Confronting affirmative action.

Because the administration argued against the two specific University of Michigan affirmative action plans before the Supreme Court last year, few appreciated at the time how much of the constitutional principle of Equal Protection it was willing to sacrifice. For example, Solicitor General Theodore Olson argued in his brief, "[T]his case requires this Court to break no new ground," in effect conceding that race preferences to achieve diversity, articulated in the Bakke case a generation ago, remained good law. And, said Olson, while not considering race per se, a school might consider such matters as "a history of overcoming disadvantage, geographic origin, socioeconomic status, challenging living or family situation," and a host of other technically race-neutral factors. Got the point (wink, wink)?

The solicitor also pointed with pride to the "percentage plan," adopted by Texas under the Bush governorship. The plan got around a circuit court decision barring race-conscious university admissions by automatically admitting high school students from any public high school graduating in the top 10 percent of their class.

Now comes word, via the New York Times, of widespread dissatisfaction with the Texas percentage plan. The reason is the plan's built-in bias in favor of students' graduating from lowachievement schools and against those in the tougher academic institutions where many superior students fail to make the top 10 percent. As a result, many turned down by the flagship University of Texas take their talents to other states.

This problem was totally predictable at the time the law was passed, as were many others inherent in percentage plans. For one thing, to "cure" the racial imbalance problem, they rely on a foundation of de facto segregated high schools. For another, they have no applicability to graduate schools or to private or out-of-state high schools.

The diversity approach sanctioned by the Court, to which dozens of colleges and universities are now scrambling to adjust, has its own set of problems. Diversity itself is a vague concept since it can relate to philosophical viewpoint, religious conviction, and areas of academic interest as easily as to race or ethnicity. Diversity is also subject to diminishing returns. The first African American in a history or political science course arguably could bring important perspectives to bear on many subjects. But what about the 2nd? Or the 17th? And suppose the course were biology? Or chemical engineering? Or medicine? There is at this point overwhelming evidence that affirmative action at medical schools has produced lower-achieving students who take longer to graduate, do less well on the medical boards, and materially less well in qualifying to practice their specialties.

As a matter of principle I continue to oppose the allocation of rights and rewards based on race. I wish the administration had shown a bit more spunk in the leading Michigan cases. But a decade of attention to this area also persuades me that this society is unwilling to accept the results of a raceneutral approach--without gimmicks or outright intellectual fraud--to college admissions. For that reason I believe the least harmful approach is a straightforward affirmative action plan modified to avert harm to innocent nonminority students. This could be achieved by enrolling one nonminority student who would otherwise have been rejected for each minority student enrolled through affirmative action. It is time to restore integrity to issues of race and university admissions.

Robert Zelnick is a research fellow at the Hoover Institution and chairman of the Department of Journalism at Boston University.