The Supreme Court got two decisions right yesterday, more or less by accident. The Court properly struck down a "points-based" system intended to achieve student diversity in undergraduate admissions at the University of Michigan, and it properly upheld a more nuanced system of preference in the University's law school that is intended to achieve the same goal. Unfortunately, it did so largely for the wrong reasons.
The 6-3 decision to strike down the undergraduate admissions policy in Gratz v. Bollinger on Equal Protection grounds may be even more clear-cut than the raw votes suggest, since Justice Stevens's dissent did not touch the merits of the case, but concerned itself exclusively with the narrow question of whether the petitioners had standing to challenge Michigan's rule. Yet many of the Court's objections to the policy—which assigned 20 of the 100 points needed to secure admission on the basis of membership in an underrepresented minority group—are puzzling.
Consider the question of whether race is a "decisive" factor, or merely "one among many." It's certainly true that a system that grants a student more than six times more points for being Native-American than it does for writing an outstanding essay is placing a lot of weight on race. But what could be the point of considering a factor at all unless it is "decisive" along some margin? Any factor that's taken into account will be decisive for some students, or there would be little point in looking at it. What the court really seemed to be objecting to was the size of the margin along which race was decisive, but it should be obvious that there are no bright lines to be found here.
The Court's focus on the distinction between individualized consideration and mechanical quotas is more promising, but doesn't get fleshed out in a satisfying way. The precedent established by the controlling Bakke decision is that considering race in admissions is fine, but quotas, or systems that de facto function as quotas, are verboten. The problem is that if a school is serious about considering race as a means to achieving diversity, its procedure must be, to some extent, quota-like and group oriented. The contribution an individual makes to the diversity of the student body is, by definition, not a function of his own characteristics alone, but of the total composition of the incoming class. True, the quota may be a fuzzy one, but to aim at diversity, of which race is one component, means necessarily to weigh race just enough for each group to get the appropriate United Colors of Benetton look on orientation day.
The phobia about precision evidenced by the Court is also mysterious on its face, but it shows the path to a more sane way of thinking about race in law. At first blush, there's something strange about saying that a precise 20 point "plus" for race is unacceptable but that it's fine to let admissions officers "eyeball" a candidate and factor in race in some ad-hoc fashion. Why is a preference applied fuzzily and implicitly less objectionable than one applied uniformly, out in the open?
The answer, and I suspect the deep root of the Court's willingness to accept the more holistic admissions procedure challenged in Grutter v. Bollinger, is that "race" is never monolithic, not well captured by the blunt sorting of check-off boxes. If we take universities invoking the diversity rationale for affirmative action at their own word, they are not pursuing the merely tokenist diversity of the Benetton ad. Rather, race is supposed to act as a proxy, albeit an imperfect one, for diversity of experience and cultural background.
There is something to this. Racial diversity is obviously not the only kind schools should (and do) worry about, but the experience of living as a member of a minority group in a predominantly white culture is unique: It cannot be captured without looking at race. Neither, however, can it be captured by looking only at race, in a mechanical way. The Court's unease with Michigan's automatic "20 points for race" policy is well motivated, because "race" is not monolithic. We care about race, not because there's something intrinsically significant about different skin pigmentations, but because of its social significance. Race is, above all, a social construct. Though notions of racial "authenticity" have been misused to bully, for example, black conservatives, our justified public interest in race is an interest in race as something that is lived, not in race as a mere genetic trait.
We are stuck looking at race through a legal lens ground in the 1950s. Following precedent, the Court in the Michigan cases saw the use of race in a state classification scheme. Since race is always a "suspect class," this automatically triggered 14th Amendment "strict scrutiny," requiring a "compelling state interest" to override the presumption against it. The sense that some racial classifications are genuinely innocuous prompted them to cast about for a compelling interest characterizing those instances, leading to the conclusion that the quest for diversity fit the bill. This is almost certainly a mistake. The language of "compelling state interests" is not unique to Equal Protection Clause jurisprudence: The same standard is used in the context of First Amendment cases, among others. Unless we are prepared to allow that a concern for diversity can also override free speech protections, or to admit that the 14th Amendment establishes a far weaker barrier than the First, we should be extremely wary of what we allow into this category.
