Race-Neutrality, Baselines, and Ideological Jujitsu After Students for Fair Admissions

Introduction

In the hand-wringing conclusion to an oft-cited but rightly infamous article, Professor Herbert Wechsler lamented that he had not yet overcome the “challenge” of defending the result in Brown v. Board of Education^[1] against the charge that there is no “basis in neutral principles for holding that the Constitution demands that the claims” of Black Americans not to be treated as outcasts prevail over the claims of white racists to employ state power to enforce segregation.^[2] Wechsler was laughably obtuse about Brown^[3] and also misguided in his insistence that neutrality should be the sine qua non of adjudication,^[4] but he was right about how litigation works.

Legal principles are abstractions that cannot be confined to the context in which they are launched. They are “neutral” (as Wechsler insisted) at least in the sense that a conservative court can use a legal rule or standard first articulated by liberal jurists to serve conservative ends, and vice-versa. The abstractness of legal principles thus enables agile lawyers to engage in a kind of ideological jujitsu—turning opponents’ strengths against them.

The Roberts Court’s treatment of Brown exemplifies such ideological jujitsu. In defining equality under the Constitution as color-blindness, full stop, the Court has repeatedly invoked Brown, despite the protests of dissenting Justices.^[5] Those dissenters have tended to view Brown as the fulfillment of both halves of the first Justice Harlan’s dissent in Plessy v. Ferguson.^[6] After all, the two sentences that precede Harlan’s statement that the “Constitution is color-blind” espouse an anti-caste principle that laws and policies promoting racial integration advance rather than violate.^[7] As color-blindness’s critics see it, the constitutional obligation for government actors to look past race has substantially less force when used to ameliorate, rather than to perpetuate, racial subordination.^[8]

The champions of color-blindness tout court have responses, of course: no use of race can simply be assumed benign without testing through the crucible of strict scrutiny;^[9] race-conscious ameliorative measures themselves reinforce the stereotypes their proponents seek to undercut;^[10] people who did not perpetrate white supremacy should not have to bear the cost of remedying it even if they benefit from it;^[11] and so forth. And the critics of color-blindness have their surrebuttals, including the accusation that the principle’s champions are simply reading their preferences into a constitutional provision proposed by a Reconstruction Congress that neither preached nor practiced color-blindness.^[12]

To those of us who study and practice constitutional law, the debate is familiar and, after a half century, tired. It is also over. Alas, we who thought race-based affirmative action sensible—or at least constitutional—have lost. Color-blindness has won, partly through the accidental timing and hardball politics of Supreme Court appointments, but also because race-based affirmative action has been unpopular for at least a generation: Even as they were delivering large majorities to Democratic candidates, voters in ostensibly liberal California forbade it in 1996^[13] and reaffirmed that judgment in 2020.^[14] In between, in Michigan in 2006, voters decisively re-elected their Democratic governor while simultaneously and slightly more decisively forbidding affirmative action.^[15] The Court’s composition and our politics could eventually change again, but for at least the medium-term the question is not whether the Constitution is color-blind but what, as a practical matter, color-blindness means.

The answer is not obvious. On the last two pages of the majority opinion in Students for Fair Admissions, Inc. v. President and Fellows of Harvard College^[16] (hereinafter SFFA), Chief Justice Roberts simultaneously invited and sought to ward off subterfuges reminiscent of both the massive resistance to Brown^[17] and hostility towards (including routine violations of) the Court’s school prayer rulings.^[18] Despite devoting nearly the entire opinion to explaining why universities^[19] like Harvard and the University of North Carolina could no longer treat student body diversity as the sort of compelling interest that justifies race-based plus factors in admissions,^[20] the critical paragraph reassured readers that the Court was not “prohibiting universities from considering an applicant’s discussion of how race affected his or her life, be it through discrimination, inspiration, or otherwise.”^[21] However, the Court warned in the very next sentence, “universities may not simply establish through application essays or other means the regime” that the Court held invalid in the balance of the opinion.^[22]

There is no caste here; the Constitution is color-blind. Universities may consider an applicant’s experience of race; they may not consider race itself. Here we see history repeating itself as both tragedy and farce.^[23] Or so it might appear to admissions officers, the lawyers who advise them, and the lawyers who will inevitably sue on behalf of rejected applicants claiming that an admissions essay ostensibly asking about applicants’ experience of race was really a covert means of discovering and illicitly weighing each applicant’s race as such.

The seemingly scholastic distinction between racial experience and race itself implicates two broader questions with important practical implications. The first is whether facially race-neutral criteria adopted with the aim of increasing or maintaining racial diversity should be deemed race-based and thus trigger strict scrutiny. Prior to SFFA, that question was often discussed in connection with admissions schemes in Texas and elsewhere that guarantee admissions to state universities to students whose high school grade point averages were in some top percentage of their respective classes.^[24] Affirmative action’s opponents tendered such programs as race-neutral alternative means of achieving racial diversity that, in their view, demonstrated expressly race-based measures were unnecessary;^[25] affirmative action’s defenders responded that the percentage plans were not actually race-neutral, chiefly because they leveraged de facto residential segregation.^[26]

Since SFFA, the political valence of these positions has flipped. In a display of ideological jujitsu, opponents of affirmative action who have turned their sights on all efforts to promote racial diversity now deploy the arguments formerly used by their adversaries to attack nominally race-neutral means of increasing or maintaining racial diversity.^[27] Diversity’s defenders respond in turn by discovering that the likes of the George W. Bush administration were right after all when they described facially race-neutral means of achieving diversity as race-neutral, full stop.^[28]

Who’s right? The question is open, but this Article argues that at least some facially race-neutral selection criteria adopted for the purpose of increasing or maintaining racial diversity should count as race-neutral rather than race-based. It practices ideological jujitsu to leverage statements made and positions embraced by conservative Justices in the pre-SFFA period for validating the rearguard efforts that affirmative action’s erstwhile champions (including the current author)^[29] must now undertake to defend facially race-neutral efforts to sustain some measure of racial diversity in higher education.

The claim that race-conscious but facially race-neutral means of increasing or maintaining racial diversity should be deemed race-neutral is not new. Consider a recent article by Professor Sonja Starr focusing chiefly on competitive admissions to secondary schools.^[30] She offers a powerful argument that a wide range of Supreme Court case law undisturbed by SFFA assumes the permissibility of what she calls facially race-neutral but race-conscious “ends-colorblindness” while forbidding only “means-colorblindness.”^[31] In addition to defending the former on substantive grounds, Professor Starr acknowledges that other scholars critical of color-blindness also sought to defend ends-colorblindness many years before the Supreme Court’s decision in SFFA.^[32] In the same vein, a forthcoming article by Professor Deborah Hellman makes a powerful argument for disentangling the presumptive invalidity of expressly race-based classifications from a separate principle forbidding the use of race-neutral means with the intent to harm.^[33]

Originality aside, it is not clear how this sort of argument will ultimately play out. Courts hostile to diversity as such could reverse their prior position to conclude that percentage plans and other facially race-neutral selection criteria are impermissibly race-based after all. Diversity’s champions must be prepared to make additional arguments in favor of sustaining such criteria. Therefore, this Article develops another basis for sustaining race-neutral criteria, one that uses ideological jujitsu to build on a more than fifty-year-old precedent that civil rights advocates have long disdained: Palmer v. Thompson.^[34]

In response to desegregation orders, Jackson, Mississippi closed all of its public swimming pools.^[35] The lower federal courts found no equal protection violation, and a 5–4 divided Supreme Court affirmed.^[36] The Court first opined categorically that an illicit purpose does not render an otherwise valid law invalid.^[37] That aspect of Palmer’s holding is clearly bad law. It was effectively repudiated just five years later in Washington v. Davis.^[38] However, an important piece of an alternative rationale offered by the Palmer Court remains good law.^[39]

Palmer’s alternative rationale has three steps: (1) Jackson had no obligation to establish public swimming pools in the first place; (2) its elimination of the public swimming pools affected persons of all races equally; and (3) therefore, it was a permissible instance of leveling down.^[40] If Palmer was wrongly decided on its facts, that is because proposition (2) is wrong: Although the closure of the public pools denied Jackson residents of all races the ability to swim in a public pool, it worked additional harms—including stigmatic harm—to Black but not white residents. However, even if (2) is wrong and Palmer was therefore wrongly decided in its bottom line, proposition (1) is valid and potentially extremely useful in defending the constitutionality of facially race-neutral means of increasing or maintaining racial diversity.

To see how, consider a schematic example. Suppose that an undergraduate admissions office that had formerly used race as a plus factor in accordance with the pre-SFFA law discovers that its enrollment of Black, Latinx, and Native American students drops substantially after it shifts to wholly race-neutral admissions criteria such as high school grades and standardized test scores.^[41] It responds by shifting again, now to an admissions lottery for which all students above a minimum threshold are eligible. As a result, the overall median test score and median GPA of the admitted class fall somewhat but enrollment of the relevant underrepresented minority groups rebounds to close to pre-SFFA levels. Correspondingly, however, the percentage of Asian American and white students declines to roughly pre-SFFA levels. Is the adoption of the lottery, therefore, a violation of the equal protection rights of the Asian American and white applicants who do not gain admission under the lottery but would have been admitted under the admissions criteria that immediately preceded it?

One way to resist that conclusion would be to emphasize that the lottery is facially race-neutral—that despite the Supreme Court’s seeming endorsement of total color-blindness, there remains a constitutionally significant distinction between aiming to perpetuate racial hierarchy and other race-conscious motivations. That is the approach of Professor Starr and the scholars whom she cites as having anticipated her argument. As noted above and elaborated below, there is much to be said for this approach, but a very conservative judiciary might well resist it.

Accordingly, diversity’s proponents and defenders need an alternative path. The heart of this Article offers one: defend the lottery by invoking Palmer to deny that it has any constitutionally significant disparate racial impact. Asian American and white applicants fare worse (on average) under the lottery than they did under the grades-plus-test-scores regime that immediately preceded it, but there is no reason to treat the immediately prior regime as the yardstick against which to detect a disparate impact. As the United States Court of Appeals for the Fourth Circuit recently observed, there is “no precedent standing for the proposition that a particular racial or ethnic group’s performance under a prior policy is ‘the proper baseline for comparison’ in a disparate impact inquiry concerning a newly enacted policy.”^[42]

Maybe not, but is there precedent for rejecting the immediately prior regime as the relevant baseline? The Fourth Circuit quoted Village of Arlington Heights v. Metropolitan Housing Development Corp.^[43] for the proposition that a policy has a disparate impact only if it “bears more heavily on one race than another.”^[44] Yet that language is not dispositive. A lottery or a similar facially race-neutral selection mechanism might bear no more heavily on one race than another as measured against each group’s proportion in the general population or applicant pool but nonetheless might bear more heavily on some groups than others relative to the prior regime. The Arlington Heights language does not resolve the key question.

To be sure, the Arlington Heights Court seemingly endorsed population proportions as the baseline for measuring disparate impact,^[45] but the Court in that case was not faced with any claim that the prior regime was the correct yardstick—and in any event, its discussion of that issue was essentially dicta, as the case went on to hold that the plaintiffs had failed to adduce sufficient evidence of race-based purpose.^[46]

Nonetheless, the Fourth Circuit was correct about the baseline question, even though the court merely asserted the point while citing inapposite authority. Case law construing statutory disparate-impact claims confirms the claim. In statutory cases, courts typically look to applicant pool or population data—not simply to whatever policy was in place before the adoption of the challenged policy—to determine the relevant denominator.^[47] And SFFA relies on the longstanding treatment of federal statutory antidiscrimination law as equivalent in key features to constitutional equal protection.^[48] Moreover, Palmer provides powerful grounds to support and justify rejecting the status quo ante as a baseline for finding disparate impact, if only diversity’s proponents would be willing to invoke it.

The balance of this Article proceeds in four parts. Using percentage plans as the key example, Part I evaluates the claim that a facially race-neutral policy adopted for the purpose of increasing or maintaining racial diversity counts as race-based and should thus trigger the same strict scrutiny as facially race-based criteria. Offering arguments that complement those made by other scholars,^[49] it suggests that notwithstanding the current Supreme Court’s embrace of color-blindness, there is room to argue that some such policies should be treated as truly race-neutral.

Parts II through IV then turn to this Article’s primary original contribution to the literature: the Palmer-based argument. After explaining why some facially race-neutral policies present idiosyncrasies that make generalization difficult, Part II refines the analysis by settling on an admissions lottery as the ideal hypothetical test case.

Part III develops the Palmer-based argument. It turns to statutory employment discrimination law as an area where courts have grappled with the baseline question and rejected the notion that disparate impact should be measured relative to whatever the status quo ante happened to be. It acknowledges that the existing status quo is sometimes the appropriate baseline for measuring a policy’s legal impact—with property rights as the primary example—but explains that this approach is inappropriate in most equal protection cases, including the hypothetical admissions lottery.

