In 1994, shortly before his retirement, Justice Harry Blackmun announced a shift in his thinking about the death penalty. After years of struggling to make the administration of the death penalty sufficiently fair to comply with the 8th Amendment’s ban on “cruel and unusual punishments,” Blackmun concluded that the Supreme Court’s “death penalty experiment has failed.” He therefore could no longer “tinker with the machinery of death.”

Nearly 30 years later, some Supreme Court Justices may be coming to a similar conclusion about the nation’s diversity experiment. This was the tone, at least, underlying many of the questions in the Court’s oral arguments for its two pending affirmative action cases—Students for Fair Admissions, Inc. v. President and Fellows of Harvard College and Students for Fair Admissions, Inc. v. University of North Carolina. In the former case, Students for Fair Admissions (SFFA) argued that Harvard’s race-conscious admissions process discriminates against Asian Americans and thus violates Title VI of the 1964 Civil Rights Act (which applies to all institutions that receive federal funding). In the latter, SFFA argued that UNC’s affirmative action program likewise violates the 14th Amendment’s Equal Protection Clause (which applies to all state actors).

At the center of both oral arguments was the meaning and constitutional validity of Grutter v. Bollinger (2003), the blockbuster case upholding the University of Michigan Law School’s affirmative action program. Both oral arguments focused in particular on one line from Justice Sandra Day O’Connor’s majority opinion in Grutter: “We expect that 25 years from now, the use of racial preferences will no longer be necessary to further the interest [in student body diversity] approved today.” Given that we are now only five years away from that 25-year mark announced in Grutter, the Justices repeatedly inquired why affirmative action programs are not showing any signs of diminishing. For example, at one point in the Harvard oral argument, Justice Amy Coney Barrett wondered whether “Grutter was grossly optimistic” about the 25-year expiration date and probed whether there is indeed an “endpoint” to affirmative action under the Grutter reasoning. Likewise, in the UNC case she asked the North Carolina solicitor general a series of questions in search of this elusive endpoint: “[W]hat are you [going to be] saying when you’re up here in 2040? Are you still defending [affirmative action] like this is just indefinite, it’s going to keep going on?” Chief Justice John Roberts made a similar observation, noting that there can be no endpoint under the Grutter framework:

I don’t see how…the program will ever end. Your position is that race matters because it’s necessary for diversity, which is necessary for the sort of education you want. It’s not going to stop mattering at some particular point. You’re always going to have to look at race because you say race matters to give us the necessary diversity.

The fact that so many of the Justices were focused on finding an endpoint to affirmative action has led many observers to predict that the Court will not only rule against Harvard and UNC but will go further by invalidating all affirmative action programs. For anyone who has seriously studied affirmative action law, this prediction has a “here we go again” quality to it. For over 40 years, scholars and pundits on the left and right alike have been predicting the death of affirmative action. My own recently published article in the Ohio Northern University Law Review, “Beyond the Law: A Four-Step Explanation of Why Affirmative Action Is Here to Stay,” catalogs a sample of publications dating back to 1978 predicting the end of affirmative action—a sample consisting of over 50 citations and several pages of law journal space.

But aside from the error rate of past predictions, there is another reason to doubt that the Harvard and UNC cases will put an end to affirmative action: although many of the Justices appeared frustrated with the task of tinkering with the machinery of diversity, not a single one seemed prepared to withdraw from the enterprise altogether and declare that our nation’s diversity experiment has failed.

This was most apparent in two features absent from the oral arguments: (1) the Justices failed to trace the origin and trajectory of our diversity discourse, and (2) the Justices failed to engage how affirmative action, in tandem with the civil rights revolution, has supplanted our constitutional order. Until the Court is willing to address these two issues, affirmative action is likely here to stay.

Diversity as Our Core Value

“Diversity is our nation’s greatest source of strength.” With those magical words, North Carolina Solicitor General Ryan Park began his argument defending UNC’s affirmative action program. As is usually the case when someone invokes the “diversity is our greatest strength” mantra, Park did not mean diversity as such; rather, he was referring to a particular type of diversity—racial diversity. Additionally, he asserted this relationship between racial diversity and national strength without providing evidence to buttress the claim. This is common when someone makes this assertion because there is no evidence as to why racial diversity, in itself, would strengthen a nation.

