To the Editors:

I have finally had a chance to read Ronald Dworkin’s essay “The Court and the University” [NYR, May 15]. In it the author cites an essay by me that had appeared in The New York Times as an Op-Ed piece. Unfortunately he makes no use of my data even when it is quite relevant to key segments of his argument. For example, Dworkin lays great stress on amicus curie briefs by some very large corporations, retired military officers, and major universities. These show, he avers, that the establishment has become convinced that higher education is greatly improved by diversity.

Unfortunately the matter is not quite so simple. As my colleagues and I point out, only 2 percent of high-level administrators, including college presidents, in the national sample we surveyed, believe that special admissions policies for minority students improve college academic standards, as against 66 percent who believe such policies have no impact and 29 percent who believe that they actually lower academic standards. And the larger the proportion of the number of black students in a university, the more likely top-level administrators are to have negative opinions regarding the students’ ability to work at a college level. Indeed their view of the quality of education at their university drops. College administrators seem to have a public voice that differs from their real convictions.

What is more, studies conducted by the University of Michigan itself contradict Dworkin’s praise of its diversity initiatives. The studies, which have not exactly been trumpeted by Michigan, were obtained by Robert Lerner and Althea Nagai. Key findings include the discovery that the longer students have been at Michigan the more likely they are to perceive “a great deal” of racial tension in residence halls, and the less likely they are to have friends from another racial group. Further, as they move from entrance year to senior year, both white and Asian students are increasingly unlikely to believe that “in the long run, greatly increased enrollment of students of color will enhance the excellence of universities.” Hispanic and, especially, African-American students perceive the situation otherwise. Thus on this issue the students become increasingly polarized over time.

Just one final point: reading Dworkin one would think that diversity is primarily about black students. In fact the primary beneficiaries of diversity in the coming years will be Hispanic students, including recently arrived students of affluent backgrounds.

Unless we deal honestly with diversity issues our efforts will fail to achieve their promised goals, whatever the Supreme Court decides.

Stanley Rothman
Mary Huggins Gamble Professor Emeritus of Government
Smith College
Director, Center for the Study of Social and Political Change
Northampton, Massachusetts

Ronald Dworkin replies:

The Supreme Court’s recent decisions were widely and correctly seen as a huge victory for affirmative action: they made it plain that universities could continue to follow the guidelines laid down by Justice Powell in the Court’s Bakke decision twenty-five years ago. It will be expensive for those institutions that use point systems to substitute less mechanical programs, but not prohibitively so. The Court therefore saved educational policies that almost all our best universities have adopted, and that are crucially important for the nation as a whole.

The decisions provoked a great deal of press and academic comment, and some misunderstanding. Professor Stanley Rothman’s response to my article, written before the decisions, suggests that the Court ignored important evidence that racial diversity does not achieve its educational goals. Some other commentators, echoing Justice Clarence Thomas’s emotional dissenting opinion in the law school case, say that the Court’s decision means that affirmative action automatically becomes unconstitutional in twenty-five years. And it is widely believed that the Court limited the justification of affirmative action to educational benefits narrowly construed, which would restrict the impact of its decisions on affirmative action programs outside higher education. Each of these opinions is mistaken.

In his New York Times Op-Ed piece that I cited in my article, Rothman relied on a study designed by himself and Professors Seymour Martin Lipset and Neil Nevitte, and he cites that study again, in his letter, to contradict my claim that America’s leading universities believe that affirmative action programs have educational benefits.1 But though Justice Thomas cited the study in his dissenting opinion, it has, on closer examination, no bearing at all on the issues in the cases.

Useful empirical research on the impact of affirmative action on education, like the survey by William Bowen and Derek Bok that Justice O’Connor mentioned in her opinion,2 has concentrated on the elite universities in which affirmative action is practiced. Rothman et al. collected data from a much broader sample of 140 American colleges and universities, selected at random from “the entire universe of qualified institutions.” Their sample therefore included schools ranging from the least to the most prestigious, and included many schools where affirmative action is not practiced at all, in some cases because there is already a substantial minority student group. Ninety-seven percent of those who responded to the study’s questionnaire “attended schools with a black student percentage that ranges from 0 to 43 percent …3 percent of the respondents attend schools which are more than 77 percent black.” The Rothman study asked a variety of questions, but none of these was designed to test the impact of affirmative action on any institution.