Part IV addresses a potentially troubling counterargument. It asks whether reliance on Palmer proves too much. Would it validate race-neutral policies adopted for the illicit purpose of disadvantaging a racial minority group so long as the disadvantage does not fall below some population-based threshold? Part IV explains that this question really asks whether Palmer was right on its facts. I suggest that advocates of diversity who wish to avoid the disturbing implication can rely on narrower critiques of Palmer that emphasize animus and stigma, while preserving Palmer’s insight that the status quo ante is rarely the appropriate baseline for measuring disparate impact.

The Article concludes with the admonition that lawyers and legal scholars should never hesitate to deploy ideological jujitsu even while acknowledging that its efficacy depends ultimately on judges’ ability and willingness to be principled.

I. How Color-Blind Is the Color-Blind Constitution?

To defend a facially race-neutral selection device against the charge that it unlawfully discriminates based on race, we must first understand exactly what that charge entails. Its rudiments are straightforward: (1) Washington v. Davis and subsequent cases make clear that if a government entity or private actor subject to anti-discrimination law adopts a selection device with the purpose of disadvantaging some racial minority group, and the selection device in fact results in disparate disadvantage to that minority group, the action is equivalent to the use of an express racial classification and will therefore be unlawful unless it can satisfy the nearly always fatal strict scrutiny standard;^[50] (2) SFFA and indeed many cases that preceded it hold that racial classifications trigger strict scrutiny whether they advantage or disadvantage racial minorities;^[51] and thus, (3) combining these two principles, a facially race-neutral selection device that is adopted for an assertedly benign race-based purpose and has its intended disparate impact—being equivalent to a facially race-based selection device—must also be subject to strict scrutiny.

Assuming one accepts the premises, the logic of what I shall call the Davis/SFFA syllogism seems compelling. Yet for years, conservatives pointed to percentage plans as a race-neutral alternative to expressly race-based affirmative action,^[52] even though a percentage plan precisely fits the paradigm just described. Indeed, under the logic of the Davis/SFFA syllogism, percentage plans would be problematic even if they did not leverage de facto residential segregation to generate the intended effect. Why, then, did affirmative action’s opponents endorse percentage plans as race-neutral alternatives? And how might they and other facially race-neutral selection criteria that aim at increasing or maintaining racial diversity now be characterized as race-neutral before a Court that will be suspicious of efforts to evade SFFA?

A. Inattention, Lesser of Two Evils, or Opportunism

One possible explanation for affirmative action’s critics’ prior acceptance of percentage plans is that they simply were not focusing on the Davis-based argument when they offered such plans as race-neutral alternative means. To be sure, inattention might seem an unlikely explanation because affirmative action’s defenders called attention to percentage plans and argued that they are not neutral in design or operation. Dissenting in Gratz v. Bollinger,^[53] Justice Souter, joined by Justice Ginsburg, stated flatly that “‘percentage plans’ are just as race conscious as” expressly race-based affirmative action, “but they get their racially diverse results without saying directly what they are doing or why they are doing it.”^[54]

Nonetheless, we should not rule out inattention. Justice Souter’s observation was a response to amicus briefs contending that the University of Michigan’s express use of race in undergraduate admissions was not narrowly tailored because percentage plans are a race-neutral alternative.^[55] The majority in Gratz did not invoke the percentage plans and so would have had little reason to defend their race-neutrality. Then, in two subsequent Supreme Court cases involving the University of Texas, the contested issue was whether supplementation of a percentage plan with some express use of race was permissible.^[56] It was taken for granted—or at least not contested—that the percentage plan itself was race-neutral.

Next, consider another possible explanation for the views affirmative action’s opponents hitherto espoused: They might have accepted percentage plans as a lesser of two evils. Perhaps they were uneasy with such plans but thought them less problematic than expressly race-based affirmative action. It is possible to think that two courses of action are both illegal but that one is in some sense more illegal than the other, or even that one is more unconstitutional than the other.^[57]

The lesser-evil hypothesis could explain why we are now beginning to see challenges to facially race-neutral selection criteria such as admissions lotteries.^[58] Having vanquished what they regard as the greater evil of express racial preferences, they can now begin a mopping-up operation to stamp out the lesser evil of facially race-neutral schemes for increasing or maintaining racial diversity, which opponents regard as proxies for the now-forbidden expressly-race-based criteria.

And of course, it is also possible that some of affirmative action’s opponents formerly touted the percentage plans disingenuously. Maybe they intended all along to mount or support challenges to the percentage plans as the second phase of a legal strategy.

In any event, we need not dwell further on the subjective motivations of opponents of affirmative action who previously seemed to accept as race-neutral selection criteria that have the purpose and effect of increasing or maintaining racial diversity. The question going forward is whether it is possible to articulate a persuasive argument against the Davis/SFFA syllogism that will appeal to a Court that is hostile to expressly race-based affirmative action. The next subpart attempts that task.

B. Purpose to Disadvantage Versus Purpose to Achieve Diversity

Despite the seeming inexorability of the Davis/SFFA syllogism, the Supreme Court has never had occasion to adopt or reject it. An argument against the syllogism might run through a narrow reading of Davis. In Davis, the use of “Test 21” as a screening mechanism by the Washington, D.C. police department had a negative disparate impact on Black applicants but did not trigger strict scrutiny because there was no showing of discriminatory purpose.^[59] What, exactly, would count as a discriminatory purpose?

Notably, the Davis Court in some places referred to “purposeful discrimination,” but even more often identified purposeful “invidious” discrimination as the missing element that would transform a mere disparate racial impact into the equivalent of an express racial classification.^[60] The modifier “invidious” or “invidiously” appears fully nine times in the Davis majority opinion.^[61] The other leading disparate impact cases likewise identify invidious discriminatory purpose as the secret sauce that renders disparate impact subject to strict scrutiny.^[62]

Thus, we have the elements of an argument to break the syllogistic chain: Whereas SFFA and other affirmative action cases require that all expressly race-based classifications must be subject to strict scrutiny, they do not bear on race-neutral means of achieving diversity or some other assertedly benign goal that in some way implicates race. To trigger strict scrutiny, it does not suffice that a policy with a racially disparate impact was adopted for the purpose of having that impact; it must have been adopted for an invidious discriminatory purpose. And the goal of increasing or maintaining racial diversity is not invidious.

Hold on. Doesn’t SFFA say just the opposite?

Actually, it does not. SFFA holds that expressly racial plus factors are not narrowly tailored to promoting racial diversity, in part because the Court found that the benefits of diversity could not be sufficiently concretely quantified to facilitate the narrow tailoring analysis.^[63] But nothing in the opinion suggests that diversity is not a legitimate government interest.

By contrast, an invidious purpose is, by definition, illegitimate. It carries with it a connotation of what the Court, in a different context, has sometimes condemned as animus.^[64]

Consider the Court’s explanation for why it applies strict scrutiny to all facially race-based classifications. As Justice O’Connor wrote for the Court in City of Richmond v. J.A. Croson Co.,^[65] “the purpose of strict scrutiny is to ‘smoke out’ illegitimate uses of race,”^[66] or what we might call invidious uses of race. By negative implication, some uses of race are therefore not invidious.

To be sure, if strict scrutiny of racial classifications is merely strict in theory but fatal in fact,^[67] then all racial classifications are impermissible in fact, or, to put it differently, all racial classifications are invidious. Maybe Justice O’Connor’s opinion in Croson allowed for the possibility that some uses of race in the affirmative action context might survive strict scrutiny because she was less hostile to affirmative action than the Court’s most committed champions of color-blindness? That is a plausible supposition because she would later write for the Court in Grutter that “[s]trict scrutiny is not ‘strict in theory, but fatal in fact.’”^[68] That statement appeared in a majority opinion that spoke for Justice O’Connor and four Justices to her left on the issue.^[69] The Court’s right-leaning Justices were in dissent.^[70] Now that the Court’s conservative super-majority has disavowed Grutter’s bottom line,^[71] we might think that strict scrutiny is once again fatal in fact in all cases, including those involving affirmative action. And if all racial classifications are invalid, then they are all invidious.

Does it follow that the Davis/SFFA syllogism holds after all? To answer that question, consider where the doctrine is headed on the assumption that the Roberts Court has fully committed to color-blindness: Once the state (or a private actor subject to an equivalent norm) uses a racial classification, a purpose to achieve racial diversity and every other assertedly benign purpose will trigger scrutiny no less strict than that which applies to racial classifications adopted for classically racist purposes. Purpose no longer constitutes a basis for adjusting the strictness of the strict scrutiny.

It does not necessarily follow, however, that the Court’s conservatives are committed to treating the purpose or goal of advancing racial diversity as equivalent to classically racist purposes in all contexts. It remains an open question whether the secret sauce that must be added to a disparate racial impact to trigger strict scrutiny is any racially related purpose—as implied by the Davis/SFFA syllogism—or whether something more, an invidious racial purpose—as described in Davis itself—is required.

But why would a Justice who thinks all racial purposes trigger genuine strict scrutiny of express racial classifications distinguish between invidious and non-invidious purposes when considering the threshold question of whether a particular disparate impact of a facially race-neutral policy results from an impermissible purpose? The answer may well be that racial classifications are especially dangerous in ways that disparate impacts—even if the intended result of a broader strategy—are not.

To state the obvious, racial classifications classify people by race, thus potentially reducing them to tokens or stereotypes.^[72] By contrast, a policy adopted because of its disparate racial impact but (by hypothesis) not invidiously so chosen, does not classify anyone. Under the percentage plan, a white or Asian student who excels at an overwhelmingly Black or Latinx high school in Texas gets the exact same admissions benefit as anyone else who attends that school. The state knew when it adopted the percentage plan that it would, in the aggregate, promote a diverse student body in the state’s public universities, but by relying on criteria that are race-neutral at the individual level, it avoided the key harms that the Court has associated with racial classifications.

Thus, even accepting the key premises underlying the Roberts Court’s endorsement of a color-blind Constitution, there remains room to argue that the Davis/SFFA syllogism does not hold—that percentage plans and other facially race-neutral policies should be deemed race-neutral, full-stop.

To be clear, I am not making any claim about the likely subjective dispositions of any particular jurists. I am not predicting that the Court’s conservative Justices will necessarily draw a distinction between benign and invidious purposes to reject the Davis/SFFA syllogism. Hostility to affirmative action or to what they might perceive as efforts to evade their rulings could lead them to invalidate percentage programs and the like.

The quite different claim I am making here has two parts: first, that the Davis/SFFA syllogism has not yet been adopted by any case, so it is an open question whether the secret sauce that subjects disparate racial impacts to strict scrutiny is any racial purpose or the higher threshold of an invidious racial purpose; and second, that adopting invidious racial purpose as the threshold would be consistent with the previously stated views of the Justices who are committed to color-blindness, given their concerns about the special harms associated with racial classifications.

II. The Variety of Facially Race-Neutral Selection Criteria

The remaining Parts of this Article explore an alternative approach to defending the constitutionality of facially race-neutral laws and policies purposely employed to increase or maintain racial diversity: relying on the logic of Palmer v. Thompson to reject the status quo ante as the appropriate baseline for measuring disparate impact. However, facially race-neutral laws and policies take a variety of forms. Some have idiosyncratic characteristics that complicate the analysis. To make the argument as broadly useful as possible, the balance of this Article uses an admissions lottery as the paradigm example.

This Part briefly explains why three other kinds of policies are suboptimal vehicles for considering the baseline question in its most general form. Those policies are: (A) the experience-of-race essay that the SFFA opinion appeared to bless; (B) Texas-style percentage plans of the sort discussed in Part I; and (C) open admissions for all applicants above a minimum threshold. This Part identifies peculiar features of each of these policies that might make them invalid or valid on grounds that do not generalize. It then explains why (D) a hypothetical lottery among minimally qualified applicants works better as a test case for considering the general question.^[73]

A. The Experience-of-Race Essay

Suppose that a university admissions office seeks to take advantage of the paragraph near the end of the SFFA opinion that seemingly approves of essay questions that ask applicants to describe how race affected their lives.^[74] Doing so will not necessarily insulate the university against liability. As the SFFA opinion itself warns, courts will scrutinize how admissions officers evaluate such essays.

For concreteness, imagine that a university admissions essay question prompts applicants to describe how “race affected [their lives], be it through discrimination, inspiration, or otherwise.”^[75] In the course of answering, many applicants will divulge their race. If a university simply uses that information as a plus factor to recreate the pre-SFFA regime, it will act unlawfully.^[76] But even if the university admissions office uses the essay in good faith, it runs a substantial litigation risk. To understand why, consider what happens if the university application asks the experience-of-race question and evaluates the answers permissibly.

