Adam Glanzman/Bloomberg/Getty Images

Protesters at a demonstration against Harvard University’s admissions process, Copley Square, Boston, October 2018

No one working in or around a US university can think of assembling a class, a set of interviewees for a job, a special issue of a journal, or a scholarly panel without making sure it is suitably “diverse.” Not only admissions officers but students, faculty, and administrators share a commitment to diversity. Even some conservatives have appropriated the concept, calling for ideological diversity at universities to ensure that their own voices are heard.

Yet in the courts, diversity is being subjected to a withering attack, one that most informed observers expect to be fatal now that Justice Anthony Kennedy has been replaced by Justice Brett Kavanaugh. According to existing Supreme Court doctrine, diversity is the only rationale sufficiently strong to justify educational institutions’ use of race in admissions. That doctrine was most recently upheld in a 4–3 decision written by Kennedy in 2016. Kavanaugh is unlikely to sustain his predecessor’s opinion, and diversity will probably lose its favored constitutional position in the near future.

Meanwhile, a highly publicized lawsuit, supported by Donald Trump’s Department of Justice, alleges that Harvard University uses diversity as a cover for admissions discrimination against Asian-Americans. The core of the claim is that if admissions were based strictly on grades and test scores, Asian-Americans would make up more than 30 percent of the student body, rather than 22.2 percent, as they did in 2017. The plaintiffs allege that Harvard gives Asian-American applicants systematically lower scores on its “personal” measure—a measure on which African-American and Latino candidates on average score higher.

The suit also points out that Harvard first made diversity a goal in admissions in order to reduce the number of Jewish students accepted in the 1920s. Although this anti-Semitic origin has been known since it was reconstructed from Harvard president A. Lawrence Lowell’s archives by a doctoral researcher in the 1970s,1 most liberals have long considered it not a reason to impugn diversity but an unfortunate historical contingency that has since been overcome by developments that have made diversity desirable. The current lawsuit challenges this interpretation by suggesting that diversity is as restrictive for Asian-Americans today as it once was for Jews.

Diversity became a central value in higher education through a judicial fiat that gradually became a widely held belief. Its origins lie in a 1978 Supreme Court decision, Regents of the University of California v. Bakke (1978). The case was brought by an unsuccessful white medical school applicant who claimed that affirmative action violated equality under the Civil Rights Act and the equal protection clause of the Fourteenth Amendment. His challenge was part of the fascinating—and in many ways horrifying—revisionist backlash against the integrationist jurisprudence of the Warren Court. In a series of close decisions, conservative majorities of the Burger and Rehnquist Courts struck down various types of affirmative action as violating the equal protection rights of whites.

These decisions undermined the rationale for affirmative action, which was, its advocates generally maintained, to make up for slavery and segregation by ensuring that African-Americans (and sometimes Latinos) would be able to obtain professional opportunities that had historically been denied them. The conservative justices repeatedly cast doubt on this argument by saying that a sufficiently close connection between past harms and current remedies had not been demonstrated. Justice Antonin Scalia, the most aggressive of those conservatives, went further, insisting that any classification by race for any reason whatever violated a principle of constitutional colorblindness that he ascribed to the equal protection clause.

In Bakke, four justices thought that the race-based affirmative action admissions method used by UC Davis’s medical school violated equality under the civil rights laws. Four other justices would have upheld affirmative action as a remedy for past discrimination. Justice Thurgood Marshall stated their position simply. “During most of the past 200 years,” he wrote,

the Constitution as interpreted by this Court did not prohibit the most ingenious and pervasive forms of discrimination against the Negro. Now, when a State acts to remedy the effects of that legacy of discrimination, I cannot believe that this same Constitution stands as a barrier.

That left the balance of power with Lewis Powell Jr., the swing justice. Powell offered a compromise position: affirmative action in admissions would be unconstitutional if it assigned explicit weight to race. It would be permissible, however, to take account of race as part of a policy that considered “diversity” as “only one element in a range of factors a university properly may consider in attaining the goal of a heterogeneous student body.”

Powell famously got the idea of using diversity as the rationale for affirmative action from the amicus brief filed by Harvard University and flagged for him by his law clerk.2 The most famous line from the brief, included in Powell’s opinion, read: “A farm boy from Idaho can bring something to Harvard College that a Bostonian cannot offer. Similarly, a black student can usually bring something that a white person cannot offer.”


The Harvard brief presented a version of diversity stripped of its anti-Semitic origins. It surely never entered Powell’s mind—or presumably that of the lawyers who filed the brief for the university—that the reference to the “Idaho farm boy” echoed the objectives of Harvard’s earlier appeals to geographical diversity in order to admit Protestant westerners at the expense of urban Jews.3

Powell’s diversity rationale entered the legal mainstream only gradually. Initially, no other justice had joined the reasoning of his compromise opinion, since the conservatives didn’t think affirmative action should continue in any form, and the liberals didn’t like Powell’s rejection of remediation as a justification for it. But eventually, in 2003, Justice Sandra Day O’Connor wrote an opinion for the majority of the Court embracing Powell’s vision of diversity.

