming 8% by virtue of his/her transcript?” The possibility that a 10% plan might have the indirect effect of desegregating public schools in this manner might be the best argument for a 10% plan. I’m not convinced it would happen, but I love the possiblity.

  34. Sheldon November 19, 2002 at 3:23 pm | | Reply

    The 10% plan exploits the reality of segregated secondary schools. What if universities instead use admissions to attack that segregated reality by giving preferences to students from diverse (racially/economically/whatever) secondary schools. If students/parents chose to attend diverse secondary schools to gain the admissions advantage, it would be a powerful tool for desgregation. If it were done in a race neutral manner (i.e., the preference is based on the composition of the student body, rather than the race of the applicant) would there be constitutional issues? It’s not clear to me what the result of such a system wold be on racial diversity at the Universtiy level, but I suspect that the benefit of increased secondary school diversity would outweigh the possible decrease in university diversity.

  35. Celeste November 19, 2002 at 4:05 pm | | Reply

    Sheldon –

    Such a system might be feasible in states/areas where there are diverse racial populations, but it in turn would punish any students living outside of large cities, and especially students living in largely homogenous rural areas. It isn’t uncommon to walk into a high school in the midwest and see a single black student there. And it isn’t because all the black kids are going to the school across the street…. there simply aren’t any black kids around to go to that school. Hell, its quite likely there aren’t any other schools around – unless you want to go private. That’s why I favor the 10% plan as much as I do. There simply isn’t any way you can force it to be a racial thing, unless you enforce segregation in the schools. Doesn’t matter if you come from a good school or a bad school, a rich or poor one, a black or white one, if you rose to the top 10% of your class, the University of Texas is willing to give you a chance.

  36. Roger Sweeny November 19, 2002 at 10:33 pm | | Reply

    I’m sorry I wasn’t clear enough. I feel fairly sure some judges would strike down the law right away without anything more. I assume they would be a minority, perhaps a small one and perhaps all overturned on appeal. But I don’t know. That’s why I added the parenthesis: (How many? Maybe John knows.)

    However, I am also fairly sure that if the law were passed with speeches about “too many dark faces” that it would be quickly and pretty unanimously struck down as racially discriminatory.

  37. John Rosenberg November 19, 2002 at 11:11 pm | | Reply

    Fritz Schrank of Sneaking Suspicions emailed the following comments. Ordinarily I would call them “thoughtful,” but since he agrees with me I think they’re extra thoughtful.

    I’ve read your post and extended comments and some of the other pieces on the 10% controversy. Personally, I believe you have the better argument on its constitutionality.

    Any normal disparate impact case, at least in the Title VII arena in which I practiced, relied heavily upon a showing of actual impact, usually expressed in terms of a statistical significant difference in results. From what I’ve read about the 10% plan, it is not the sole means by which Texas residents can gain admission to UT-Austin or other state colleges. Before one discusses motive, therefore, can we at least see some statistical showing that the percentage plan has a disparate impact? I’m not suggesting a return to the kind of group performance defense found wanting in Teal v. Connecticut, but how can the opponents of the plan show a disproportionate impact on whites and Asians, say, by the use of the 10% plan?

    There are plenty of other points to argue about the plan other than its constitutionality. It’s a blunderbuss approach to school admissions, with no consideration for the equally unfortunate disparity in education quality made available to students throughout Texas. OTOH, if the effect of the plan is to force others to seek to upgrade educational opportunities elsewhere than in their own sheltered environments, even if only done for selfish motives, that would be a good thing.

    Fritz’s point reinforces Brendan’s about the need to look closely at the actual effects of the plan.

    Another reader emailed a point about the numbers that I think is widely shared but, as I hope we’ve seen, open to debate:

    First, and most obviously, there are only a limited number of slots at the institutions applying the 10% rule; thus, every person admitted solely as a result of that rule who would not have been admitted otherwise represents an unfair result. This is unfair not just for that one student with better test scores and grade point averages who would have been admitted in the first applicant’s place, but for EVERY such student who might have been admitted.

    Two points: 1)The by-now familiar one that “unfair” does not necessarily mean illegal or unconstitutional; and 2) the alleged unfairness here both assumes and requires that in the absence of the top X% plan students “better test scores and grade point averages” would have been admitted. Why? No state has any obligation to admit on merit. Perhaps more athletes or legacies or out-of-state students would have been admitted had the top X% plan not been in effect. In short, it is hard to identify actual victims of the alleged discrimination.