Park also added a significant “but” following his assertion. This is another feature of diversity talk: whenever someone says “diversity is our greatest strength,” it is almost always to explain away a serious problem. Park argued that diversity “poses unique challenges to the American experiment,” in that different groups “have to learn to live together and unite in common purpose.” In other words, racial diversity makes us so divided that we need the government to be involved in managing our most intimate affairs and teaching us how “to live together and unite in common purpose.” No one asked why something that requires constant governmental meddling and intervention is somehow still a strength and not a weakness.

To non-lawyers, this must seem like a strange, even bizarre, way to begin an oral argument—by using diversity as code for something else, by asserting without any evidence a relationship between diversity and national strength, and by drawing attention to a feature of diversity that ostensibly undermines that relationship. But to lawyers, this was perfectly sensible, because the Court’s first major affirmative action case, Regents of the University of California v. Bakke (1978), made diversity the cornerstone of the Supreme Court’s affirmative action jurisprudence and, in turn, of the nation’s very identity.

In his Bakke opinion, Justice Lewis F. Powell, Jr., proclaimed that public university affirmative action programs, like all racial classifications, must satisfy “strict scrutiny” under the 14th Amendment’s Equal Protection Clause. This means that a public university may practice affirmative action only if the program is narrowly tailored to achieve a compelling government interest. In the context of higher education, Powell found such an interest in the pursuit of a racially diverse student body. Bakke thus made diversity the only government interest that is sufficiently powerful to overcome the Constitution’s ban on racial classifications. Indeed, the Court has held that matters like national security, the welfare of children, and prison safety are not sufficiently powerful to justify race-conscious classifications. But diversity is. That raises a fundamental question critical to understanding the trajectory and future of affirmative action: how did diversity come to define our constitutional order?

Origins of Diversity

The story begins with Appendix A of Powell’s Bakke opinion, which summarized Harvard’s admissions program for the purpose of illustrating the educational value of racial diversity. Over the years, scholars and pundits have treated Appendix A as a formal Harvard admissions office document, because Appendix A has an official-sounding title (“Harvard College Admissions Program”), cites internal university documents in describing how the Harvard admissions committee operates, and matches verbatim an appendix that Harvard College included in its Bakke amicus brief.

Scholars have generally neglected, however, that before that language appeared in the appendix of Harvard’s amicus brief in Bakke, it appeared on pages 14-17 of the body of Archibald Cox’s amicus brief for Harvard in DeFunis v. Odegaard (1974), an affirmative action case that the Supreme Court dismissed on jurisdictional grounds four years before deciding Bakke. This is significant because the distinction between the body of a brief and an appendix to a brief is a distinction between legal advocacy and factual description. Appendix A thus represents the transformation of Cox’s advocacy for his client in DeFunis into the Supreme Court’s factual record in Bakke of how Harvard admissions actually operated.

This migration from advocacy to fact is troubling enough. Even more troubling is that the excerpted portion of the Cox brief was not even a persuasive piece of advocacy. Indeed, in that excerpted portion, Cox relied on the similarity between regional diversity and racial diversity in creating an optimal educational environment. But this analogy did not fit at the time with how Harvard admissions operated. Here is the relevant language from Cox’s brief (reproduced in Bakke’s Appendix A):

When the Committee on Admissions reviews the large middle group of applicants who are “admissible” and deemed capable of doing good work in their courses, the race of an applicant may tip the balance in his favor just as geographic origin or a life spent on a farm may tip the balance in other candidates’ cases. 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.

This analogy between the Idaho farm boy and a black applicant apparently had a significant influence on Powell. Berkeley Law professor David Oppenheimer’s superb article “Archibald Cox and the Diversity Justification for Affirmative Action” (Virginia Journal of Social Policy & the Law, 2018) explores Powell’s archives to understand how the Justice ended up taking a portion of Cox’s brief in DeFunis and turning it into Appendix A in Bakke. Oppenheimer discovered that one of Powell’s clerks had written a memo that, in Oppenheimer’s words, “repeatedly cite[d] as authority the ‘Brief for Harvard College in DeFunis,’ and refer[red] to the ‘Idaho farm boy’ analogy.” In that memo, Powell wrote in the margin: “This is the position that appeals to me. Use DeFunis.” According to Oppenheimer, Powell even intended in the Bakke oral argument to ask Cox (who argued on behalf of U.C. Davis) “how Harvard went about choosing an Idaho farm boy over a Boston first-family son, and whether there was a guarantee of a certain number of seats for farm boys,” but Powell “didn’t get to the question before time ran out.”