Powell had emphasized that courts should defer to universities’ stated educational mission. O’Connor preserved the appeal to diversity, but subtly shifted the argument away from the educational benefits of diversity and toward its broader social effects among the graduates of elite institutions. Invoking both businesses and the US military, she insisted that “in order to cultivate a set of leaders with legitimacy in the eyes of the citizenry, it is necessary that the path to leadership be visibly open to talented and qualified individuals of every race and ethnicity.”

In 2016, after O’Connor’s retirement, Kennedy wrote an opinion that endorsed diversity and repeatedly cited O’Connor’s opinion—even though he had not joined it in 2003. Kennedy was unable to offer his own substantive vision of why diversity has value, signaling rather that it possesses an inchoate, almost mystical quality. “Diversity takes many forms,” he wrote—without ever specifying what those might be. “Formalistic racial classifications may sometimes fail to capture diversity in all of its dimensions.”

Although Kennedy did not say just what diversity was, he nevertheless did accept that diversity could be captured by “holistic review” of candidates for admission. His idea seemed to be that admissions officers would weigh the unique qualities of every student, rather than assembling a class based on demographic goals. In reality, elite institutions do seek to evaluate applicants individually. Yet they simultaneously manage to produce demographically balanced classes according to established proportions of race, geography, and wealth, which do not necessarily correspond to the demographics of the larger population. The balance typically differs only a little from year to year and even less when one compares the classes admitted by all schools of similar standing in any given year.

Those consistent results would be highly improbable if the admissions officers were merely evaluating individual applicants on a “holistic” basis, without attention to overall numerical goals. The balanced classes are sometimes defended (although not by Harvard) as creating a “critical mass” of minority students. Presumably, the similarity across institutions satisfies the universities’ desire to include students of many backgrounds while simultaneously protecting them from the criticism that they got the balance wrong by appealing to shared institutional norms.

Because diversity has been the only legally permissible rationale for affirmative action since Bakke, universities have taken the idea seriously. Admissions offices and administrators heralded diversity as their goal. Over time, faculty and students began to use the same language of diversity. And during the period in which diversity became a liberal value, it gradually supplanted the idea that affirmative action was justified as a remedy to past wrongs. A few stalwarts of the liberal golden age of the Warren Court, like Owen Fiss at Yale Law School, have never stopped insisting that affirmative action was intended to be (and remains morally justifiable as) a remedy to structural discrimination.4 But by the time I was Fiss’s student, in the mid-1990s, this view had already come to seem like a relic of the Marshall-Brennan era. Today, most people who openly invoke remediation are advocates of reparations for African-Americans, intended to go much beyond affirmative action. Advocates of diversity may still quietly think of affirmative action in remedial terms, but that rationale has essentially disappeared from official institutional discourse.

Roughly speaking, the most common defense of diversity has two parts. The first focuses on the educational and social benefits of diversity. The second attempts to show the inherent value of a diverse environment, one that is in some sense representative of the diversity of the American, or perhaps global, population.

Consider this passage from a 2016 report by a committee headed by the dean of Harvard College, endorsed by the Harvard Faculty of Arts and Sciences:


We emphatically embrace and reaffirm the University’s long-held and oft-expressed view that student body diversity—including racial diversity—is essential to our pedagogical objectives and institutional mission. Our diverse student population enhances the education of our students of all races and backgrounds and prepares them to assume leadership roles in the increasingly pluralistic society into which they will graduate.

The passage invokes both Powell’s version of diversity, by highlighting its educational benefits (“enhances the education of our students”), and O’Connor’s version, by emphasizing the social benefits of diversity (“leadership roles in…pluralistic society”). Yet the report also goes well beyond the views of both justices, making diversity not just a feature of who is admitted and who graduates but a core value, “essential” to the university’s mission. The report includes sections on diversity in curriculum, research, student living, and extracurricular activities, all described as “contexts where students interact with ‘other’ [sic], with those having different realized and recognized identities.” The interactions “ensure that our students truly engage with other people’s experiences and points of view, [and] that they truly develop their powers of empathy.” This account strives to make diversity not only useful but an inherent good for an educational community, since different identities generate the different ideas that are necessary for a proper education.

In practice, the idea that a diverse environment provides concrete educational benefits is a tricky one, raising as it does the question of who, exactly, is supposed to benefit. Diversity would seem intended to benefit members of the majority, who gain the opportunity to learn from people with different backgrounds and experiences. Members of minorities, whether racial, ethnic, or based in gender and sexuality, are mostly accustomed to interacting with and understanding the culture of the majority. To say that a diverse environment exposes the majority to new perspectives tends to make minorities into a means to enhance the education of the majority, rather than treating the benefits to minorities as the main goal. (Though of course in principle members of one minority might benefit from exposure to members of other minority groups whom they otherwise would not encounter.)