    And speaking of legacies, in one of his provocative comments Roger Sweeney posits a hypothetical in which Ole Miss offers automatic admission to direct descendants of alumni and asks if anyone doubts the courts would strike this down. Actually, I’m not sure what courts would do, at least in the absence of compelling evidence that the actual intent was to keep out minorities. We’ve already discussed the Grandfather Clauses, which this at least superficially resembles. What we haven’t discussed, and what this hypothetical also very closely resembles, is veterans’ preferences, which have been upheld even though when the Supreme Court case was decided (1979) something like 99% of veterans were males. [Personnel Administrators of Massachusetts v. Feeney, 442 U.S. 256 (1979)]

  38. Roger Sweeny November 20, 2002 at 2:41 pm | | Reply

    John,

    I’m surprised you think the courts might give a free pass to a plan that automatically admitted all direct descendents of Ole Miss alumni to Ole Miss.

    What about a plan that automatically admitted all direct descendents of Ole Miss students who graduated before 1965? I would think that that would be obviously unconstitutional because Ole Miss was legally all-white before that date.

    So how about 1975? You’ve got a few black alumni at that point.

    And how about 1985? Which, since those born in 1985 are just 17 this year would have the same practical effect as a plan that admitted the direct descendents of all alumni (though the effects would diverge as time goes on).

    If “the law is concerned with substance, not form,” it seems to me they all fall.

  39. John Rosenberg November 20, 2002 at 2:49 pm | | Reply

    Roger,

    Before 1965: analogous to Grandfather Clauses, and would no doubt be rejected. What purpose and intent could such a policy have other than to exclude blacks:

    After 1975 or so: analogous to veterans’ preferences, which courts have consistently upheld even though 99% of veterans were male. Your hypothetical of an extreme legacy preference would no doubt be justified by some hypothetical legislators who supported it as beneficial to the university by building loyalty across generations, increasing donations, etc., i.e., the same way the nearly ubuquitous actual (though milder) legacy preferences are actually justified.

  40. Brian Fitzpatrick November 21, 2002 at 12:14 am | | Reply

    This is an amazing site. I have read the debate over the Texas Ten Percent Plan with great interest. I am convinced the Plan is not only unconstitutional, but clearly so. For those who have an interest in this legal question, you may want to see my law review article on the subject, which lays out the legal argument against the Plan in gory detail: Strict Scrutiny of Facially Race-Neutral State Action and the Texas Ten Percent Plan, 53 Baylor L. Rev. 289 (2001).

    I would like to respond to a few of the issues that have arisen on this site:

    First, just because a law is facially race-neutral does not mean that it cannot be struck down as unconstitutional racial discrimination. The reporters are full of such cases, many of which I cite in my law review article, and two of which I cited in my Op-Ed in the Washington Times.

    Second, on the other hand, it is unquestionably true that a law cannot be struck down as unconstitutional racial discrimination based solely on the motivations of the legislators who enacted it. The Supreme Court has made crystal clear, as many on this site have noted, that in order to be unconstitutional, a facially race-neutral law must have both the purpose and the effect of racial discrimination.

    Third, I agree that it is difficult to make out a case that the Plan causes the “effect” of a disparate impact. In my law review article, I do spin out an argument that it does, but the argument is weak.

    Fourth, nonetheless, and contrary to some of the posts on this site, a disparate impact is not the only “effect” that counts in the Supreme Court’s purpose/effect test. This became clear in the 1990’s with the cases on gerrymandered electoral districts. The Court held that 1) laws drawing electoral districts are facially race-neutral because they classify people according to where they live; 2) the laws under consideration did not cause a disparate impact because they did not dilute voting strength of one race relative to another; 3) but the laws were unconstitutional because they caused a different kind of racial effect, and were enacted with the purpose of so doing; 4) this racial effect was to change the racial composition of voters in the electoral district. That is, the Supreme Court held in these cases that changing a racial composition alone–even without a cognizable disparate impact–is a racial effect that, when combined with the purpose to cause the effect, invalidates a statute.

    Five, in light of these cases, there can be no doubt that the Plan is unconstitutional. It changed the racial composition of Texas universities, and that was the purpose for which it was enacted. End of story.

    Six, some have raised the question of what is the point of seeking to invalidate the Ten Percent Plan when another state could do the exact same thing for a different purpose. It is true, for example, that Texas could, perhaps, reenact the Ten Percent Plan under the auspices of geographic diversity. But even that end-run represents a victory for color-blindness. What a legislature says when passing laws is important. If it is acceptable to talk about life through the prisim of race, then not only will more people talk about life as such, but they will think about life as such. If, on the other hand, such sentiments were forced into the closet, then the hope is that race-conscious thoughts will die out more quickly. Ushering in a future where people not only act but think without regard to skin color is a very important goal. Regulating what it is acceptable for legislators to say when passing laws is one way to help bring about that future a little sooner.