Had Powell asked this question, the entire trajectory of affirmative action law might have been different. It might have revealed the extent to which Cox’s analogy in DeFunis was a lawyer’s argument to justify affirmative action, not a factual description of how the Harvard affirmative action program operated. Indeed, Cox used the analogy between the educational value of “black students” and “Idaho farm boys” for the specific purpose of concealing Harvard’s black quota (which, by 1978, had been in place for a full decade). The analogy made it seem that Harvard’s racial preferences were merely part of a holistic, individualized interest in academic diversity. But that was not the case. In fact, there is no evidence that Harvard had at this point any sort of formal admissions preference whatsoever for rural applicants. But it did have a firm black quota in place, as noted in Alan Dershowitz and Laura Hanft’s 1979 Cardozo Law Review article, “Affirmative Action and the Harvard College Diversity-Discretion Model: Paradigm or Pretext.” And to produce that quota, Harvard had to institute a roughly 200-point preference on the SAT for black applicants, as Jerome Karabel details in his book, The Chosen: The Hidden History of Admission and Exclusion at Harvard, Yale, and Princeton (2005). None of this, however, was evident from Cox’s brief.

Moreover, the excerpted portion of Cox’s DeFunis brief (which became Appendix A in Bakke) cites only two authorities in explaining why Harvard valued racial diversity and how its affirmative action program operated. It seems that Powell simply reproduced the excerpted portion into Appendix A and did not bother to do his own research by reading these two documents (which were reports submitted in 1960 and 1968 by the outgoing admissions deans). Had Powell read them, he would have discovered that they had nothing to do with race and its relationship to academic diversity. In short, the entire edifice of the Supreme Court’s affirmative action jurisprudence is based on the educational value of racial diversity, which was based on a piece of advocacy that cited only two documents for this value—and, as it turns out, those two documents had nothing to do with the educational value of racial diversity.

That brings us back to the recent Harvard and UNC oral arguments, where the Justices missed an opportunity to expose the rot at the core of our diversity regime. The closest they came was when Justice Samuel Alito noted his own research on the source for Appendix A:

Harvard submitted a brief in Bakke, along with a number of other colleges. I went back and I looked at it and noticed that the brief talked about Harvard’s program going back 30 years, but it didn’t say anything about President Lowell or what Harvard had done back in the 1920s. So my question is, did Harvard sell Justice Powell a bill of goods? Do you think Justice Powell would have championed, would have held up the Harvard program as a model, as an exemplar for the whole country if he knew about the origins of the holistic program?

According to Alito, Appendix A represented a fraud on the Court because Harvard failed to disclose the relationship between Harvard’s affirmative action program in the 1970s and its Jewish quota from two generations earlier. This effort to taint Harvard’s affirmative action in the ’70s with the same nefarious motives that animated its Jewish quotas in the early 20th century is a popular rhetorical ploy among conservative critics of affirmative action. But it is an ineffective tactic, given the tenuous connection, at best, between Harvard’s discrimination against a disproportionately successful religious group in the ’20s and discrimination in favor of a disproportionately unsuccessful racial group in the ’70s.

What actually made Appendix A a “bill of goods” is that it treated a lawyer’s advocacy for affirmative action as a factual description of how Harvard’s admissions program operated. Appendix A thereby made it appear that Harvard’s affirmative action program consisted of an individualized assessment of each applicant on the holistic basis of how the candidate contributes to the college’s pursuit of a regionally, socio-economically, and culturally diverse student body. But that is not how Harvard’s program operated. Rather, it was designed simply to increase the number of black students through a race-based treatment of academic qualifications.

Sophisticated Data

Getting the Bakke diversity story right is important because it gets to the core of how diversity discourse operates as a cover for raw racial politics. Indeed, diversity has become a code for making institutions less white, so that the less white an institution becomes, the more diverse it becomes.

An example can be found in the diversity report cards used by ESPN. For the racial diversity of its players, the National Basketball Association gets a higher diversity grade than Major League Baseball, despite the fact that the NBA is one of the most racially homogeneous sports leagues, whereas MLB’s racial breakdown closely matches the nation’s percentages. The only way to make sense of this ranking is that the NBA is considered more diverse than MLB because it is less white (the NBA is around 18% white, whereas MLB is around 62% white). To be more diverse is to be less white.