Initially, we might wonder why a university would legitimately ask about an applicant’s experience with race. The answer must be that experience with race is relevant to something that the university values other than race itself. What might that be?

Perhaps the university seeks applicants with grit,^[77] as demonstrated by overcoming obstacles, and considers race discrimination among the most substantial obstacles an applicant could overcome. Or perhaps the university takes the view that in an increasingly multicultural and diverse country, substantial experiences with race will lead to more successful interactions with classmates, co-workers, and fellow citizens.

However, even if the university asks about experiences of race for these or other legitimate reasons—and not simply as a means of acquiring racial data for illicit use—the resulting admissions pattern may be difficult to distinguish from what one would see if race were used illicitly. After all, in a society in which race matters, applicants of color will typically have had more and deeper experiences involving race.

To be sure, applicants of all races will have had some experience of race. However (and at the risk of stereotyping), a cliché-filled admissions essay by an upper-middle-class white applicant describing the summer they spent as a volunteer aid worker in a developing country (e.g., “what these people who had so few material goods taught me about our shared humanity was much more valuable than the labor I performed”) will be legitimately less impressive than the essay of an applicant of color describing the impact of systemic racism on their entire life. Consequently, disadvantaged minority applicants will, on average, receive more of a boost from the experience-of-race essay than will other applicants. Yet, viewed after the fact, that statistical benefit will be difficult to distinguish—at least without substantial discovery and cross-examination—from an illicit counting of race qua race.^[78] Even if a lawsuit against the university by SFFA or like-minded groups would ultimately fail, it would survive a motion to dismiss and probably survive a motion for summary judgment.

The goal of the foregoing analysis is not to advise university admissions offices against accepting the SFFA Court’s invitation to inquire about experiences of race.^[79] Rather, the point is simply that the chief nominally race-neutral admissions tool the Court identified in SFFA—the experience-of-race essay—presents a serious evidentiary issue. Therefore, we should use a different example to interrogate broader questions about the permissibility of facially race-neutral means of increasing or maintaining racial diversity.

B. Percentage Plans

Part I observed that if the Davis/SFFA syllogism is valid, it would operate to invalidate Texas-style percentage plans even if such plans did not leverage residential segregation. But because these plans do in fact leverage residential segregation, they are a poor vehicle for interrogating the general permissibility of facially race-neutral approaches to increasing or maintaining racial diversity. To see why, it may be useful to briefly recount the history of such plans.

The best-known percentage plan was adopted by the Texas legislature in 1997^[80] in response to a 1996 Fifth Circuit ruling that, in its central features, anticipated SFFA.^[81] Subsequent Supreme Court opinions made reference to the Texas percentage plan,^[82] and the Supreme Court addressed the constitutionality of a modified version of it that added a race-specific plus factor in two cases involving the same rejected applicant in 2013^[83] and 2016.^[84] But the Court has never considered a direct constitutional challenge to a percentage plan as such.

Now consider how such a challenge might invoke the leveraging of de facto segregation, regardless of whether the courts accept the Davis/SFFA syllogism more broadly. In Texas, as in many other states, the public and private elementary and secondary schools exhibit a high degree of de facto racial segregation.^[85] Percentage plans increase or maintain the racial diversity of universities through simple arithmetic. If nearly all of the students who attend a particular high school are Black or Latinx, then nearly all of the students in the top ten percent (or some other sufficiently large percentage) of that high school’s graduating class will also be Black or Latinx. By guaranteeing top students from each of the state’s high schools admission to a state university, a percentage plan effectively conducts separate intra-racial competitions in violation of even the pre-SFFA case law, which required that all applicants be considered for all positions.^[86]

To be sure, a percentage plan does not facially segregate applicants into separate racial pools. The equivalence arises as a consequence of de facto segregation. Under the Supreme Court’s precedents, de facto segregation is not unconstitutional so long as it is not a traceable legacy of the de jure segregated schools that Texas formerly operated—and cases since the 1980s and 1990s have increasingly attributed de facto segregation to private choice (so-called white flight) rather than to state action.^[87] Even so, however, the deliberate decision of Texas (and other states that employ percentage plans) to leverage de facto segregation can be thought to impermissibly implicate the state in what would otherwise be private choices.

Consider Palmore v. Sidoti.^[88] There, the Supreme Court invalidated a family court judge’s award of custody to a child’s father, rather than mother, because the mother was part of an interracial couple and the judge thought it in the child’s best interest to be shielded from prejudice against such couples.^[89] “Private biases may be outside the reach of the law,” Chief Justice Warren Burger wrote for a unanimous Court, “but the law cannot, directly or indirectly, give them effect.”^[90] The family court judge apparently did not share the prejudiced views of the people from whom he sought to protect the child, but that was not dispositive. So too here we may assume that the Texas legislature that adopted the percentage plan did not approve of the private prejudices that led to de facto racial segregation of the state’s schools, but, as in Palmore, that is not a justification for giving them effect.

Admittedly, the analogy to Palmore is not perfect. There, the family court judge expressly considered the race of the child’s stepfather, whereas percentage plans take advantage of, but do not expressly incorporate, racial considerations. For that reason, in defending against a challenge to a percentage plan, a state could seek to distinguish Palmore and plausibly claim that the plan is at least facially neutral in a way that the custody determination in Palmore was not. However, the leveraging of de facto segregation makes the Palmore analogy at least plausible. One might therefore have reasons for concluding that a percentage plan is unconstitutional that are specific to percentage plans and not representative of the general category of facially race-neutral means of increasing or maintaining racial diversity.

C. Admit All Applicants with Minimum Qualifications

In his separate opinion (concurring in part and dissenting in more substantial part) in Grutter v. Bollinger,^[91] Justice Thomas described a hypothetical admissions policy of “accepting all students who meet minimum qualifications,” which, he said, would allow the University of Michigan Law School to achieve substantial racial diversity “without the use of racial discrimination.”^[92] It is certainly notable that Justice Thomas—who is arguably the most pro-colorblindness Justice ever to sit on the Supreme Court—thought (and apparently still thinks) that at least one facially race-neutral policy remains race-neutral even when used to increase or maintain racial diversity. That fact suggests that the argument of Part I for deeming at least some such policies truly race-neutral might succeed.

Nonetheless, a policy of accepting all applicants who meet minimum qualifications is not ideal for considering the broader issue because such a policy could be deemed permissible on the idiosyncratic ground that it does not disadvantage anyone. If a law school or other institution literally accepts all applicants above the threshold, then no one who would have been granted admission as more highly qualified under more selective criteria will be excluded based on race or otherwise. The minimum-qualifications policy might have a disparate impact on members of disadvantaged racial minority groups because they will be disproportionately less likely to satisfy the minimum; however, that effect occurs despite rather than because of the policymakers’ goal of increasing or maintaining racial diversity.^[93] Thus, the minimum qualifications policy would be deemed race-neutral under the conventional approach to analyzing disparate impact.

Put differently, a minimum-qualifications policy is atypical of facially race-neutral policies aimed at increasing or maintaining racial diversity because such policies typically involve scarce resources—such as jobs or seats in a university class. Although most community colleges and some other institutions of higher education allow open enrollment by applicants with high school diplomas or equivalency degrees,^[94] questions concerning the legality of race-neutral alternatives to race-based affirmative action in university admissions arise in institutions with highly selective admissions criteria. Yet, as Justice Thomas frankly acknowledged in his separate opinion in Grutter, the University of Michigan Law School would have to sacrifice selectivity to utilize a minimum-qualifications policy. He thought that sacrifice was required if the law school wished to pursue what he denigrated as its “aesthetic” interest in racial diversity because, he averred, a state lacks a compelling interest in maintaining a public law school at all, much less a highly selective one.^[95]

Suppose that we accept, arguendo, Justice Thomas’s view that universities must be prepared to sacrifice selectivity if they wish to pursue racial diversity. Even so, covered actors will not invariably be able to do so by adopting a minimum qualifications policy. Limited financial and other resources will often mean that there are fewer spaces available than qualified applicants. Accordingly, a fully generalizable example of a race-neutral admissions policy will deny admission to at least some minimally qualified applicants.

D. Admissions Lottery

That brings us to what the remaining Parts of this Article use as the paradigmatic example of a race-neutral university admissions policy adopted for the purpose and having the effect of increasing or maintaining the racial diversity of the student body: an admissions lottery for which all applicants above a minimum threshold are eligible.

To make the example useful, we need to specify a few characteristics. To begin, we assume that the minimal qualifications do not leverage de facto segregation in the way that percentage programs do. Suppose that to be eligible for the lottery, an applicant for undergraduate admissions must be on pace to receive a high school diploma or equivalent with at least a minimum threshold grade point average and must score above some minimum threshold on a standardized test such as the SAT or ACT.

Policymakers will set the grade and test score thresholds low enough to ensure that the qualified pool is racially diverse but high enough to ensure that everyone admitted has a substantial chance of succeeding upon enrollment. Perhaps the minimum grades and scores match those that were previously associated with the tenth percentile of admittees. No one will be admitted whose grades and test scores would have made them categorically ineligible under the pre-SFFA regime. Under that prior regime, applicants just above the minimum threshold would have had little likelihood of admission unless they were recruited varsity athletes, alumni or donor children, beneficiaries of race-based affirmative action, or truly extraordinary in some other way. Now, by contrast, everyone at or above the minimum thresholds has an equal chance of admission. Accordingly, mean and median grades and test scores of admitted applicants drop from where they stood when the admissions office differentiated between applicants with higher versus lower grades and test scores.

There are many reasons why a selective university would not choose to use an admissions lottery. The decline in the average academic qualifications of the student body will likely lead to fewer distinguished graduates; it will adversely affect the university’s ranking in influential publications like U.S. News and World Report; abandoning preferences for alumni and donor children (as would be required for a truly random lottery) will negatively impact the university’s finances; and so forth. Understandably, then, the possibility of an admissions lottery appears almost exclusively on lists of hypothetical race-neutral alternatives that critics of pre-SFFA affirmative action programs argued should have been used,^[96] not as an actual selection device—at least not in the United States.^[97]

Nonetheless, the simplicity of an admissions lottery makes it a good test case for examining the core question of what baseline to use to determine whether a facially race-neutral admissions scheme adopted for the purpose and having the effect of increasing or maintaining racial diversity is constitutional.

III. Rehabilitating Palmer to Change the Baseline

This Part uses Palmer and statutory antidiscrimination law to argue that the status quo ante is rarely the correct baseline for measuring disparate impact—and that without a disparate impact, a racial purpose, even if illicit, is, at worst, what Professor Laurence Tribe has called “an unsuccessful attempt to violate the Constitution.”^[98] Deploying the hypothetical admissions lottery as a test case, the Part begins by explaining that the basic logic of Palmer is both sound and general: There are good reasons not to measure constitutional impacts relative to whatever the status quo ante happened to be. This Part then turns to statutory antidiscrimination law to suggest what an appropriate baseline might look like. Finally, it acknowledges that there are exceptional contexts in which the status quo ante is the proper baseline but also explains why those contexts are, well, exceptional.

A. Palmer’s Sound Rejection of the Status Quo Ante Baseline

Consider a more concrete version of our hypothetical undergraduate admissions lottery. Assume that from some time in the late twentieth century through 2023, the highly selective Generic State University (GSU) admitted students based on a holistic review that used race as a plus factor as permitted by Bakke and Grutter. Following SFFA, for the 2023–2024 admissions cycle, GSU leaves other aspects of its admissions criteria in place: it assigns weight to high school grades, standardized test scores, and extracurricular activities; it gives a boost to recruited varsity athletes and children of alumni, donors, and faculty; but it no longer counts race as a plus factor. Under this approach, GSU ends up enrolling substantially fewer first-year Black, Latinx, and Native American students for fall 2024 than in prior years when race was a plus factor; Asian American and white enrollment correspondingly increases. GSU faculty and administrators are unhappy with this result. Consequently, for the 2024–2025 admissions cycle, and with the acknowledged purpose of restoring as much of the pre-SFFA racial diversity as it can, GSU shifts to a random lottery among applicants who satisfy a minimum threshold of high school grades and standardized test scores. Average academic qualifications decline, but representation of Black, Latinx, and Native American students in the class that enrolls for fall 2025 rebounds to roughly pre-SFFA levels with corresponding declines in Asian American and white enrollment. Does the GSU lottery trigger strict scrutiny?

The argument of Part I, if accepted, would avoid strict scrutiny on the ground that the symmetry demanded by the Court with respect to expressly race-based classifications does not apply to facially race-neutral policies. If accepted, that argument would mean that the lottery—despite its race-conscious purpose—would not trigger strict scrutiny even if it had a disparate impact on some racial group(s) (here, Asian American and white applicants). But the Supreme Court might not be receptive to that argument. What then?