  41. John Rosenberg November 21, 2002 at 4:45 pm | | Reply

    Liberals think diversity so important that it is acceptable (some argue it is mandatory) to discriminate on the basis of race to achieve it. The University of Michigan will be arguing this position in the Supreme Court. Critics of race preferences usually argue that diversity is nice, maybe even extra nice, but not nice enough to justify racial discrimination in order to achieve. Achieve it by non-discriminatory means, we have argued, just as many who oppose race preferencess have argued in favor of class- or income-based prefs.

    The Comments here, however, have revealed that many critics of diversity believe that anything a state does whose purpose is to promote racial/ethnic diversity is, by virtue of the purpose alone, impermissible. Despite Brian Fitzpatrick’s powerful argument, I continue to believe that view is probably wrong as a prediction of what courts will do, and almost certainly wrong as a matter of politics. Ironically, the only area where I find the criticism of the top X% plans compelling is educational policy, where I find them deeply disappointing.

    I look forward to reading Fitzpatrick’s Baylor Law Review article. Based on his Washington Times OpEd and his comment here, I’m sure it will be impressive. Since I have already given my opinions on his numbered points (and repeated some of them already) I’ll not respond to them all here. I also would like to commend Fitzpatrick’s comparison to the redistricting cases, since those cases and the questions surrounding the top X% plans do overlap to a substantial degree.

    Fitzpatrick may well be right about how courts will respond to the top X% plans, and I should (but won’t) wait to comment further until reading his article, but as I read the redistricting cases the answer is not as “clear,” “end of story” as his comment asserts. (It is possible that this is because the last major case, Hunt v. Cromartie, 121 S. Ct. 1452, was decided in 2001, the same year his Baylor article was published. I don’t know if that case was covered in the article.)

    The redistricting cases arising from North Carolina’s 12th district are fascinating. They involve, as does our discussion here, the degree to which it is legitimate for the state’s purpose to include certain racial outcomes. The Supreme Court examined that district in four separate times, and each time it reversed the district court’s opinion, although not in the same direction. The earlier cases resulted in the district being invalidated because race was the “predominant factor” in how the district was drawn; the last case upheld the district lines, even though race was taken into account and the legislature (or at least an influential legislator) had mentioned “racial balance” as one of its goals. That these issues are not “clear,” at least to the current court, is indicated by the fact that these decisions were 5-4, with Sandra Day O’Connor being the swing vote.

    In one sense, I believe Fitzpatrick may exaggerate the similarity of these cases to the top X% plans, for in the redistricting cases race was much more explicitly involved. That is, the legislature assigned different precincts to different districts based in large part on their racial makeup, arguing that it was trying to create a safe Democratic district and blacks were the most reliable Democratic voters. Briefly and too simply, this explanation was ultimately accepted by the Supremes.

    As Justice Breyer wrote for the majority in Hunt v. Cromartie, the 2001 case, the test is not whether racial motives were present but whether “the legislature’s motive was predominantly racial.” Moreover, plaintiffs “must show that a facially neutral law is unexplainable on grounds other than race.” He quoted Justice O’Connor’s opinion from one of the earlier cases — “If district lines merely correlate with race because they are drawn on the basis of political affiliation, which correlates with race, there is no racial classification to justify” — and concluded that the test of racial predominance is “a demanding one.” Where is the “racial classification” in the top X% plans?

    Breyer’s conclusion (in a majority opinion that was majority because O’Connor, the author of the original opinion invalidating these racial gerrymanders, signed on) is directly relevant to our discussion here:

    We can put the matter more generally as follows: In a case such as this one where majority-minority districts (or the approximate equivalent) are at issue and where racial identification correlates highly with political affiliation, the party attacking the legislatively drawn boundaries must show at the least that the legislature could have achieved its legitimate political objectives in alternative ways that are comparably consistent with traditional districting principles. That party must also show that those districting alternatives would have brought about significantly greater racial balance.

    I am not arguing here that Breyer and O’Connor were right. The Rehnquist/Scalia/Thomas/Kennedy dissent was powerful indeed. But the fact that they were in the minority on an issue close in many ways to the top X% issue does at least suggest, I think, that this issue is not “end of story” clear.