This is exactly how the lawyers and Justices used the term in the Harvard and UNC oral arguments. Cameron Norris, representing SFFA—the group challenging Harvard’s affirmative action program—emphasized how his proposed Harvard admissions formula “would boost underrepresented minority representation” and “would lower the number of white students on campus” by providing admissions preferences on the basis of socio-economic factors.

The Justices treated this proposal as race-neutral, even though the proposed formula was designed explicitly to reduce the number of white students. The Justices’ principal objection to Norris’s proposal was that, under the gerrymandered formula, “[b]lacks wouldn’t increase,” as Justice Sonia Sotomayor wryly observed. Norris insisted, however, that even if it were true that “blacks wouldn’t increase,” his approach would still be beneficial, because the “number of white students would decrease” and “I think you’d see lots of benefits in that.”

Norris assured the Justices that Harvard could manipulate this proposed socio-economic formula to ensure that blacks would be more than 10% of the student body. The only reason he couldn’t gerrymander the admissions formula to guarantee black representation above “an absolute floor” of 10% is that his experts did not have access to the “sophisticated data that Harvard has.” The implication was that if the experts had access to Harvard’s data, they could further manipulate the admissions formula to reach what the Justices and lawyers seemed to agree is the desired outcome of university admissions—to raise black representation and reduce white representation.

It is worth reminding the reader that this argument—on how to use “sophisticated data” for the express purpose of manipulating university admissions criteria to guarantee fewer white students and more non-white students—was the argument challenging Harvard’s affirmative action program. There is no hope for eradicating affirmative action when this is what it means to be race-neutral.

It is also important to keep in mind the actual Harvard admissions data at issue in this case. Table 5.3 of the expert report submitted by Duke University economist Peter Arcidiacono showed that, if Harvard admitted students according to a purely academic index (as we would expect an academic institution to do without racial diversity pressures), Harvard would be only 0.76% black (assuming that a selective institution like Harvard would not need to go beyond the top 10% of applicants). But with Harvard’s racial preferences, more than 15% of the admitted class was black.

In other words, the challengers to Harvard’s affirmative action program were willing to deviate so far from merit that, under their preferred admissions formula, the college could produce more than ten times the number of black admissions warranted under strictly academic standards. The real fight, then, between the challengers and defenders of affirmative action was whether Harvard would have a 10% or 15% black quota. With challenges like this, affirmative action isn’t in much danger.

Another notable feature of Table 5.3 is that, of the four major racial groups in the United States, whites already are the most underrepresented group at Harvard, constituting under 40% of Harvard admissions and roughly 40% of the enrolled student body. This is the case, incidentally, at all of the nation’s leading institutions. Despite the fact that whites constitute around 60% of the nation’s population, the undergraduate programs at Yale, Princeton, and Stanford are, respectively, 38%, 41%, and 26% white. Nevertheless, in an oral argument preoccupied with diversity and with helping under-represented groups, no one mentioned that whites, the one group that everyone agreed should be diminished in representation, are already the most under-represented group at elite institutions.

Moreover, though the Harvard oral argument focused on how Harvard is preferring whites over Asians, the Harvard data suggest that this preference is a negligible feature of how affirmative action at the college operates. If Harvard abandoned affirmative action altogether and admitted only the top academic decile of applicants, it would have almost no effect on the white population (white admissions would be reduced by about one percentage point). Affirmative action is now, at Harvard at least, a battle between two over-represented groups: Asians (who are over-represented because of their academic excellence) and blacks (who are over-represented because of their privileged status under affirmative action). None of this was evident, however, from the Harvard oral argument, which made it seem that Asians were vastly under-represented because of white students stealing their spots.

Just like the Harvard argument, the UNC argument focused on how to manipulate the university’s admissions formula to generate the preferred racial composition. In one important way, the UNC argument revealed, even more than the Harvard case, the rot at the core of the diversity regime. This arose when U.S. Solicitor General Elizabeth Prelogar explicitly rejected any obligation of public institutions to consider granting admissions preferences for people from poorer backgrounds because, in the experience of the service academies, class- or income-based affirmative action “would actually increase the number of white men.” No one probed whether the government acts race-neutrally if it rejects an admissions criterion explicitly, and apparently solely, because it would increase the number of white students.

