In my criminal procedure class this year, we tried to decide whether a driver who has been pulled over by the police because of his race has suffered a constitutional injury. "The problem isn't being pulled over," said one African American student. "It's what happens after. You have to do this step-'n'-fetch-it routine, showing that you're subordinate to the officer, to make sure he won't arrest you. Even if the officer is black, it's incredibly degrading."
The comment changed the way the class and I thought about the constitutionality of racial profiling. And it illustrates a proposition that now strikes me as obvious: There can be a meaningful correlation between racial diversity and diversity of viewpoints in the classroom. Before I started teaching, I was more skeptical of this correlation--and expressed some skepticism in these pages. And, unfortunately, some judges, who have not experienced the classroom dynamic I've described, remain skeptical. At the end of March, a federal district judge in Michigan, Bernard A. Friedman, struck down the University of Michigan Law School's affirmative action program, noting that "the connection between race and viewpoint is tenuous, at best." If the Supreme Court ultimately accepts this reasoning and forbids affirmative action in all public universities, it will not only be substituting its untested judgment for the classroom experience of teachers across the nation. It will also enmesh the courts in an impossible effort to disentangle the relationship between racial diversity and viewpoint diversity--a relationship that, as recent voting rights decisions show, is best defined by the political process.
The connection between racial diversity and viewpoint diversity is central to the legal future of affirmative action because of the peculiarities of Justice Lewis Powell's compromise opinion in Regents of the University of California v. Bakke. The real reason most public universities have affirmative action programs is that they feel it would be politically untenable to admit a class much less diverse than the state as a whole. But Powell rejected this representational justification, calling it " discrimination for its own sake." Instead, he announced, universities may legitimately seek a racially diverse student body because the diversity of viewpoints that resulted would have educational benefits for all students. For nearly 20 years after Bakke was decided, everyone accepted Powell's reasoning as binding Supreme Court precedent. But, in 1996, a Texas appellate court decided that this part of Powell's opinion hadn't commanded five votes in 1978 and wouldn't command five votes in the Supreme Court today, a prediction Judge Friedman embraced as well.
In trying to predict how the Supreme Court would decide Bakke today, Friedman and the Texas court have acted more like pundits than lower-court judges (and in 1978 five justices clearly joined Powell's conclusion that " the State has a substantial interest that may legitimately be served by a properly devised admission program involving the competitive consideration of race and ethnic origin"). But, regardless of how you interpret the constitutional history of affirmative action, its future will be decided by Justice Sandra Day O'Connor, who has inherited Powell's position in the middle of a divided Court. For this reason, defenders of affirmative action are intent on convincing O'Connor that there is, in fact, a meaningful connection between race and viewpoint.
On this point, the University of Michigan's litigation strategy may not have helped its case. Instead of sticking to the modest claim that there's often some correlation between racial diversity and viewpoint diversity in the classroom, the university made a more ambitious argument: that students who experience the most racial diversity in and out of the classroom show the greatest academic and intellectual growth. The university submitted a study by Patricia Gurin, a psychology professor, who reported that students who attend ethnic-studies courses and participate in racial-awareness workshops are "better prepared to become active participants in our pluralistic, democratic society." But, as the National Association of Scholars pointed out in a rebuttal, Gurin doesn't establish how many ethnic minorities, if any, actually attended the workshops; and, in any event, ethnic-studies workshops surely aren't what Powell had in mind when he talked about the benefits of an integrated classroom.
Because those benefits aren't always reflected in grades, test scores, and other tangible measures of educational achievement, the National Association of Scholars, like Judge Friedman, concludes that "intellectual diversity bears no obvious or necessary relationship to racial diversity." But this claim is too strong. It's true that, in some classes, there's no clear correlation; hearing African Americans talk about their experiences with police is more important in a criminal procedure class than in a corporate tax class. Moreover, the University of Michigan, which gives preference to some racial and ethnic groups and not others, isn't interested in a complete diversity of viewpoints. It might be useful to hear from native Puerto Rican students in a classroom discussion about annexation cases, for example, but the University of Michigan doesn't appear to distinguish between island and mainland students of Puerto Rican descent. Nevertheless, the constitutional question isn't whether it's possible, with unlimited resources and a very different applicant pool, to achieve an even greater diversity of views than would emerge in an affirmative action system that uses broad racial categories as crude proxy for viewpoint diversity. The question is whether minority students with very different political and social backgrounds bring enough of the distinctive perspective that comes from being a minority in America to create real educational benefits for everyone.