The study found that a higher percentage of all students, faculty, and administrators “expressed dissatisfaction with the quality of education and work ethic of their peers” in schools with more black students than in those with less. In his Op-Ed piece Rothman said that this refuted the claim that diversity has educational benefits. “If diversity works as advertised,” he said, “then those at institutions with higher proportions of black enrollment should rate their educational and racial milieus more favorably than do their peers at institutions with lower proportions.” But that is not true. It is unsurprising that students in schools in which blacks make up 43 percent of the class think the quality of their education worse than students do in other institutions where blacks are a substantially lower percentage. Unfortunately, it probably is worse.

Rothman said that even among comparably “selective” schools more students are dissatisfied with their education as the percentage of black students increases. But he measures selectivity by the proportion of its applicants that an institution accepts, and since students self-select in their applications, a school with a relatively low percentage of acceptances to applications but a high percentage of black students may well be educationally inferior to a school with a higher rate of acceptances but a lower percentage of blacks. In any case, the study indicates nothing about whether increasing the proportion of minority students in elite schools like Michigan from a trivial to a nontrivial level, through a carefully constructed program aimed at recruiting the most qualified minority applicants, improves or lowers the quality of education offered. Only a study that charts the effect of such programs at such a school could do that.

In his letter Rothman cites a different conclusion of the study: that only 2 percent of “high-level” academic administrators in the study’s sample believe that affirmative action improves “college academic standards” while 29 percent of them think they lower those standards. (In their published papers the authors say that “all” administrators, not just “high-level” ones, were included in the sample.) Many of the administrators, however, are from schools that have no experience with affirmative action. In any case, the claim Rothman says he contests—the claim made by the studies O’Connor cited—is not that diversity raises academic standards as traditionally defined but that it benefits students and makes their education more valuable in other ways. The most interesting of Rothman et al.’s findings, in fact, is that 68 percent of the administrators, 60 percent of the faculty, and 63 percent of the students who responded thought that affirmative action does not lower academic standards.

Rothman says that “studies conducted by the University of Michigan itself contradict Dworkin’s praise of its diversity initiatives.” These studies were the first findings, presented in 1994, in a continuing Michigan Student Study. They showed that a lower percentage of students thought there was “quite a bit” to “a great deal” of racial interaction on the Michigan campus at the end of their second year than did at the end of their first. But as Michigan’s expert witness, Professor Patricia Gurin, reported, the full Student Study showed an actual increase over each of four years in the percentage of students who reported their own close friendships with students of another race. Students’ perception of the general racial situation, that is, did not match the true situation revealed when they reported on their own experience. Drs. Robert Lerner and Althea Nagai, whom Rothman cites, concluded from the data that “friendships among students of different races continue to decline, all the way to the fourth year.” They made the simple mistake of confusing perceptions with fact.3

O’Connor did say that “we expect that 25 years from now, the use of racial preferences will no longer be necessary to further the interest approved today.” She made plain, however, that this expectation was not an automatic “sunset” deadline, but rather a prediction that by that time minority applicants would be sufficiently qualified on traditional criteria to produce a diverse student body without special programs, or that experimentation in various states would have discovered admissions techniques that produce adequate diversity without relying so explicitly on race.

Justices Breyer and Ginsburg, whose votes were necessary to make O’Connor’s opinion the opinion of the Court, emphasized the difference in Ginsburg’s concurring opinion: “From today’s vantage point,” she wrote, “one may hope, but not firmly forecast, that over the next generation’s span, progress toward nondiscrimination and genuinely equal opportunity will make it safe to sunset affirmative action.” Of course, if racially sensitive admissions policies are still in use in twenty-five years, O’Connor’s statement will place a heavier burden of proof on those who then defend them. But it will also make a successful challenge more difficult before then, even if the Court’s composition shifts further to the right.


O’Connor emphasized that the only university interest that justifies affirmative action is its interest in “attaining a diverse student body,” which has suggested to many commentators that only immediate classroom benefits count. In fact her opinion achieved a subtle but important expansion of the benefits of diversity, because she mentioned not only the improved understanding that racial diversity brings to students, but the further benefits of providing more qualified minority leaders for all sections of the nation’s economy and government. She referred to the briefs of leading American corporations and of the military, which emphasized the importance of more racial diversity among executives and officers, and concluded that

high-ranking retired officers and civilian military leaders assert that a highly qualified, racially diverse officer corps is essential to national security. Moreover, because universities, and in particular, law schools, represent the training ground for a large number of the Nation’s leaders,…the path to leadership must be visibly open to talented and qualified individuals of every race and ethnicity. Thus, the Law School has a compelling interest in attaining a diverse student body.

In that way, O’Connor enforced a critical distinction between a policy of seeking racial balance either for its own sake or to compensate for past social injustice, which she rejected, and a policy of improving the racial mix of different dimensions of American leadership for everyone’s benefit, which she accepted.

This Issue

August 14, 2003