The solicitor general’s position captures the complementarity between the Harvard and UNC cases: whereas the Harvard oral argument focused on how elite private institutions could use socio-economic factors to become less white, the UNC oral argument focused on why less elite public institutions should not use socio-economic factors because that would make them more white. The two premises underlying the arguments were that the more racial diversity the nation has, the stronger it will become, and the less white the nation becomes, the more racial diversity it will have. These two premises combine to form a syllogism with dangerous implications for how to strengthen a polity.

Why Colleges Now Exist

Affirmative action is unique in Supreme Court case law as the only permissible government discrimination based on race. But affirmative action is also unique in American politics. It is the only public program in all of American history (to my knowledge) that has expanded in breadth and strengthened in force in the face of growing resistance from the American people, state legislatures, and federal courts.

When polling on affirmative action began in the late 1970s, a little over 50% of Americans opposed it. That percentage has increased considerably over the years, so that now about 75% of Americans oppose affirmative action. Most government programs gain acceptance over time, but affirmative action has decreased in popularity over nearly 50 years of polling. Reflecting this growing opposition, nine states over the past 30 years have banned affirmative action. Likewise, in nearly 50 years of affirmative action litigation, the Supreme Court has heard over a dozen cases on the subject, and the vast majority of these decisions have invalidated the particular form of affirmative action before the Court as unconstitutional.

Despite this resistance, affirmative action has managed not only to endure but to expand. And it has grown not simply to include just about all colleges and universities, but to define the very enterprise of higher education itself. The pursuit of diversity, as opposed to knowledge or truth, now constitutes the essence of what it means for a college or university to exist. Higher education has shifted from the trivium of grammar, logic, and rhetoric to the trivium of diversity, equity, and inclusion.

In the Harvard and UNC arguments, the Justices seemed oblivious to these phenomena. Indeed, the Justices acted as though affirmative action began with the Bakke decision and could be eliminated through a single Supreme Court decision, a mere turning of the judicial switch. The Harvard and UNC arguments thus obscured how affirmative action has evolved in concert with the civil rights revolution to become a defining feature of our constitutional order. Just as the Civil Rights Act of 1964 is “the law that ate the Constitution,” affirmative action is the civil rights program that ate our educational institutions and, in turn, our national identity.

To understand how this happened, we must understand the origins of affirmative action, which requires a definition of what it is. Affirmative action programs prefer historically disadvantaged groups in the provision of employment, educational, or other professional benefits, and provide these benefits for the purpose of either redressing a past injustice or creating future economic or social equality. These two features explain why we do not treat all employment and educational preferences as affirmative action. Indeed, we may like or dislike legacy, in-state, and veteran preferences, but whatever the case, we do not treat these as affirmative action programs.

When did programs consisting of these two features first appear in American law? The term “affirmative action” first appeared in federal law as part of the Wagner Act (1935), a law dealing with trade unions and labor rights, but its first usage in a racial context appeared in President John F. Kennedy’s Executive Order 10925. The policy, however, began in the 1940s, more than a decade before Kennedy’s executive order gave the practice a name. As John Skrentny explains in his book, The Ironies of Affirmative Action: Politics, Culture, and Justice in America (1996), several anti-discrimination measures in the 1940s and ’50s were framed in race-neutral terms, but the administration of these measures relied on affirmative action methods, creating race-based hiring goals even in areas where there was no evidence of past employment discrimination. Likewise, while we often think of affirmative action in higher education as beginning around the time of the Bakke case, it is more accurate to trace affirmative action to 30 years before Bakke, in 1948, when Harvard became the first college to make black recruitment an administrative priority under Dean John U. Monro. According to David Oppenheimer, Monro sought “10 Black recruits per freshman class,” and “[t]his should be rightfully regarded as the beginning of race-conscious affirmative action at Harvard.” In the 1950s, these efforts expanded under the Taconic Foundation’s “Gamble Fund,” which, as reported in the January 7, 1966, Harvard Crimson, admitted students to Harvard with “rock-bottom College Board scores” and provided special training at Andover to prepare them for Harvard.