Powell held that, as long as universities avoid explicit quotas and two- track admissions systems (in which minority candidates are compared only with each other), they may use race as a "plus factor" in admissions decisions. His opinion has been justly criticized for failing to define how much race consciousness is too much--at what point does a plus factor become the predominant factor? But, in the two decades since Bakke, it has become obvious that federal courts are not institutionally equipped to identify this point with precision.
In this sense, affirmative action cases teach the same lesson as the Supreme Court's voting rights cases: that race and viewpoint are intertwined in such complicated ways that courts can't possibly disentangle them. In a series of decisions beginning with Shaw v. Reno in 1993, the Court held that state legislatures could use race as one of several factors, but not as the predominant factor, in drawing legislative districts. Four conservative justices made it clear they would never allow districts to be drawn in order to maximize the voting strength of racial minorities. Four liberal justices made it clear they would allow these districts to be drawn without close judicial supervision. This meant that every voting district in the country had to await the reaction of the swing justice, O'Connor, whose criterion for distinguishing permissible districts from impermissible ones seemed to be: You know it when you see it.
But, on April 18, Justice O'Connor finally decided that she could live with the redrawn North Carolina district she had first expressed concern about in Shaw v. Reno nearly a decade ago. She provided the fifth vote for Justice Breyer's opinion, which concluded that politics, not race, had been the predominant factor in drawing the district. Breyer emphasized that because African Americans in the state register and vote Democratic more than 95 percent of the time, it was hard to distinguish a legislature's effort to create a majority-black district from a legislature's effort to create a safe Democratic one. And because white registered Democrats more frequently cross over to vote Republican than African American registered Democrats do, a legislature trying to secure a safe Democratic seat could plausibly include more heavily African American precincts for political, not racial, reasons. " Race in this case correlates closely with political behavior," Breyer concluded, reasoning that judges should therefore give legislatures leeway to balance racial and political considerations without assuming one was predominant over the other.
Republicans, of course, don't dispute the close correlation between racial identification and political affiliation when it suits their interests. The same week the Supreme Court handed down its decision in the North Carolina case, New Jersey Republicans announced they were challenging the decision by the state apportionment commission, supported by some Democratic black and Hispanic state representatives, to reduce the minority population in three Essex County districts and redistribute it among several surrounding ones. It's been obvious since the early '90s that the Democratic strategy of creating as many minority districts as possible is a self-defeating way for the party to isolate its most reliable voters. Now that Democrats--at least in New Jersey--have finally come to their senses and decided to follow their political rather than ideological interests, it's the Republicans who are insisting that all black voters think and vote alike, and therefore can't be represented by whites.
By taking the opposite position in affirmative action cases and by pretending there is no meaningful connection between race and viewpoint, Republican judges and politicians may think they are eliminating race from the admissions process. But this is naive; the political demand for racial diversity is too great to be resisted. What affirmative action's opponents are likely to do, ironically, is undermine meritocracy. The most alarming part of Judge Friedman's opinion is his suggestion that the University of Michigan should have experimented with "race-neutral" alternatives to affirmative action, including "decreasing the emphasis for all applicants on undergraduate GPA and LSAT scores, using a lottery system for all qualified applicants, or a system whereby a certain percentage of the top graduates from various colleges or universities are admitted." This abandonment of objective standards of academic measurement is precisely what has occurred in California, Florida, and Texas in the wake of initiatives and court decisions forbidding affirmative action. The political pressure to maintain the number of minorities in the states' flagship universities was so strong that, in February, the chancellor of the University of California called for the abolition of the SAT, and the Universities of Texas and Florida now guarantee admission to a certain percentage of all high school graduates in the state. These reforms may well turn the leading state universities into remedial academies. The quality of high schools varies so widely across the state of Texas, for instance, that the Austin campus, the flagship of the University of Texas system, is now admitting a lower caliber of applicants from all ethnic groups than it did under affirmative action, when exceptions were made primarily for African American and Hispanic students.
In arguing that affirmative action represents a lesser compromise of meritocratic standards than the alternatives that are almost certain to follow its elimination, I don't mean to deny that there are costs to broad generalizations about the connection between racial diversity and viewpoint diversity. One of the most committed conservative students in my recent constitutional law class was an African American woman who, like Clarence Thomas, bitterly resented the claim that African Americans think and vote alike. But her position, too, contributed to the diversity of the class. By presuming to identify the mystical point at which racial diversity does or doesn't lead to a richer classroom experience, Judge Friedman has interjected the federal courts into a complicated question of educational policy and social psychology best left to the University of Michigan. If the Supreme Court agrees with him, it will trigger an abandonment of objective admissions standards that may destroy America's great public universities. What an odd position for self-styled defenders of meritocracy to be applauding.
Jeffrey Rosen is legal affairs editor at The New Republic and president and CEO of the National Constitution Center.