We can use Palmer to argue that the lottery does not have a disparate impact at all. Yes, on average, Asian American and white applicants are less likely to be admitted under the lottery than under the race-blind protocol used in the 2023–2024 cycle,^[99] but there is no good reason to measure disparate impact relative to that particular protocol.

One might think that I have loaded the dice by making the status quo ante the procedure that GSU used for only one admissions cycle. But I did so only for the purpose of making the hypothetical match the timing of SFFA. Almost nothing depends on how long GSU used the race-blind grades/scores/non-race-based-boosts regime.^[100] Just as the Palmer Court did not pause to consider whether Jackson had previously operated public pools for a year or for decades, so too here, we need not concern ourselves with the duration of the prior regime. The key factor is that nothing in the Constitution required Jackson to operate public pools, just as nothing in the Constitution required GSU to apply the particular race-blind admissions criteria it used in 2023–2024. Indeed, nothing in the Constitution required the state to operate public universities at all.^[101]

But wait. Isn’t Palmer problematic?

It may well be—as I discuss in subpart IV(B)—but it is not problematic in its rejection of the status quo ante as the baseline for measuring disparate impact. The key to understanding why may be found in another case that has long made progressive supporters of civil rights uneasy: Washington v. Davis.^[102]

Davis rejected the claim that a disparate racial impact, standing alone, suffices to trigger strict scrutiny.^[103] The decision can be and has been subjected to the criticism that in so doing, the Court ignored systemic racism—patterns of institutional bias that perpetuate racial hierarchy even in the absence of express individual prejudice.^[104] I have no quarrel with that critique.

However, the Davis Court had a further rationale for its ruling. Under Title VII as it existed in 1976 when Davis was decided and today, once a plaintiff shows a disparate impact based on a protected characteristic (such as race), the burden shifts to the defendant to validate the selection mechanism as useful for distinguishing between less and more qualified candidates.^[105] The Davis Court declined to read that approach into the constitutional obligation of equal protection, partly out of concern that Congress should take the lead in deciding whether to prescribe the burden-shifting approach.^[106] The Court also worried about the potentially “far-reaching” implications of a contrary holding, which “would raise serious questions about, and perhaps invalidate, a whole range of tax, welfare, public service, regulatory, and licensing statutes that may be more burdensome to the poor and to the average black than to the more affluent white.”^[107]

Whatever one makes of that concern in the context of Davis, measuring disparate impact by reference to the baseline of the status quo ante would greatly magnify it because just about every policy change will have a disparate impact on some protected group by that measure.

Suppose that a state university has hitherto given admissions preferences to children of alumni, donors, and faculty, all of whom are disproportionately white relative to both the applicant pool and the population. Those preferences were not a denial of equal protection because (we will assume) the university provided them despite rather than because of their disparate racial impact.^[108] But now suppose that the university wishes to abolish the preferences at least in part because it has come to regard the disparate racial impact as contrary to its inclusive mission.^[109] Eliminating the preferences will have a disparate racial impact on white applicants, as measured against the prior regime of preferences. And because the purpose of eliminating the preferences is inextricably intertwined with race, that would mean the preferences’ elimination must be subject to strict scrutiny—at least if one rejects the argument of Part I for treating racially inclusive purposes differently from express racial classifications. Yet, however one gets there, that result seems preposterous. Surely an action that merely takes away a privilege disproportionately enjoyed by whites does not count as a negative disparate impact on whites simply because they formerly disproportionately enjoyed that privilege.

QED?

Maybe not. Perhaps readers are once again thinking I have loaded the dice here by choosing an example in which the status quo ante is manifestly unfair. The children of alumni, donors, and faculty did not deserve the admissions preferences in the first place. Thus, the preferences’ elimination simply puts their erstwhile beneficiaries on an equal footing with all other applicants. By contrast, this objection continues, when GSU abandons its 2023–2024 admissions scheme for a lottery, its action disproportionately disadvantages Asian American and white applicants relative to a meritocratic approach in which they deserved to be admitted.

The objection should fail, even if we change our core lottery hypothetical somewhat to imagine that the 2023–2024 admissions criteria were based entirely on grades and test scores—i.e., that they were entirely meritocratic (as such things are typically conceived). The reason the objection fails is that meritocracy is only one, highly contestable,^[110] conception of fairness, but there are a great many others, including allocating scarce resources through a mechanism that gives everyone capable of using those resources an equal chance at them—i.e., the lottery that GSU adopts for 2024–2025. Unless GSU was somehow required to use the criteria it did in 2023–2024—and again, it plainly was not—then the mere fact that it happened to do so prior to adopting the lottery is no reason to make those criteria the baseline for measuring disparate impact.

Let us engage in one more bit of ideological jujitsu by using for progressive ends a Supreme Court loss for affirmative action. In Schuette v. Coalition to Defend Affirmative Action,^[111] a 6–2 Court rejected a challenge to a Michigan ballot initiative that banned race-based affirmative action in state entities.^[112] In response to the Court’s rulings in Gratz, the University of Michigan had modified its undergraduate admissions criteria but, as allowed by Grutter, it did not completely eliminate consideration of race.^[113] The ballot initiative overturned that decision, replacing the race-as-plus-factor regime allowed by Grutter with race-blind admissions.

In assessing a challenge to the ballot initiative, the Court fractured, producing a three-Justice plurality, a concurrence, two concurrences in the judgment, and a dissent.^[114] However, nearly all of the disagreement concerned whether and how to give effect to a line of cases articulating the principle that certain racially motivated alterations to the political process are invalid.^[115] Notably, however, neither the plaintiffs nor the dissenters claimed that the ballot initiative was invalid only in virtue of the fact that it disadvantaged minority applicants to the University of Michigan relative to the admissions policy that had preceded the affirmative action ban. The dissenters argued that the ballot initiative unfairly “changed the basic rules of the political process,”^[116] not that a state that once practiced race-based affirmative action creates a constitutionally significant racially disparate impact when it ceases to do so. And of course, none of the Justices in the majority gave any consideration to the wildly implausible claim that on average, disadvantage for a racial group relative to a prior but completely optional policy is, ipso facto, disparate racial impact.

In sum, whether or not Palmer was rightly decided on its facts, the Court in that case was right to reject out of hand the notion that a disparate impact can be demonstrated simply by comparing the racial statistics before and after a change from one otherwise lawful policy to another.

B. Statutory Baseline Measurement

If the status quo ante is not the appropriate baseline for measuring a policy’s disparate impact, what is? Statutory cases construing antidiscrimination statutes provide considerable guidance. Indeed, statutory cases expounding the meaning of disparate impact do more than provide guidance. Many of them apply of their own force—which is why Harvard, a private actor, was a respondent in SFFA.

The Supreme Court in SFFA reaffirmed Bakke’s treatment of Title VI as coextensive with equal protection,^[117] but with respect to disparate impact cases, there appears to be some daylight between the constitutional and statutory obligations.^[118] In equal protection cases, a plaintiff must show disparate impact and discriminatory intent to trigger strict scrutiny.^[119] In statutory employment discrimination cases under Title VII, by contrast, a showing of disparate impact, by itself, suffices to shift the burden to the defendant to validate the selection mechanism that produces the disparate impact.^[120] To be sure, Title VI does not expressly include a disparate impact provision, but longstanding regulations apply one.^[121] And while those regulations are not enforceable through private lawsuits,^[122] they are enforceable administratively.

Here, then, is the worry that directly implicates Title VI, even if one concludes per Part I that race-neutral means that have the purpose of increasing or maintaining racial diversity really count as race-neutral: A future (Republican) administration might conclude that because such means have a disparate impact relative to the status quo ante, they are ipso facto forbidden by the Title VI regulations. To ward off that conclusion, therefore, it is important to develop the argument that the status quo ante is not the right baseline for measuring disparate impact. Accordingly, I turn now to the statutory case law to develop that argument.

The Supreme Court has sometimes assumed that disparate impact should be measured simply by comparing the success rate of applicants of various races. For example, in a leading Title VII case best known for its holding that absence of “bottom line” disparate impact does not vitiate disparate impact at an earlier stage of the selection process, the Court simply stated without further elaboration that the challenged test had a disparate racial impact because Black applicants passed at a rate of approximately 54% whereas white applicants had a nearly 80% pass rate.^[123] However, while the applicant pool will often be the right baseline, that is not inevitably so. The applicant pool might be an improper baseline for two sorts of reasons.

First, there could be racial disparities in the relevant qualifications of the applicant pool. As the U.S. Court of Appeals for the Second Circuit put the point in a 2020 case quoting a 1989 Supreme Court opinion, in a typical Title VII case alleging disparate impact in hiring, “the relevant comparison is between ‘the racial composition of the at-issue jobs and the racial composition of the qualified population in the relevant labor market.’”^[124]

Arguably, looking at the “qualified population” unfairly disadvantages disparate impact plaintiffs by double counting. After all, in a disparate impact case, the plaintiffs’ claim is that the defendant has, via a race-neutral selection device, disproportionately screened out more members of one or another race. The claim, in other words, is that these plaintiffs are qualified even though the defendant has used criteria that deem them unqualified.

That objection calls for a refinement, not a complete rejection, of looking at the qualified applicant pool. What the courts really mean—what they ought to say—is that the baseline should be the otherwise qualified applicant pool, where the other qualifications are not challenged. For example, suppose a trucking company administers a test to prospective employees that allegedly has a racially disparate impact. The relevant baseline for measuring such an impact can uncontroversially be restricted to the pool of applicants or potential applicants who have the legally mandatory license to drive trucks. Because plaintiffs are challenging the test but not the license requirement, it is permissible to narrow the baseline pool based on who has a license.

In the previous paragraph, I deliberately referred to a pool of not merely applicants but also potential applicants. That brings us to the second way in which the applicant pool, standing alone, is not necessarily the right baseline for measuring disparate impact: The challenged selection process itself will frequently shape the applicant pool.

Consider International Brotherhood of Teamsters v. United States.^[125] The defendant argued that persons who had not applied for a position should not be able to complain about not being selected for it, but the Court rejected the argument on the ground that the discriminatory policy itself could have discouraged Black and Latinx applicants.^[126] Although the issue arose in that case at the remedial phase of a Title VII claim, it can also affect a showing of disparate impact in the first place. Thus, the Supreme Court relied on International Brotherhood in a disparate impact case, Dothard v. Rawlinson,^[127] to conclude: “The application process might itself not adequately reflect the actual potential applicant pool, since otherwise qualified people might be discouraged from applying because of a self-recognized inability to meet the very standards challenged as being discriminatory.”^[128]

It is easy to see how this principle could apply in university admissions. Suppose that a university announces it will admit all applicants with a high school GPA above 3.8 and an SAT score above 1400, while rejecting all other applicants. A few students who fall below one or the other threshold might apply because they do not carefully read the admission criteria; but given the formulaic nature of the process, nearly everyone who would be rejected simply will not apply. Applicants of all races will have a nearly 100% admissions rate, even though the criteria have a substantial racially disparate impact relative to the potential applicant pool.

As this brief discussion suggests, there is no one-size-fits-all rule for determining the correct baseline for measuring disparate impact. Thus, the guidelines for the Department of Justice, which is one of the agencies that enforces Title VI, acknowledge that “[d]etermining the population to which the challenged policy is applied or area the policy actually affected can present a challenging, fact-intensive element of proof.”^[129]

There is also, no doubt, a substantial element of subjective judgment in determining, for example, the relevant geographic pool. If a prestigious university in California has traditionally drawn 75% of its students from California, should disparate impact be measured separately for in-state versus out-of-state applicants? Should the baseline be the same for all out-of-state applicants or must those be further subdivided by the region or state from which applicants hail? Answering such questions involves more than finding facts.

To be sure, in many circumstances, the baseline question is relatively easy to answer. Federal courts have been deciding disparate impact cases for over five decades. Rarely do they divide badly or even stumble over the question of how to set the baseline against which to measure disparate impact.

Moreover, and crucially, even if the baseline question can sometimes be difficult, one aspect of it is remarkably simple: It is never—not in a single case ever—simply whatever the status quo ante happened to be in virtue of it being the status quo ante. The principle we derived from Palmer is thus consistent with and gains considerable force from statutory antidiscrimination law.

C. Exceptions

In some contexts, the status quo ante is the appropriate baseline for measuring constitutional and other legal wrongs. Property rights provide a useful comparison. Whether a government regulation of real property that forbids “all economically beneficial use of land” amounts to a taking for which the Constitution requires the payment of just compensation depends on whether the regulation is new (in which case it is a taking) or already “inhere[s] in the title itself” and thus merely recapitulates “restrictions that background principles of the State’s law of property and nuisance already place upon land ownership” (in which case it is not a taking).^[130] Thus, an identical regulation might be a taking in one state but not another, depending on each state’s prior property and/or nuisance law. The legal status quo ante makes a difference.