The Supreme Court’s decision in Brown v. Board of Education (1954) was the turning point in spreading affirmative action throughout higher education. In their article “The Origins of Race-conscious Affirmative Action in Undergraduate Admissions: A Comparative Analysis of Institutional Change in Higher Education” (Sociology of Education, 2014), sociologists Lisa Stulberg and Anthony Chen examined various primary sources (such as school periodicals, press releases, administrative reports, internal memoranda, and private correspondence) to determine what prompted undergraduate institutions to adopt affirmative action in the 1950s and ’60s. They found that between 1954 (the year Brown was decided) and 1964 (the year the Civil Rights Act was passed), ten private and public institutions formally adopted affirmative action programs, and many of these schools explicitly linked the creation of their programs to Brown and the civil rights movement.

The civil rights movement quickly created a new competition among elite colleges—a competition over who could attract more black students. For example, as early as 1960 Swarthmore’s president acknowledged that, to increase black enrollment, it was the college’s practice to “lean[] over backwards” and “make concessions in matters such as Board score performance,” but in 1962 the college sought to go further, hiring John C. Hoy to make black enrollment a priority as the new dean of admissions. In two years Hoy managed to nearly triple Swarthmore’s black enrollment. This prompted Wesleyan University in 1964 to lure Hoy away from Swarthmore. At Wesleyan, Hoy transformed admissions policies. As he reported in 1967, “[s]tandard selection procedures have been found inadequate” to increase black enrollment and therefore “test scores were not to be used in the same way” for black applicants. Under this new approach, black enrollment increased 1,457%, from 0.7% of the 1964 freshman class (when Hoy was hired) to 10.9% of the 1967 freshman class.

As documented in Karabel’s The Chosen, this competition to enroll black students became a critical component of how Harvard, Yale, and Princeton operated in the 1960s. Between 1963 and 1966, Princeton had a roughly 200-point SAT difference for black students and the overall student population (for this period, Princeton’s black students averaged 550 verbal and 590 math on the SATs; the class overall averaged 650 verbal and 695 math). Not to be outdone, Harvard admitted 90 black freshmen in 1969, almost 8% of the student body. Princeton pushed ahead in 1970, using racial preferences to increase blacks to 10.4% of the student body, the highest percentage of any of the Big Three. Princeton thus went from being a school that did not admit a single black student for three consecutive years in the 1950s to being more than 10% black in 1970. Well before Bakke, American higher education was on the diversity path.

The Price of Academic Standards

It is easy to look at this period and blame these colleges and universities for creating the racial preferences that would end up swallowing higher education. But it is important to remember that elite schools were actually reluctant to abandon their academic standards. In the ’50s, many of these schools fought tooth and nail to maintain their academic requirements while boosting black enrollment. They experimented with various efforts, such as outreach strategies and special training programs, that could increase black enrollment while maintaining at least similar SAT requirements.

For example, Karabel recounts how, in 1960, after Yale’s black recruitment efforts managed to yield only five black students out of a class of 1,000, Yale considered whether it should formally lower admissions standards for black applicants. The dean of admissions rejected this proposal on the grounds that black applicants should be expected to “meet the same standards required of other applicants.” Yale seemed to take this expectation seriously, as reflected in how, after Yale’s black recruitment program found a student who ranked number one in his high school class of 500, Yale still rejected him because he scored “only 488 on the SAT,” well below the school’s bottom tenth percentile. This rejection of a black student that Yale had specifically recruited was, according to the admissions office, “the price we pay for our academic standards.”

Schools like Yale ultimately became unwilling to pay the price of uniform academic standards, once the civil rights revolution made black representation a source of prestige and it became clear that lowering standards for black applicants was the only way to increase black representation. A breaking point came in 1966, when the first major collection of data on the black-white achievement gap appeared in the Equality of Educational Opportunity study (now known as the Coleman Report, after its lead author, Johns Hopkins sociologist James S. Coleman). The Coleman Report was a product of Section 402 of the Civil Rights Act of 1964, which provided that, within two years, the Commissioner of Education must submit a report to the president and Congress “concerning the lack of availability of equal educational opportunities for individuals by reason of race, color, religion, or national origin in public educational institutions.” As a co-author of the report recalled, there was a clear agenda driving the study—to show “that the South was discriminating by having lousy schools for poor and minority kids.”