Instead, perhaps we have reached the point where we can drop the assumption, necessary half a century ago, that any state consideration of race is automatically suspect, requiring the same kind of heightened scrutiny that would be triggered by Jim Crow: The Sequel. Instead, the role race plays in decisions, the way in which race is considered, must shape the legal approach. Mechanical tokenism deserves the strictest scrutiny, and the Benetton diversity it would achieve is scarcely compelling. Consideration of "lived" race, on the other hand, may not merit the same degree of resistance.
The typical objection to affirmative action is that it is unfair to "more qualified" Caucasian or Asian students to pass them over in favor of "less qualified" minority applicants. But this makes sense only if one accepts a reified notion of "merit," according to which SAT scores and grade point averages constitute a uniquely authoritative and just way of allocating spaces at public universities. Whether or not one approves of a government role in higher education, the purpose of state schools is manifestly not to hand out gold stars and smiley faces to the smartest students, as measured by standardized testing. Neither does the narrow conception of SAT-score meritocracy accurately reflect the way college admissions have ever been handled in the United States, by public or private institutions. The value of a college education is a function of the caliber of the student body no less than of the faculty, and the benefit each student gets from his fellow students has to do with far more than test scores. Lived race, as opposed to mere pigment variation, is as valid a component in building a varied student community as diversity of student viewpoints. Schools don't and shouldn't seek to maximize any single dimension—musicality, athleticism, activism, test scores. They try to achieve the balance that gives the college experience a texture and richness that students themselves have come to expect and demand.
The distinction between lived-race and tokenist policies does not map neatly onto that between race-conscious and ostensibly race-neutral policies. Opponents of any form of affirmative action often suggest that, as an alternative to explicitly race-conscious admissions, schools may adopt something akin to the "ten percent plan" used in Texas, where slots at state schools are guaranteed for any student graduating in the top ten percent of her high school class. This is as stupid as it is disingenuous: there is no good reason to break down applicants by origin school, taking the top tenth of each class rather than the top tenth of the applicant pool as a whole, except as a means of indirectly achieving a desired racial balance. Only the most hardened formalist could see any advantage in carrying out the same old policies via a Rube Goldberg-like roundabout that trades on the persistence of de facto geographical segregation at the high school level.
Contrast that with the approach taken by Rice University following the 5th Circuit's 1996 Hopwood decision, which forbade Texas schools to engage in explicit affirmative action. In addition to intensifying efforts to encourage qualified minority students to apply, Rice added to their application form an essay on "cultural background." Jeffrey Rosen derides this as a dodge akin to 10 percent plans in his recent New York Times Magazine article "How I Learned to Love Quotas." But Rosen's derision is misplaced. Rice was forced to rethink what "race" meant, and though minority enrollment dropped off precipitously immediately after the ruling, it rose within a few years to pre-Hopwood levels. The difference was that the school was no longer considering race according to the crude dictates of the checkbox, and admissions officials concede that the quality of the student body has been improved as a result.
As Rosen notes, this approach is not without its disadvantages. It perversely encourages upper-middle class minority students to attempt to portray themselves as victims of racism, dredging up any slight as evidence of systematic discrimination. There is also the danger of encouraging an artificial identification with race as one more resumé bullet-point: the student who might prefer to devote time to the National Honor Society decides she'd better aim for the presidency of the Black Student Union instead. But this may be a transitional phenomenon: Considering lived race can mean much more than merely rewarding narratives of victimization as mechanically as crude group-membership once was.
The Supreme Court's mixed rulings are bound to touch off years of confused revision of university admissions policies. The Justices' sound instincts have been distorted by a legal framework that lacks the crucial distinction between regard for lived-race and tokenism. That same deficit will be reflected in the policies shaped by their ruling. In the narrowest sense, the Court made the right decision. At what cost remains to be seen.