    The essential question here, I believe, is whether or not promoting racial/ethnic diversity can be a legitimate state objective if pursued by facially neutral policies that provide advantages and disadvantages to all racial and ethnic groups (remember those rural white who benefit). If a state’s attempt to promote diversity by racially neutral means is invalid because of its very purpose, then by definition no state attempt to promote diversity is legitimate.

    Finally, I still doubts courts will reach this conclusion, and I continue to think that is not a goal critics of racial preferences should pursue.

  42. Curtis Crawford December 3, 2002 at 12:51 am | | Reply

    Mr. Rosenberg, in your post of 11/18, 4:26 pm, you say that you are “as committed to neutral color-blindness as anyone you’ll find.” Now, whether the Texas plan is good or bad, it surely is not color-blind. You and your interlocutors agree that it was designed to increase the proportion of blacks and Hispanics admitted to Texas universities, and has had that effect.

    But you believe that the plan, though race conscious, is better than formal racial preference, worth securing as a political compromise, and may be constitutional. You also point out that anti-preference critics of the Texas plan seem to hold positions concerning the desirability of racial diversity and the original meaning of affirmative action that are race-conscious.

    I am not a moral absolutist in my opposition to racial discrimination (different treatment based on race or ethnicity). But it does surprise me when people strongly opposed to racial discrimination defend the Texas plan. I cannot imagine their defending such a plan, if the intended and effected racial results were reversed.

    I agree with you that there is an inconsistency in adopting racial diversity (meaning an increase in minority participation) as an end and repudiating racial discrimination as a means. However, there need be no contradiction between believing a change desirable, and rejecting it as a goal of public policy. For example, other things being equal, I think it would be better if the racial gap in educational and social achievement between blacks, Latinos and whites were smaller. But I think this reduction should not be a goal of public policy.

    I also agree with you that there is an inconsistency between nondiscrimination and some early ways of understanding affirmative action. If the reason for widening outreach to possible applicants was to increase the proportion of minority applicants, the purpose was indeed discriminatory. But if it was to reach a larger proportion of the persons, regardless of race, who might be qualified for the position, its purpose was non-discriminatory. Clegg rightly emphasizes that President Johnson’s executive order required affirmative action to ensure that decisions were made regardless of race.

    Racial diversity, defined as increasing the proportion of blacks and Latinos in higher education, business and the professions, is a popular social goal, generally uncontested in the universities, the law schools, professional organizations, the media and the political parties. So you may be right that it is unrealistic to reject this goal. But this is a profoundly race-conscious goal. I believe that race-conscious policy is fundamentally unsound, especially for a country that depends on the loyalty and cooperation of persons of many races and ethnicities.

    Congratulations for stimulating such an extensive discussion of this issue on your site. May I invite you to visit http://www.DebatingRacialPreference.org Your comments most welcome.

  43. John Rosenberg December 3, 2002 at 1:10 pm | | Reply

    Curtis,

    First, thanks for your thoughtful comments. They prove, as do a number of other very perceptive comments on this topic that take issue with my position, just how difficult an issue it is. Attempting with what I am sure will be only limited success not to repeat myself, I will offer only a few additional points here.

    First, you say the Texas plan “surely is not colorblind.” I’m not so sure. The intent is surely to promote diversity, and thus is not colorblind, but the means are colorblind: a preference is given to those of whatever color who graduate in the top 10% of their classes.

    You thus argue that any program whose intent is to increase the number of minorities (or, you would presumably agree, whites) is ipso facto discriminatory, no matter the means. Thus you would seem to have to agree with the affirmative action forces that a university giving preferences to legacies, or oboe players, is racial discrimination if the university’s alumni or the pool of oboe players contain disproportionately few blacks. I don’t agree.

    And then there is the problem of intent, which is a somewhat slippery concept. There is the obvious meaning of doing something on purpose, or to accomplish a conscious purpose. But it is also sometimes thought to cover things less obvious: doing something with the foreknowledge that it will have a certain effect, even if that effect is not your reason for doing it. Under this conception of intent, it would be discriminatory to require SATs if the univ. knew that doing so would reduce the number of qualifed black applicants, just as it would be discriminatory to stop giving the SATs since doing so would increase the number of qualified black applicants. It would also seem to call into question a program, say, to equalize schools across an entire state, at least insofar as a purpose for doing so is to close educational gaps between minorities and others.

    Since the Texas plan benefits all races — poor whites from rural schools also benefit — and since its means are colorblind, I do not find the program discriminatory simply because one of its intended effects is to increase the number of minorities in Texas colleges.

Say What?