On the Contrary:

Nearly 25 years ago, in Regents of the University of California v. Bakke, the U.S. Supreme Court held that educational institutions may consider the race of their applicants in making admissions decisions. But the Court didn't clarify the constitutional justification for racial preferences. Are they permissible only when offered as remedies for previous discrimination (to level the playing field)? Or may racial preferences be employed in order to achieve a university administration's vision of racial balance or diversity? (I'm using the term "racial" loosely, as we often do.)

There's a profound difference between these competing notions of affirmative action. Envisioning racial preferences as purely remedial greatly limits their use: An institution that hasn't discriminated against racial minorities in the past may not discriminate in favor of them in the present, and institutions that are guilty of prior discrimination may, in theory, only employ racial preferences remedially, until the prior offense is cured. But if racial preferences are justified in the name of diversity, they may be used by any institution, regardless of its history, and they may be used forever, to benefit any demographic group favored by institutional authorities. It's not surprising that liberals generally promote affirmative action as a means of achieving diversity, whereas conservatives, if forced to accept affirmative action at all, would only allow it to be used remedially.

Because affirmative action pits the rights of disadvantaged groups against the rights of individuals (and equality against liberty), the problems it poses are not easily or peacefully resolved. And so decades-old battles over group preferences haven't diminished in ferocity, as a recent federal court ruling shows. On May 14, in Grutter v. Bollinger, a closely and bitterly divided 6th Circuit Court of Appeals in Cincinnati upheld the University of Michigan law school's affirmative-action program, which has no remedial purpose. The law school's admission policies were expressly designed to achieve diversity.

In Grutter, the 6th Circuit narrowly ruled that the Supreme Court had endorsed diversity-based affirmative action in Bakke. (Other federal courts have disagreed, and the Court is likely to address this conflict soon.) The majority in Grutter also concurred with the University of Michigan that racial diversity advances democracy and benefits all students. (Those rejected for being insufficiently diverse may disagree.) Grutter held that the University of Michigan plan does not impose a constitutionally prohibited quota system so much as it enshrines a lawful effort to establish a comfortable, "critical mass" of students from target groups. The Michigan plan was characterized as a "plus" system, which allows the race of an applicant to be considered in admissions decisions along with other "soft" factors, such as the applicant's life experiences, talents and extracurricular achievements. The fact that members of some racial groups might be admitted with lower grade point averages and LSAT scores than applicants from other groups did not convince the court that the university maintained a two-track admission system based on race.

Considering the arbitrary nature of admissions decisions, with or without racial preferences, it's hard not to sympathize with the majority view. Before colleges and universities employed racial preferences, they relied on class preferences, favoring graduates of particular schools or the children of alumni, as advocates of affirmative action regularly point out. When I entered law school in 1972 (just before the implementation of federal equal-education guarantees), graduate and professional schools openly maintained generous affirmative-action quotas for men. These generated virtually no opposition from conservatives.

But if the fallacy underlying much conservative opposition to sexual or racial preferences is the assumption that without them life would be a meritocracy, liberal advocacy of affirmative action often reflects another fallacy: the assumption that the use of group preferences is cost-free and that the socially desirable goal of racial and ethnic diversity can be met without harming individuals or violating fundamental liberties. Racial (or sexual) profiling is, at the very least, problematic, whether it's employed by bigoted police officers or well-meaning educational administrators.

The dissent in Grutter v. Bollinger focused on the harm affirmative action can do individuals even as it benefits particular groups: "The fundamental premise of our society is that each person is equally 'diverse'" and endowed with equal rights, which are violated by official racial quotas. A "critical mass" is nothing more than a quota, the dissent asserted, and in any case the concept is spurious. The number of students deemed to constitute a critical mass of a particular group is usually determined by the group's share of the population. So the quest for critical mass masks a quest for proportional representation. (That's why the critical mass the university deems necessary for the comfort of Native Americans is so much smaller than the critical mass sought for African Americans.) Exactly how does the university determine which applicants belong to a preferred racial group, anyway? Should university administrators second-guess an applicant's self-identification? If you have an African-American or Latino grandfather, do you qualify for a racial preference? As the dissent pointed out, the path toward diversity through racial preferences can lead to official determinations of racial status reminiscent of the Nuremberg race laws. This doesn't mean that affirmative action is never justified; I have always considered remedial forms of it regrettable necessities. But it does demonstrate the thoughtless perversity of assuming that allocating rights, resources and privileges on the basis of race or sex to engineer diversity is an unquestionable, unmitigated good.