Property rights are not unique. Various federal constitutional provisions and doctrines take as their antecedent what the law previously was. A leading scholarly treatment identifies “cases under the Contract, Due Process, and Takings Clauses, and under Article II” as “exemplif[ying] a broad set of situations in which federal constitutional or statutory law operates to protect an entitlement created primarily, if not exclusively, by state law.”^[131]

Notably, however, broad does not mean universal. There is no general requirement that states (or other government entities) maintain the legal status quo as it previously existed. Indeed, the default is the opposite. Basic principles of democratic self-government entail legislative (and other governmental) freedom to change the law, at least prospectively.

Equal protection and statutory antidiscrimination cases fall comfortably within the default. A’s claim that some law or policy unequally burdens A relative to B demands that A and B receive the same treatment, not that A be treated in some way that A previously was treated. Moreover, the Supreme Court’s repeated statements that government has the discretion to choose whether to cure equal protection violations by leveling up or down^[132] further demonstrate that the yardstick for measuring equality is the treatment of a comparator, not some past state of the law.

What about reliance? In various contexts, reasonable reliance on the law as it was counts against changing it, thus entrenching the status quo ante.^[133] Might reliance justify using the status quo ante as the baseline for measuring a disparate impact? Although changes in university admissions policy can frustrate expectations, the answer is no.

Consider our hypothetical GSU’s adoption of the admissions lottery for the 2024–2025 admissions cycle. Many applicants will have worked especially hard in high school or taken SAT or ACT preparatory classes on the assumption that doing so would increase the likelihood that they would be admitted to GSU or another highly selective university. They might therefore feel cheated, perhaps even justifiably so.

However, while that fact could provide a policy reason for GSU to announce changes to its admissions criteria as far in advance as possible, GSU is entitled to balance that accommodation of applicants’ expectations against other policy goals. No one has ever suggested that the Constitution limits such changes simply on the ground that they are changes.

And in fact, universities routinely change their admissions criteria in ways that heretofore have not been thought to violate any legally protected interests of applicants who relied on the prior criteria. During the first year of the COVID-19 pandemic, many selective universities went test-optional,^[134] thereby frustrating expectations (while no doubt leading some applicants to feel a sense of relief). Some universities have continued to change their admissions criteria quite rapidly, giving potential applicants little time to adjust. For example, in May 2020, the University of California (UC) changed its policy to make admission to all UC campuses SAT/ACT-optional for fall 2020 and 2021, test-blind for 2023 and 2024, and subject to a new test thereafter.^[135] Then, in November 2021, the Board abandoned plans for a new test, making admissions test-blind indefinitely.^[136]

No doubt, those and other rapid changes created confusion and consternation for applicants who had prepared for the prior regime. Universities can be fairly criticized on policy grounds for changing their admissions policies without sufficient notice—although, as noted above, the cost of disappointing applicants’ expectations must be balanced against the benefits associated with the new criteria.

Meanwhile, even if disappointed reliance of prospective applicants had a constitutional dimension (and in this example it apparently does not), that would at most call for a delay in any change to a new admissions policy, not to its outright prohibition. Yet that conclusion underscores the fact that the reliance of prospective applicants on a continuation of the universities’ prior admissions policy has nothing to do with their equal protection claim: One does not remedy inequality by delaying the inequality.

Hence, the conclusion reached in the previous subparts remains valid. Although the Constitution and other sources of law sometimes preserve prior law or make it the yardstick for measuring legality, the status quo ante is not the proper baseline for measuring disparate impact. On that point, Palmer and six decades of statutory antidiscrimination law are completely sound.

IV. Implications for Invidious Discrimination

This Article has focused on facially race-neutral means of increasing or maintaining racial diversity. However, the arguments have potential implications for other cases, including, disturbingly, facially race-neutral policies adopted with invidious racially discriminatory purposes. Suppose that a university adopts some facially race-neutral admissions policy that affects the incoming class composition, not because it wishes to increase the proportion of the class that are underrepresented minorities but because its leaders believe that the prior regime admitted too few white students and too many Asian American students, based on pernicious anti-Asian stereotypes.^[137] Do the arguments set forth above validate that sort of policy too?

Not necessarily. This Part explains that: (A) the argument of Part I—rejecting the Davis/SFFA syllogism—clearly does not validate invidious discrimination; (B) on its face, the argument of Part III—rejecting the status quo ante as the baseline for measuring disparate impact—does have that potential implication; but (C) that implication potentially can be avoided by recognizing the real problem with the outcome in Palmer, which is that Jackson’s closure of the swimming pools in fact did have a disparate racial impact as measured against a neutral baseline: It imposed stigmatic harm on Black but not white residents. Subpart IV(D) then addresses and rebuts an objection to the argument of subpart (C): that it has unacceptable remedial consequences, such as a judicial order that Jackson must maintain public swimming pools in perpetuity.

A. Distinguishing Purposes

Part I of this Article argued that, consistent with existing case law, courts could distinguish between, on one hand, facially race-neutral policies with a disparate impact that were adopted for an invidious racial purpose, and, on the other hand, facially race-neutral policies with a disparate impact that were adopted for a race-conscious but non-invidious purpose, such as achieving or maintaining racial diversity. That argument, by its terms, necessarily distinguishes between invidious racially discriminatory purposes and non-invidious race-conscious purposes.

Thus, under the Part I approach, adopting a facially race-neutral policy with the purpose of disadvantaging Asian American applicants relative to white applicants would render the purpose racially invidious, so that if the policy had the intended effect, it would be subject to (and surely fail) strict scrutiny on that basis. By contrast, as discussed above, adopting a facially race-neutral policy with the purpose of increasing or maintaining racial diversity would not be racially invidious, so achievement of its ends would amount only to a disparate racial impact under Washington v. Davis.

The challenge, however, is that, as noted at the end of Part I, the conservative Justices who have endorsed a color-blind approach to equal protection (and to antidiscrimination statutes like Title VI) might reject the distinction between invidious racially discriminatory purposes and non-invidious race-conscious purposes. If they were to do so, the hypothetical race-neutral policy that is adopted for the invidious purpose of disadvantaging Asian American applicants in order to benefit white applicants would be invalid—a result that conforms with what I take to be widely shared expectations of what equality demands—but so would the race-neutral means of achieving diversity discussed throughout this Article, such as percentage plans and lotteries. We would have avoided the unintended consequence of an argument to preserve some means of increasing or maintaining racial diversity at the very substantial cost of having failed to preserve any such means.

B. Bite the Bullet?

The argument laid out in Part III—that lotteries and many other facially race-neutral means of achieving racial diversity do not have a disparate impact when measured against the correct baseline, which is not simply the status quo ante—does have the dreaded implication. In our hypothetical example, so long as Asian American applicants continue to be admitted in proportion to their numbers in the qualified applicant pool (where “qualified” is duly understood in accordance with the caveats discussed above), there is no legally relevant disparate impact, and the policy would satisfy equal protection and statutory antidiscrimination law for that reason. Meanwhile, Palmer itself would also be correct in its outcome.

Well, so what? Would accepting that Palmer is correct not only in its rejection of the status quo ante as the baseline but also in its result really be so bad? In the more than half century since Palmer was decided, the case has rarely been invoked to defend laws or policies based on clearly invidious racial motives—perhaps because, during that same period, it has become socially unacceptable for defendants to admit to them.

Consider that SFFA’s case against Harvard included evidence purporting to show that Harvard evaluated the applications of Asian Americans using pernicious stereotypes.^[138] Harvard contested the allegations as a factual matter.^[139] But suppose that Harvard, or some other university, was to adopt a facially neutral admissions policy for the purpose of admitting more white and fewer Asian Americans. It is extremely unlikely that a defendant university would admit that was its actual purpose and invoke a Palmer-based critique of the status quo ante baseline as its legal justification. Instead, the defendant would contest the allegation of racial bias in the selection of the particular race-neutral criterion. Accordingly, we might conclude that conceding the theoretical availability of the baseline defense does little damage in practice.

Yet that conclusion would overlook how litigation actually works. In a world in which plaintiffs must show an intended disparate impact as measured against a baseline of the qualified applicant pool, a defendant university would make an argument in the alternative. We did not choose this admissions criterion based on racial animus, it would say while pointing to evidence purporting to validate that claim, but even if we had (and we didn’t!), the result would be legally permissible. Because it is possible to make the Palmer-based argument without conceding illicit motive, we would expect defendants to do just that.

Might we still want to bite the bullet? We could weigh the benefits of a Palmer-based argument that preserves the ability of universities to pursue racial diversity through race-neutral means against the cost of validating the result in Palmer itself and allowing the adoption of race-neutral policies for invidious race-based reasons, so long as they do not result in a disparate impact relative to the qualified applicant pool. If the former outweighs the latter, then bullet biting would be acceptable.

However, it would be better still to have our cake and eat it too—to preserve the Palmer-based argument for rejecting the status quo ante baseline without approving Palmer itself or the animus-motivated lottery. How might we do that?

C. The Real Problem in Palmer: Racially Disparate Stigma

We can make clear that reliance on Palmer’s reasoning as a ground for the conclusion that the status quo ante is not the proper baseline for measuring disparate impact is not a full endorsement of the result in Palmer. True, after Jackson closed its public swimming pools, nobody of any race had the opportunity to swim in a public pool in Jackson. However, it does not follow that the closure had the same impact on people of all races. For one thing, as Professor Tribe has observed, “the effect of the pool closing was anything but racially neutral, for it seems clear that more whites than [B]lacks had alternative places to swim in Jackson once the public pools were closed.”^[140]

However, one might object that the disparity in alternative swimming venues resulted from private segregation and thus was not unconstitutional because it was not the product of state action. The strength of that objection is not entirely clear. Just as one might think that percentage plans impermissibly leverage private segregation and thus render the state accountable for their racial impact, so too, one might think that when the Jackson authorities closed the public swimming pools, they were at least knowingly taking actions that disproportionately impacted the city’s Black residents.^[141] And while knowledge is not enough to render a disparate impact unconstitutional—because the contested action must be taken because rather than in spite of its disparate impact—it might be enough to render the authorities accountable for the impact, given their intent to stymie desegregation.^[142]

Yet even if one is persuaded by Professor Tribe’s argument that the racially disparate alternative swimming opportunities should have counted as a constitutionally disparate impact in Palmer (as I am persuaded), that argument does not generalize to all of the wide variety of facially race-neutral selection devices that might be used in university admissions and other contexts. Consider our dreaded hypothetical example in which a university adopts a facially race-neutral admissions criterion for the purpose of reducing Asian American enrollment while increasing white enrollment. Even if Asian American applicants have a great many opportunities at other selective universities that do not act with such invidious intent, surely we would nonetheless want to conclude that Asian American applicants to this university have suffered a disparate impact, notwithstanding the fact that they are admitted in proportion to their numbers in the qualified applicant pool. Professor Tribe’s critique of Palmer (was not intended to and thus) does not generate that result.

All is not lost, however. There is another way in which Jackson’s closure of the swimming pools had a disparate racial impact: It stigmatized Black but not white residents of Jackson. As Professor Randall Kennedy has noted: “Insistence that there be no desegregation at the pool amounted to a public declaration that, in the eyes of officials, it would be degrading to whites to have to share a bathing and swimming facility with blacks.”^[143] The real problem, the disparate impact, arose out of the “social meaning” of closing Jackson’s swimming pools to avoid desegregation.^[144] Doing so was a means by which the city’s white elected officials “express[ed] and perpetuat[ed] the superiority of whites over [B]lacks.”^[145] Surely that expression and that perpetuation had a negative impact on Black Jacksonians that it did not have on white ones.

Is that enough? Can a law or policy be invalid by virtue of its invidious social meaning? Professor Kennedy drew a damning analogy between Palmer and Plessy v. Ferguson.^[146] Moreover, he was surely correct that, just as the Plessy Court was obtuse in describing Jim Crow’s “badge of inferiority” as merely a “construction” “the colored race” chose to put on state-mandated segregation,^[147] so too was the Palmer Court at best insensitive in looking only to the fact that persons of all races were deprived of the opportunity to swim in public pools in Jackson while overlooking stigmatic harm.^[148]

And yet, there are two potentially important differences between Plessy and Palmer. First, Louisiana expressly classified people by race as a condition of entering railway cars, whereas the swimming pool closure did not facially classify anybody.^[149] Second, Louisiana mandated the segregation of passenger rail cars; it did not shut down the railroads in response to an order to desegregate.^[150] Had Jackson maintained segregated pools, that clearly would have been unconstitutional; indeed, the order to desegregate was the impetus for the decision to close the pools.