The hope was to use the report as a retroactive justification for the Civil Rights Act by showing that testing differences between white and black students could be remedied through federal funding and integration measures. The authors went into the study expecting that region, school infrastructure, and funding would be bigger factors than race in terms of academic performance. Coleman publicly stated that he was sure that the data “would show that the low Negro test scores were the fault of poor schools.” The statistician on the survey recounted, “[W]e really thought that the Northeastern Negro would score higher than the Southern whites.” They were shocked to discover that race itself (as opposed to school region or funding) was the driving factor behind the achievement gap. In the statistician’s words, “the magnitude of the black-white difference and the uniformity over the country was mind-boggling.”

The following year, a Harvard study on race and SAT performance concluded that “only 1.2 percent of the nation’s male black high school graduates could be expected to score as high as 500 on the verbal section of the SAT and a mere three-tenths of one percent as high as 550.” To put that in perspective, that same year the median SAT scores for Harvard-admitted students were 697 verbal and 708 math. Only about 1% of the nation’s black male students were able to score on the SAT roughly 400 points below Harvard’s average at the time.

Such studies on race and test scores reinforced the idea that if schools wanted to increase black representation, they would need to treat scores differently on the basis of race. But even as schools began doing this more aggressively and explicitly in the late 1960s, a significant percentage of university administrators may have thought that these preferences would not be in place for long. Even in the face of the Coleman Report, many elites assumed that the black-white achievement gap would quickly be reduced through various public and private programs.

The achievement gap did not go away as expected, so the racial preferences did not diminish as hoped. Over the next generation university racial preferences actually strengthened in force. In their book, No Longer Separate, Not Yet Equal (2009), sociologists Thomas J. Espenshade and Alexandria Walton Radford document how racial preferences grew in the 1980s and ’90s. For example, in 1997 black applicants to private colleges had a 310 SAT advantage over white applicants—meaning that “[a] black candidate with an SAT score of 1250 could be expected to have the same chance of being admitted as a white student whose SAT score is 1560, all other things equal.” That is 50% greater than the 200-point preference that schools like Princeton had applied 25 years earlier.

Diversity and Polarization

We are now more than two generations removed from when affirmative action took over the nation’s leading colleges and universities, and yet the achievement gap obstinately remains. Notwithstanding the transformations to American life wrought by the civil rights revolution—including the overhauling of our constitutional order through anti-discrimination, education, and housing programs—group differences persist. If academic institutions insist on racial diversity—which has come to mean, among other things, a student body that is at least 10% black—schools will need to consider race in admissions decisions, either through the front door of explicitly applying different admissions criteria to different racial groups (as many liberals prefer) or through the back door of manipulating admissions criteria in outcome-oriented ways (as many conservatives prefer).

The similarity between the above-mentioned 1967 Harvard study and the data recently collected on Harvard admissions for the pending affirmative action litigation is revealing. According to both the 1967 Harvard study and the expert report in the pending Harvard case, if the college admitted students only in the top decile of academic performers, Harvard would be less than 1% black. In other words, in more than two generations of civil rights measures, creating what Christopher Caldwell has aptly described as “the mightiest instrument of domestic enforcement the country had ever seen” and “the largest undertaking of any kind in American history,” racial preferences are still necessary—to virtually the same extent—to produce a “critical mass” of black students at Harvard.

But something has changed over this time—what it means to produce that “critical mass.” In 1967, the push was to get close to proportionality and make Harvard 8% black. Now the push is for Harvard to go beyond proportionality. Indeed, even though blacks would still be less than 1% of the Harvard student body without affirmative action, they now constitute over 15% of the student body with affirmative action.

So here we are, with a Court, and a citizenry, exhausted from tinkering with the machinery of diversity, but unsure how our institutions can be extricated from this enterprise. This has come to be the antinomy of a nation defined by its devotion to racial diversity. One part of the population is increasingly frustrated with what diversity requires of us as a nation, and another part responds with a strengthened commitment to using government and corporate power to manage and control the diversity experiment. The frustration with and commitment to diversity fuel one another, increasing with each turn of the wheel the possibility of a major conflict.

Whatever comes of this rising tension, one thing is clear: neither the Court nor the colleges are willing to let the chips fall according to merit. It is therefore almost inconceivable that the Supreme Court will “eliminate” affirmative action—not this summer, not by the 25-year Grutter mark, which will arrive in 2028, and not even by 2053, 25 years after that point. This is what the civil rights revolution has come to mean: affirmative action now, affirmative action tomorrow, affirmative action forever.