Justice Powell
Justice Powell; drawing by David Levine

Another familiar problem with the diversity rationale is that it implicitly relies on the idea that each person who is part of a diverse environment represents some particular set of demographic characteristics—some identity with a specific worldview attached. For students of color in particular, this can create pressure to speak and identify in classrooms and on campuses as representative minorities, even when they might prefer not to do so. This implied representativeness is at odds with the belief that students are unique and should be able to express their views independent of what anyone expects. It undercuts the classic liberal arts notion that an educational experience is supposed to offer students access to a set of ideas and values that they did not possess when they began their studies. In any case, we should not be asking minority students to speak about their “experiences” in educational environments for the purpose of enlightening mostly upper-middle-class white students.

The theory of diversity that supports the value of being exposed to different ideas derived from different life experiences also cannot be limited to minorities who have faced discrimination. It should logically include anyone whose beliefs are connected to identity and differ strongly from those of the majority. On today’s elite university campuses, that is coming to include conservatives and Christian evangelicals. Similar “diversity” could extend to white nationalists, anti-evolutionists, flat-earthers, climate-change deniers, and indeed almost anyone whose views would be unusual on college campuses. Exposing students to these beliefs would unquestionably broaden their knowledge, encourage lively and vigorous debate, and maybe teach empathy. Yet most advocates of diversity feel skeptical about appealing to diversity in order to admit more people who hold such views. Not only do they object to the views, they want diversity to be limited to its original function of justifying affirmative action designed as a remedy for past discrimination.

One possible response to these troubling aspects of diversity is to argue for the inherent value of having classrooms that “look like America.” Rather than assume that diverse membership guarantees diverse points of view, this approach suggests that an educational institution should resemble the community it serves. But what real-world community should be the model for universities? Is it the United States? Some region thereof? Or something else? And what of overrepresented minorities on elite campuses, such as Asian-Americans, who make up 5.6 percent of the US population?

Finally, the argument for the inherent value of a university that mimics the diversity of the wider population is to some degree at odds with the idea that academic endeavor is fundamentally about the pursuit of truth. We may want elite universities to resemble a demographic average, but we want excellent students and faculty, not average ones. A university may justifiably select its members for their capacities to contribute to the search for knowledge. And while disagreement of all kinds has a vital place in a political community, the ends of a scholarly community are not always well served by disagreement fueled by racism or sexism, or discord over fundamental issues like the importance of critical thinking and scientific method.

These problems with our current thinking on diversity should remind us that we embraced the idea in the first place mostly because the courts allowed it as a justification for race-based affirmative action. Now that the courts are poised to take away that rationale, we are going to need new approaches to ensure continued representation of African-Americans and Latinos without violating the law or our conception of what goals universities should pursue. It would be too ambitious to lay out a comprehensive solution here. But a good place to start is with the suit against Harvard, which suggests some of the complexity of the issue.

The premise of the suit is that grades and standardized test scores are the necessary and sufficient determinants of admission. But this argument is too simple, both with respect to measuring the merit of individual applicants and to fulfilling the broader social functions of the university. A good and fair admissions process should consider not only applicants’ grades and test scores but also the obstacles they have confronted. These include the experience of structural and economic inequality, which in the United States is inevitably inflected by race.

In this sense, historical racial discrimination should be a factor in admissions—not as backward-looking remediation, but because of its ongoing effects on current applicants. There is nothing unlawful about considering the effects of racism on applicants, provided that such effects are considered on a case-by-case basis using information that could be gleaned from application essays and sociological evaluation of applicants’ life circumstances. This would mean, for example, that applicants’ race would not in itself affect admissions, but would do so only indirectly, insofar as intergenerational discrimination has shaped their experiences and opportunities. Standardized tests themselves should also probably be de- emphasized, a trend that is already well underway as numerous universities have made the SAT and ACT optional. The results of those tests remain closely linked to socioeconomic status, and they do not sufficiently track aptitude, which should be distributed equally through the population.

Finally, although the plaintiffs in the suit assert that affirmative action for African-Americans and Latinos is the reason that more Asian-Americans are not admitted, it is very unlikely to be the case. Given current balancing goals in admissions, Asian-Americans are being compared with and balanced against white applicants, not African-American or Latino ones. If there exists unconscious, stereotype-based admissions discrimination against Asian-American applicants, as the suit alleges, that problem must be addressed by examining the unspoken bias and ensuring it does not persist. That might conceivably lead to fewer white admissions and more Asian-American ones. It should not reduce the numbers of African-Americans and Latinos admitted.

An elite university does indeed serve the social function of educating a broad-based set of leaders. As Justice O’Connor suggested, this provides a legitimate reason to seek strong and qualified candidates from all backgrounds. But diversity should not be an end in itself. Rather, diversity should be the indirect result of the goal of inclusion based on forms of merit found in every part of the population and not purely reducible to grades and scores. The university need not look precisely like America. But its classes should reflect a conscientious effort to find and admit the students who are best placed to pursue truth and become future leaders, weighing their abilities and potential in the light of the challenges they have faced. That should logically result in a wide range of students. Diversity could then live on—as an outcome, not as an objective.