Accordingly, a skeptic might argue that while segregation in Plessy and the public pool closures in Jackson both imposed stigma, the segregation (judged under current equal protection doctrine) resulted in actionable harm—exclusion of Black passengers from the railway cars designated for white passengers—while the swimming pool closures did not. This skeptical response would trade on broader skepticism of proposals to make a law’s validity ever turn on social meaning.^[151]

Yet even if one accepts that critique—that is, even if one accepts that social meaning alone does not give rise to an actionable injury^[152]—it does not follow that the unequal stigma is not actionable. Consider that even as the Supreme Court has denied general-purpose Article III standing to anyone offended by a government policy, it has recognized that where plaintiffs are “‘personally denied equal treatment’ by the challenged discriminatory conduct,” “they have standing to litigate their claims based on the stigmatizing injury.”^[153] Both the Black residents of Jackson in the actual Palmer case and the Asian American applicants in my hypothetical case are personally impacted by the stigmatizing policy, despite its formal race-neutrality.

To be sure, the language quoted in the last paragraph could be further distinguished. Perhaps “discriminatory conduct” exists only where there is an express racial classification or there is an intent to discriminate and a disparate impact with respect to concrete outcomes beyond stigma.

But if so, that is really only another way of saying that Palmer was rightly decided on its facts—which is the counterintuitive proposition this subpart aims to dispel in order to make reliance on Palmer for its insight about baselines more palatable. Professor Kennedy’s observation that Jackson stigmatized Black but not white residents provides the most persuasive grounds for concluding that Palmer was wrongly decided. It should hardly come as a surprise that there are grounds for resisting that conclusion, given that Palmer has never been repudiated.^[154]

D. Remedial Consequences

Before concluding that we can indeed have our cake and eat it too, we should consider one last objection. If Palmer was wrongly decided on its facts, does that mean that Jackson was obligated to keep its desegregated swimming pools open indefinitely? If not, for how long? One might worry that rejecting the result in Palmer entails rejecting the longstanding proposition that a government actor may remedy an equal protection violation by leveling down^[155]—at least where, as with swimming pools, there is no freestanding obligation to provide the government service in the first place.

The objection from remedial consequences has some force, but we might overcome it for any one or more of three reasons. First, the fact that a party has no obligation to do X as a freestanding matter does not mean that party cannot be ordered to do X as a remedy for violating the law. Traditional equitable principles as expounded, especially in the school desegregation cases, make clear that a court’s remedial power extends beyond ordering lawbreakers to comply with their pre-existing obligations.^[156] As typically formulated, that remedial power runs out only when the violation has been remedied—which need not and often does not mean simply that the action giving rise to the violation has ceased.^[157] Accordingly, the fact that Jackson had no obligation to establish public swimming pools in the first place should not have prevented a court from ordering it to maintain desegregated public pools for some time.

How long? The desegregation cases say until the violation is cured but not longer.^[158] Unsurprisingly, that answer cannot be reduced to a numerical formula.^[159]

Nonetheless, one might think that the problem is qualitatively different in a case like Palmer than in a typical desegregation case. In the latter, the school district must immediately cease operating de jure segregated schools and then take court-ordered measures that go beyond its freestanding obligation until it has rooted out the vestiges of the prior regime.^[160] However, in a case like Palmer, reopening the now-desegregated pools immediately eliminates the vestiges of their closure, so it is not clear what a court must measure before allowing the city discretion to close them again.

Answering that puzzle brings into play a second response to the remedial objection: The issue is in no way unique to a case like Palmer but inheres in all cases in which illicit legislative intent converts an otherwise permissible action into an unlawful one.

Consider a garden-variety case involving a facially race-neutral policy that was adopted for racially discriminatory purposes and has an uncontroversially disparate impact as measured by any reasonable baseline. For concreteness, suppose that a city council sites a locally undesirable land use, such as a sewage treatment plant, in the heart of a minority neighborhood. Suppose further that evidence adduced at trial shows unequivocally that: (a) The city could have legitimately chosen the site based on race-neutral considerations of efficacy, safety, and cost; but (b) racial hostility (as manifested by express statements thereof by crucial decision makers) was in fact a but-for cause of the decision to choose the site. It is blackletter law that the siting decision would be invalid.^[161]

Then what? Is the city forever barred from siting the treatment plant where it previously chose to? What if subsequent analysis by independent experts who lack any racial bias reveals that the site is indeed far superior to any alternative? In practice, these are tricky questions^[162] to which the answers depend as much on how one chooses to construct legislative intent as it does on what one finds in the world.^[163] Answers will also depend critically on who bears the burden of proof. But at least in the abstract it is not difficult to identify the end point: An action that was once forbidden because of illicit racial intent—whether that action is siting a sewage treatment plant, closing the public pools, or with racial animus adopting an admissions policy in which, when compared to the status quo ante, Asian American applicants are disadvantaged relative to white ones—remains forbidden until the evidence shows that the action is now being undertaken despite rather than because of its racial effect.

If readers find that proposition difficult to swallow, their objection is not to the notion that Palmer is wrongly decided on its facts, but to the entire post-Washington v. Davis legal apparatus in which illicit racial intent plus disparate racial impact equals an invidious racial classification subject to and almost certain to fail strict scrutiny.

Third and finally, recall that we are considering an objection to the remedial consequences of concluding that Palmer was wrongly decided on its facts. A court cannot order Jackson to keep the pools open, the objection goes, because we have no way of knowing when the order should expire.

However, injunctive relief is not the only possible remedy. If one worries that a court should not order Jackson to keep its public pools open, other remedial options are available. Indeed, given the traditional rule that equity acts only when there is no adequate remedy at law,^[164] one might think that a damages remedy would be preferred.

Although civil rights lawyers in earlier periods typically made a strategic choice to seek only injunctive and declaratory relief,^[165] there is no reason in principle why persons injured by unconstitutional policies could not sue for damages. Indeed, courts have sometimes adjudicated claims that sought very small damages amounts when a claim for injunctive relief had been mooted, including in important cases challenging affirmative action programs.^[166]

Sauce for the opponents of racial diversity is, or at least should be, sauce for its proponents. Accordingly, even if one thinks that courts face insuperable obstacles in crafting injunctive remedies in cases like Palmer and our hypothetical facially race-neutral admissions policy that aims to disadvantage Asian American applicants relative to the status quo ante, the possibility of other forms of relief—including not only damages but also a declaratory judgment—means that courts can nonetheless hold such policies unlawful.

Conclusion

SFFA does not invalidate facially race-neutral means of advancing racial diversity, but through litigation already pending and additional cases likely to be filed, racial diversity’s critics will press hard towards that result. In pushing back, diversity’s champions can practice ideological jujitsu on two tracks. First, we can deploy the arguments for facially race-neutral means that were formerly advanced by diversity’s opponents as grounds for deeming race-based approaches unnecessary and thus unlawful. Second, we can redeploy Palmer v. Thompson and other landmark cases—that, when decided and on their facts, acted as a brake on civil rights claims—to resist the claim that programs that aim at diversity have a disparate racial impact simply because they result in a change from the status quo ante.

Will it work? That remains to be seen. Ideological jujitsu relies on the same dynamic as Wechsler’s conception of neutral application of legal principles.^[167] It appeals to the equality norm at the heart of a system of precedent: the demand that like cases be treated alike. However, equality is, if not exactly an empty concept, one that invites supplementation with values that will vary depending on the ideological druthers of whoever is doing the supplementing.^[168] Accordingly, jurists who oppose efforts to promote diversity may conclude—whether disingenuously or sincerely—that race-neutral means of achieving diversity are not relevantly like Jackson’s closure of the swimming pools or any other precedents diversity’s champions invoke.

But that risk inheres in all legal scholarship and advocacy that rely on judicial precedent. In a doctrinal mode, all that scholars and lawyers can do is point to the logical and practical consequences of the principles that jurists espouse. If those jurists do not follow where the principles they espouse lead, that is on them.

1. . 347 U.S. 483 (1954). ↑
2. . Herbert Wechsler, Toward Neutral Principles of Constitutional Law, 73 Harv. L. Rev. 1, 34 (1959). ↑
3. . See Charles L. Black, Jr., The Lawfulness of the Segregation Decisions, 69 Yale L.J. 421, 424 (1960). As Professor Black explained:[I]f a whole race of people finds itself confined within a system which is set up and continued for the very purpose of keeping it in an inferior station, and if the question is then solemnly propounded whether such a race is being treated ‘equally,’ I think we ought to exercise one of the sovereign prerogatives of philosophers—that of laughter.

   Id. ↑

4. . See, e.g., Cass R. Sunstein, Neutrality in Constitutional Law (with Special Reference to Pornography, Abortion, and Surrogacy), 92 Colum. L. Rev. 1, 50–52 (1992) (arguing that neutrality in the sense of valueless legal judgments is not an attainable aspiration, and one we are “better off without,” but “is unobjectionable insofar as it is a call for internal consistency”). ↑
5. . Compare, e.g., Parents Involved in Cmty. Schs. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 746–47 (2007) (“What do the racial classifications at issue here do, if not accord differential treatment on the basis of race?”); Students for Fair Admissions, Inc. v. President & Fellows of Harvard Coll., 143 S. Ct. 2141, 2160–62, 2175 (2023) (invoking Brown to argue that “[e]liminating racial discrimination means eliminating all of it”); Fisher v. Univ. of Tex. at Austin, 570 U.S. 297, 326–27 (2013) (Thomas, J., concurring) (quoting the plaintiffs’ arguments in Brown to support the claim that the Constitution is color-blind), with Parents Involved, 551 U.S. at 799 (Stevens, J., dissenting) (“The Chief Justice rewrites the history of one of this Court’s most important decisions.”); id. at 803 (Breyer, J., dissenting) (arguing that, contrary to the majority’s characterization, the challenged “plans represent local efforts to bring about the kind of racially integrated education that Brown . . . long ago promised”). ↑
6. . 163 U.S. 537 (1896). ↑
7. . Id. at 559 (Harlan, J., dissenting) (“[I]n view of the Constitution, in the eye of the law, there is in this country no superior, dominant, ruling class of citizens. There is no caste here.”). ↑
8. . See, e.g., Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 407 (1978) (Blackmun, J., writing separately) (“In order to get beyond racism, we must first take account of race. There is no other way. And in order to treat some persons equally, we must treat them differently. We cannot—we dare not—let the Equal Protection Clause perpetuate racial supremacy.”); Parents Involved, 551 U.S. at 799, 803 (Stevens, J., dissenting) (stating that the majority’s reliance on Brown in striking down racial classifications rewrote that case’s history and that “no Member of the Court that [Justice Stevens] joined in 1975 would have agreed with today’s decision”); Adarand Constructors, Inc. v. Peña, 515 U.S. 200, 243, 245 (1995) (Stevens, J., dissenting) (arguing that the Court’s “concept of ‘consistency’” assumes “no significant difference between a decision by the majority to impose a special burden on the members of a minority race and a decision by the majority to provide a benefit to certain members of that minority notwithstanding its incidental burden on some members of the majority” and that such logic “would disregard the difference between a ‘No Trespassing’ sign and a welcome mat”). ↑
9. . See City of Richmond v. J. A. Croson Co, 488 U.S. 469, 493 (1989) (stating that “the purpose of strict scrutiny is to ‘smoke out’ illegitimate uses of race by assuring that the legislative body is pursuing a goal important enough to warrant use of a highly suspect tool”). ↑
10. . See, e.g., Students for Fair Admissions, 143 S. Ct. at 2189, 2202 (Thomas, J., concurring) (arguing that because all racial groups are heterogenous in other significant respects, “all racial categories are little more than stereotypes, suggesting that immutable characteristics somehow conclusively determine a person’s ideology, beliefs, and abilities”); Adarand Constructors, 515 U.S. at 241 (Thomas, J., concurring in part and concurring in judgment). According to Justice Thomas:So-called “benign” discrimination teaches many that because of chronic and apparently immutable handicaps, minorities cannot compete with them without their patronizing indulgence. . . . These programs stamp minorities with a badge of inferiority and may cause them to develop dependencies or to adopt an attitude that they are ‘entitled’ to preferences.

    Id.; cf. Richard Sander & Stuart Taylor, Jr., Mismatch: How Affirmative Action Hurts Students It’s Intended to Help, and Why Universities Won’t Admit It 3–4 (2012) (arguing that affirmative action admissions policies admit minority students who are unprepared for the academic standards of the admitting institution, leading to disproportionate difficulties for those students). ↑

11. . See, e.g., Stephen Kershnar, Justice for the Past 51 (2004) (arguing that mere benefit “from an unjust injuring act or being a member of a community that owes a debt of compensation to racial minorities . . . is not a sufficient ground to override the duty owed to the white male”). ↑
12. . See Students for Fair Admissions, 143 S. Ct. at 2225–29 (Sotomayor, J., dissenting) (arguing that the history of the Fourteenth Amendment, the Freedman’s Bureau Act, the Civil Rights Act of 1866, and appropriations bills of the 1860s indicate that the Reconstruction Congress was not color-blind, but instead purposefully pursued race-conscious legislation); id. at 2264–65 (Jackson, J., dissenting) (contending that the Reconstruction amendments and acts were specifically meant to benefit Black Americans). ↑
13. . Prohibition Against Discrimination or Preferential Treatment by State or Other Public Entities, 1996 Cal. Stat. A-294 (codified as Cal. Const. art. I, § 31). Also known as Proposition 209, California voters passed the measure with about 55% of the vote. Bill Jones, Cal. Sec’y of State, Statement of Vote Nov. 5, 1996, at xii–xiii (1996). That same year, California voters also adopted relatively liberal ballot initiatives increasing the minimum wage, permitting medical marijuana use, and limiting campaign contributions and spending. Id. at xii. ↑
14. . Resolution Chapter 23, 2020 Cal. Stat. A-1 (repealing Cal. Const. art. I, § 31) (defeated as Proposition 16 on Nov. 3, 2020). Fifty-seven percent of eligible Californians voted no on this measure that would have repealed the state constitutional language adopted by Proposition 209. Alex Padilla, Cal. Sec’y of State, Statement of Vote Nov. 3, 2020, at 67 (2020). That same year, about 64% of Californians voted for Democrat Joseph R. Biden for President. Id. at 8. ↑
15. . Michigan Civil Rights Initiative, 2006 Mich. Pub. Acts 2515 (codified as Mich. Const. art. I, § 26). Michigan voters backed this measure by a 58 to 42% margin. Jocelyn Benson, Mich. Sec’y of State, 2006 Michigan Election Results, https://mielections.us/election
    /results/06GEN/ [https://perma.cc/U7PE-CY4D]. Compare to the 56 to 42% margin by which they re-elected Democratic Governor Jennifer Granholm the same year. Id. ↑
16. . 143 S. Ct. 2141 (2023). ↑
17. . Id. at 2175–76. For a brief overview of historical resistance to Brown and the view that the decision had no direct effect in ending discrimination in public schools, especially in the American South, see generally Gerald N. Rosenberg, The Hollow Hope: Can Courts Bring About Social Change? (1991). See also James T. Patterson, Brown v. Board of Education: A Civil Rights Milestone and Its Troubled Legacy 94–99 (2001) (detailing the history of resistance and subterfuge in response to Brown). ↑
18. . See Jesse H. Choper, Consequences of Supreme Court Decisions Upholding Individual Constitutional Rights, 83 Mich. L. Rev. 1, 78 & nn.523–24 (1984) (noting that the Court’s school prayer decisions initially generated great “political hostility and popular noncompliance”). ↑
19. . Race-based affirmative action has heretofore been practiced in undergraduate admissions at colleges and universities, see Students for Fair Admissions, 143 S. Ct. at 2155 (2023) and Gratz v. Bollinger, 539 U.S. 244, 253–54 (2003), as well as in graduate and professional programs, see Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 274–76 (1978) and Grutter v. Bollinger, 539 U.S. 306, 316 (2003). For simplicity, this Article will refer to all such institutions as “universities,” a term that is not meant to exclude exclusively undergraduate institutions or standalone graduate and professional schools. ↑
20. . Students for Fair Admissions, 143 S. Ct. at 2166–67. ↑
21. . Id. at 2176. ↑
22. . Id. ↑
23. . See Karl Marx, The Eighteenth Brumaire of Louis Bonaparte 9 (Daniel de Leon trans., Charles H. Kerr & Co. 3d ed. 1919) (1852) (“Hegel says somewhere that all great historic facts and personages recur twice. He forgot to add: ‘Once as tragedy, and again as farce.’”). ↑
24. . For a description of the Texas plan, see Fisher v. Univ. of Tex. at Austin, 570 U.S. 297, 305 (2013) (“The Texas State Legislature . . . enacted a measure known as the Top Ten Percent Law,” granting “automatic admission to any public state college . . . to all students in the top 10% of their class at high schools in Texas that comply with certain standards.”). Likewise, long before SFFA, California and Florida also adopted percentage plans in response to, or in anticipation of limits on, affirmative action. See Cheryl I. Harris, Mining in Hard Ground, 116 Harv. L. Rev. 2487, 2519 n.57 (2003) (reviewing Lani Guinier & Gerald Torres, The Miner’s Canary (2002)) (describing California’s “Eligibility in Local Context” plan and Florida’s “Talented Twenty Plan”). Versions of both programs remain in effect. Statewide Guarantee, Univ. of Cal. Admissions, https://admission.universityofcalifornia.edu/admission-requirements/freshman-requirements/california-residents/statewide-guarantee/ [https://perma.cc/354J-UEJS]; Talented Twenty Program, Fla. Dep’t of Educ., https://www.fldoe.org/schools/family-community/activities-programs/pre-collegiate/talented-twenty-program/ [https://perma.cc/A5DG-DCNU]. Less than a year after the ruling in SFFA, Wisconsin enacted a percentage plan. See 2023 Wisconsin Act 95, 106th Wis. Leg., Reg. Sess. (codified in scattered sections of Wis. Code §§ 36, 118, 119), https://legiscan.com/WI/text/SB367/id/2936231 [https://perma.cc/Y29V-HYYN] (guaranteeing admission to Wisconsin universities to graduates in the top ten percent of their Wisconsin high school classes but limiting guaranteed admission to the University of Wisconsin-Madison to the top five percent). ↑
25. . E.g., Brief for the United States as Amicus Curiae Supporting Petitioners at 17–18, Gratz v. Bollinger, 539 U.S. 244 (2003) (No. 02-516) (arguing that Texas’s percentage enrollment plan is an example of a race-neutral alternative that precludes the use of race in admissions because it maintained, and in some cases increased, minority enrollment); Brief for the CATO Institute as Amicus Curiae in Support of Petitioner at 32, Fisher, 579 U.S. 365 (2016) (No. 14-981) (arguing that the Texas percentage program is race-neutral and that “the operation of that law ensures broad diversity in every dimension”). ↑
26. . See, e.g., Gratz, 539 U.S. at 304 n.10 (Ginsburg, J., dissenting). As Justice Ginsburg explained:Percentage plans depend for their effectiveness on continued racial segregation at the secondary school level: They can ensure significant minority enrollment in universities only if the majority-minority high school population is large enough to guarantee that, in many schools, most of the students in the top 10% or 20% are minorities.

    Id.; Fisher, 570 U.S. at 335 (Ginsburg, J., dissenting) (Texas’[s] percentage plan was adopted with racially segregated neighborhoods and schools front and center stage. . . . It is race consciousness, not blindness to race, that drives such plans.”). ↑

27. . See, e.g., Appellee’s Response Brief at 33–34, 36, 58, Coal. for TJ v. Fairfax Cnty. Sch. Bd., 68 F.4th 864 (4th Cir. 2023) (No. 22-1280) (arguing that a facially race-neutral admissions plan for a public magnet high school was in effect race conscious based on alleged evidence of disparate impact and racially discriminatory intent, thus triggering strict scrutiny). For thoughtful discussions of this litigation and similar cases, as well as their broader implications, see generally Jonathan D. Glater, Reflections on Selectivity, 49 Fordham Urb. L.J. 1121 (2022) and Sonja Starr, The Magnet School Wars and the Future of Colorblindness, 76 Stan. L. Rev. 161 (2024). ↑
28. . Compare Opening Brief of Appellant at 21, Coal. for TJ., 68 F.4th 864 (No. 22-1280) (arguing that a facially race-neutral admissions policy, focusing on family income and underrepresented middle schools, whose designers hoped would increase racial as well as other forms of diversity, did not amount to impermissible racial balancing), with Brief for the United States as Amicus Curiae Supporting Petitioners at 18–19, Gratz, 539 U.S. 244 (No. 02-516) (contending that “if the University genuinely seeks candidates with diverse experiences and viewpoints, it can focus on numerous race-neutral factors”). ↑
29. . See, e.g., Brief of Amicus Curiae Association of American Law Schools in Support of Respondents at 14, Grutter v. Bollinger, 539 U.S. 306 (2003) (No. 02-241) (contending, in a brief co-authored by this author, that race-neutral admissions mechanisms are not in fact race-neutral and would require sacrificing other critically important law school interests). ↑
30. . Starr, supra note 27. ↑
31. . Id. at 180–81. ↑
32. . See id. at 170 n.32 (citing Elise C. Boddie, The Constitutionality of Racially Integrative Purpose, 38 Cardozo L. Rev. 531, 546 (2016) (“arguing that racial integration is an objective permitted by constitutional doctrine”)); Samuel R. Bagenstos, Disparate Impact and the Role of Classification and Motivation in Equal Protection Law After Inclusive Communities, 101 Cornell L. Rev. 1115, 1156–57 (2016); Kim Forde-Mazrui, The Canary-Blind Constitution: Must Government Ignore Racial Inequality?, 79 Law & Contemp. Probs. 53, 55–57 (2016); Reva B. Siegel, Race-Conscious but Race-Neutral: The Constitutionality of Disparate Impact in the Roberts Court, 66 Ala. L. Rev. 653, 655–56 (2015); Reva B. Siegel, From Colorblindness to Antibalkanization: An Emerging Ground of Decision in Race Equality Cases, 120 Yale L.J. 1278, 1351–52 (2011); Michelle Adams, Is Integration a Discriminatory Purpose?, 96 Iowa L. Rev. 837, 843 (2011); Richard Primus, The Future of Disparate Impact, 108 Mich. L. Rev. 1341, 1347, 1386 (2010); Kimberly Jenkins Robinson, The Constitutional Future of Race-Neutral Efforts to Achieve Diversity and Avoid Racial Isolation in Elementary and Secondary Schools, 50 B.C. L. Rev. 277, 283 (2009); Andrew M. Carlon, Racial Adjudication, 2007 BYU L. Rev. 1151, 1155–57; R. Richard Banks, The Benign-Invidious Asymmetry in Equal Protection Analysis, 31 Hastings Const. L.Q. 573, 574–75 (2003); Richard A. Primus, Equal Protection and Disparate Impact: Round Three, 117 Harv. L. Rev. 493, 495, 500 (2003); Kim Forde-Mazrui, The Constitutional Implications of Race-Neutral Affirmative Action, 88 Geo. L.J. 2331, 2334–37 (2000). The fact that so many liberal scholars defended ends-colorblindness while some race-based affirmative action was still considered lawful might be thought to undercut my suggestion above that the ideological valence of claims regarding facially race-neutral programs that aim to increase or maintain racial diversity flipped post-SFFA. It does not. We might better understand the liberal scholars’ defense of race-neutral means as offered in astute anticipation of an increasingly conservative Court’s invalidation of nearly all facially race-based affirmative action. ↑
33. . See Deborah Hellman, Diversity by Facially Neutral Means, 110 Va. L. Rev. (forthcoming 2024) (manuscript at 25), https://ssrn.com/abstract=4742118 [https://perma.cc/44XK-HZF8] (“[T]he only relevant intention is the intent to harm.”). I describe Professor Hellman’s article as similar to the work of Professor Starr and the other scholars they each cite despite the fact that Professor Hellman to some extent criticizes Professor Starr and those other scholars and distinguishes her view from theirs. Id. at 19–24. Despite important differences of nuance, all of these scholars propose that equal protection doctrine should treat race-neutral means of achieving racial diversity differently from race-neutral means of intentionally disadvantaging members of an underrepresented racial minority group. ↑
34. . 403 U.S. 217 (1971). For a fairly recent argument that Palmer should be relegated to the anti-canon of repudiated white supremacist decisions such as Dred Scott v. Sandford, 60 U.S. (19 How.) 393 (1857) (enslaved party), superseded by constitutional amendment, U.S. Const. amend. XIV, Plessy v. Ferguson, 163 U.S. 537 (1896), and Korematsu v. United States, 323 U.S. 214 (1944), see Randall Kennedy, Reconsidering Palmer v. Thompson, 2018 Sup. Ct. Rev. 179, 179, 199–200. As Part IV explains, it is possible to credit Professor Kennedy’s critique of Palmer and conclude that the case was wrongly decided on its facts without discrediting the aspect of its holding on which this Article relies. ↑
35. . Palmer, 403 U.S. at 219. ↑
36. . Id. ↑
37. . See id. at 224 (“[N]o case in this Court has held that a legislative act may violate equal protection solely because of the motivations of the men who voted for it.”). ↑
38. . 426 U.S. 229, 244 n.11 (1976). Rejecting one rationale for Palmer, the Court explained:To the extent that Palmer suggests a generally applicable proposition that legislative purpose is irrelevant in constitutional adjudication, our prior cases—as indicated in the text—are to the contrary; and very shortly after Palmer, all Members of the Court majority in that case joined the Court’s opinion in Lemon v. Kurtzman, 403 U.S. 602 (1971), which dealt with the issue of public financing for private schools and which announced, as the Court had several times before, that the validity of public aid to church-related schools includes close inquiry into the purpose of the challenged statute.

    Id.; see also Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 265 (1977) (“Proof of racially discriminatory intent or purpose is required to show a violation of the Equal Protection Clause.”); Hunter v. Underwood, 471 U.S. 222, 232 (1985) (rejecting appellants’ reliance on Palmer on the ground that “where both impermissible racial motivation and racially discriminatory impact are demonstrated, Arlington Heights and Mt. Healthy [City Sch. Dist. Bd. of Educ. v. Doyle, 429 U.S. 274 (1977)] supply the proper analysis”). ↑

39. . See Kennedy, supra note , at 180 (acknowledging that “the Court has never rejected Palmer’s holding, which remains ‘good law’”). ↑
40. . Supreme Court precedent permits government actors to cure an equal protection violation by leveling down or up. See, e.g., Sessions v. Morales-Santana, 137 S. Ct. 1678, 1698 n.23 (2017) (“Because the manner in which a State eliminates discrimination ‘is an issue of state law,’ upon finding state statutes constitutionally infirm, we have generally remanded to permit state courts to choose between extension and invalidation.” (citation omitted) (quoting Stanton v. Stanton, 421 U.S. 7, 18 (1975))); Orr v. Orr, 440 U.S. 268, 272 (1979) (“In every equal protection attack upon a statute challenged as underinclusive, the State may satisfy the Constitution’s commands either by extending benefits to the previously disfavored class or by denying benefits to both parties . . . .”). ↑
41. . It is too early to tell how widespread this pattern will prove to be, but it is representative of at least some highly selective institutions. During the first post-SFFA undergraduate admissions cycle, Amherst College, Brown University, and the Massachusetts Institute of Technology experienced substantial declines in enrollment of Black and Latinx students, while Duke University, the University of Virginia, and Yale University did not. John S. Rosenberg, Admissions After Affirmative Action, Harv. Mag. (Sept. 9, 2024), https://www.harvardmagazine.com/2024
    /09/admissions-after-affirmative-action [https://perma.cc/PMK9-GTVZ]. ↑
42. . Coal. for TJ v. Fairfax Cnty. Sch. Bd., 68 F.4th 864, 880 (4th Cir. 2023) (quoting Coal. for TJ v. Fairfax Cnty. Sch. Bd., No. 22-1280, 2022 WL 986994 at *3 (4th Cir. Mar. 31, 2022) (Heytens, J., concurring)), cert. denied, No. 23-170, 2024 WL 674659 (Feb. 20, 2024). In his dissent from the denial of certiorari, Justice Alito, joined by Justice Thomas, described the Fourth Circuit decision as holding “that intentional racial discrimination is constitutional so long as it is not too severe.” Coal. for TJ, 2024 WL 674659, at *1 (Alito, J., dissenting). As I explain below, that is a mischaracterization: If there is no disparate impact relative to the correct baseline, there is no discrimination at all, severe or otherwise. ↑
43. . 429 U.S. 252 (1977). ↑
44. . Coal. for TJ, 68 F.4th at 880 (quoting Arlington Heights, 429 U.S. at 266). ↑
45. . See Arlington Heights, 429 U.S. at 269 (“The impact of the Village’s decision does arguably bear more heavily on racial minorities. Minorities constitute 18% of the Chicago area population, and 40% of the income groups said to be eligible for Lincoln Green.”). ↑
46. . Id. at 270 (“Respondents simply failed to carry their burden of proving that discriminatory purpose was a motivating factor in the Village’s decision.”). ↑
47. . See infra subpart III(B). ↑
48. . Students for Fair Admissions, Inc. v. President & Fellows of Harvard Coll., 143 S. Ct. 2141, 2156 n.2 (2023) (“‘We have explained that discrimination that violates the Equal Protection Clause of the Fourteenth Amendment committed by an institution that accepts federal funds also constitutes a violation of Title VI.’” (quoting Gratz v. Bollinger, 539 U.S. 244, 276 n.23 (2003))). To be sure, the Court added that no party had asked it to re-examine the equivalence between
    equal protection and Title VI, id. at 2156 n.2, thereby perhaps implying that it might do so if it were asked. But even if it were to do so, statutory cases would likely remain an important source of guidance for constitutional ones. ↑
49. . See sources cited supra note . ↑
50. . See, e.g., Washington v. Davis, 426 U.S. 229, 240 (1976) (“The school desegregation cases have also adhered to the basic equal protection principle that the invidious quality of a law claimed to be racially discriminatory must ultimately be traced to a racially discriminatory purpose.”); Arlington Heights, 429 U.S. at 265–66 (“[R]acial discrimination is not just another competing consideration. When there is a proof that a discriminatory purpose has been a motivating factor in the decision, this judicial deference [to a legislative body’s decision-making] is no longer justified.”). ↑
51. . See, e.g., Students for Fair Admissions, 143 S. Ct. at 2162 (stating that all uses of race “must survive a daunting two-step examination known in our cases as ‘strict scrutiny’” (quoting Adarand Constructors, Inc. v. Peña, 515 U.S. 200, 227 (1995))); City of Richmond v. J.A. Croson Co., 488 U.S. 469, 495–96, 500 (1989) (holding that all racial classifications are suspect and liable to strict scrutiny regardless of whom they advantage or disadvantage or the “benign” purposes of the legislature). ↑
52. . For example, the George W. Bush administration touted percentage plans as an effective race-neutral approach to diversity after Grutter. See Off. for C.R., U.S. Dep’t of Educ., Achieving Diversity: Race Neutral Alternatives in American Education (2004), https://files.eric.ed.gov/fulltext/ED486351.pdf [https://perma.cc/P6Y2-FTEV] (calling percentage plans the “best-known recent development in race-neutral approaches to diversity”); see also Brief for the United States as Amicus Curiae Supporting Petitioners at 14, 18, Gratz, 539 U.S. 244 (2003) (No. 02-516) (arguing that Texas’s percentage enrollment plan is an example of a race-neutral alternative that precludes the use of race in admissions because it maintained, and in some cases increased, minority enrollment); Brief for the CATO Institute as Amicus Curiae in Support of Petitioner at 32, Fisher v. Univ. of Tex. at Austin, 579 U.S. 365 (2016) (No. 14-981) (arguing that the Texas Top Ten Percent Law is race-neutral and that “the operation of that law ensures broad diversity in every dimension”); Brief for the State of Florida and the Honorable John Ellis “Jeb” Bush, Governor, as Amici Curiae in Support of Petitioners at 7–8, Grutter v. Bollinger, 539 U.S. 306 (2003) (Nos. 02-241, 02-516) (trumpeting Florida’s Talented Twenty Program, which offered automatic admission to a state school for students graduating in the top twenty percent of their class, as a race-neutral alternative obviating the need for explicit consideration of race to achieve diversity). ↑
53. . 539 U.S. 244 (2003). ↑
54. . Id. at 298 (Souter, J., dissenting). ↑
55. . Id. at 297. ↑
56. . See Fisher v. Univ. of Tex. at Austin, 570 U.S. 297, 303 (2013) (holding that the court below had applied an insufficiently rigorous version of strict scrutiny to the university’s supplemental race-conscious admissions program); Fisher v. Univ. of Tex. at Austin, 579 U.S. 365, 377, 388 (2016) (finding that the university’s “sui generis” supplemental race-conscious admissions program satisfied strict scrutiny). ↑
57. . See Neil H. Buchanan & Michael C. Dorf, How to Choose the Least Unconstitutional Option: Lessons for the President (and Others) from the Debt Ceiling Standoff, 112 Colum. L. Rev. 1175, 1219 (2012) (articulating three criteria for making a choice when only unconstitutional options are available: “minimize the unconstitutional assumption of power; minimize sub-constitutional harm; and preserve, to the extent possible . . . the ability . . . to undo or remedy constitutional violations”). ↑
58. . The cases cited supra note 42 were first filed shortly before SFFA but after the Supreme Court’s composition had changed sufficiently to portend its result. ↑
59. . Washington v. Davis, 426 U.S. 229, 245–46 (1976). ↑
60. . Id. at 239. ↑
61. . Id. at 236, 239–43. ↑
62. . See Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266 (1977) (examining whether “invidious discriminatory purpose” was a motiving factor); Pers. Adm’r of Mass. v. Feeney, 442 U.S. 256, 274 (1979) (evaluating whether statute’s adverse effect reflected “invidious gender-based discrimination”). ↑
63. . See Students for Fair Admissions, Inc. v. President & Fellows of Harvard Coll., 143 S. Ct. 2141, 2175 (2023) (invalidating challenged admissions programs because they “lack sufficiently focused and measurable objectives warranting the use of race, unavoidably employ race in a negative manner, involve racial stereotyping, and lack meaningful end points”). ↑
64. . See United States v. Windsor, 570 U.S. 744, 770 (2013) (invalidating the federal Defense of Marriage Act on the ground that it was motivated by “improper animus”); Romer v. Evans, 517 U.S. 620, 632 (1996) (finding that a state constitutional amendment banning anti-discrimination ordinances on the basis of sexual orientation was so discontinuous with the reasons offered that it could only be explained by “animus” against the class it affected); see also U.S. Dep’t of Agric. v. Moreno, 413 U.S. 528, 534 (1973) (“[I]f the constitutional conception of ‘equal protection of the laws’ means anything, it must at the very least mean that a bare congressional desire to harm a politically unpopular group cannot constitute a legitimate governmental interest.”). ↑
65. . 488 U.S. 469 (1989). ↑
66. . Id. at 493 (plurality opinion). ↑
67. . Gerald Gunther, Foreword: In Search of Evolving Doctrine on a Changing Court: A Model for a Newer Equal Protection, 86 Harv. L. Rev. 1, 8 (1972) (describing the Warren Court’s doctrine as “‘strict’ in theory and fatal in fact”). ↑
68. . Grutter v. Bollinger, 539 U.S. 306, 326 (2003). ↑
69. . Id. at 310. ↑
70. . Id. at 310, 378. ↑
71. . This Article need not and does not take a position on whether SFFA overruled Grutter or merely declared that the time allowed by Grutter for temporary measures had expired. Compare Students for Fair Admissions v. President & Fellows of Harvard Coll., 143 S. Ct. 2141, 2207 (2023) (Thomas, J., concurring) (“The Court’s opinion rightly makes clear that Grutter is, for all intents and purposes, overruled.”), with id. at 2224 (Kavanaugh, J., concurring) (“[T]he Court’s decision today appropriately respects and abides by Grutter’s explicit temporal limit on the use of race-based affirmative action in higher education.”). ↑
72. . See, e.g., Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 298 (1978) (plurality opinion) (“[P]referential programs may only reinforce common stereotypes holding that certain groups are unable to achieve success without special protection based on a factor having no relationship to individual worth.”); City of Richmond v. J.A. Croson Co., 488 U.S. 469, 493 (1989) (plurality opinion) (“Classifications based on race carry a danger of stigmatic harm. Unless they are strictly reserved for remedial settings, they may in fact promote notions of racial inferiority and lead to a politics of racial hostility.”); Grutter, 539 U.S. at 353 (Thomas, J., concurring in part and dissenting in part) (“The Constitution abhors classifications based on race, not only because those classifications can harm favored races or are based on illegitimate motives, but also because every time the government places citizens on racial registers and makes race relevant to the provision of burdens or benefits, it demeans us all.”); Adarand Constructors, Inc. v. Peña, 515 U.S. 200, 214 (1995) (“‘[D]istinctions between citizens solely because of their ancestry are by their very nature odious to a free people whose institutions are founded upon the doctrine of equality . . . .”’ (quoting Hirabayashi v. United States, 320 U.S. 81, 100 (1943))); Rice v. Cayetano, 528 U.S. 495, 517 (2000) (“One of the principal reasons race is treated as a forbidden classification is that it demeans the dignity and worth of a person to be judged by ancestry instead of by his or her own merit and essential qualities.”). ↑
73. . The examples discussed in this Part do not exhaust all of the possibilities. For example, a university might aim at forms of diversity that correlate with race by practicing class-based or geographical affirmative action. ↑
74. . See Students for Fair Admissions, 143 S. Ct. at 2176 (“[N]othing in this opinion should be construed as prohibiting universities from considering an applicant’s discussion of how race affected his or her life . . . .”). ↑
75. . Id. ↑
76. . See id. (“[U]niversities may not simply establish through application essays or other means the regime we hold unlawful today.”) ↑
77. . For an argument that grit reflects perseverance and predicts achievement, see generally Angela Duckworth, Grit: The Power of Passion and Perseverance (2016) and Paul Tough, How Children Succeed: Grit, Curiosity, and the Hidden Power of Character (2012). Not everyone is persuaded. For an example of recent scholarly work opposing evaluations of “grit” in the college admissions process, see generally Debunking the Grit Narrative in Higher Education: Drawing on the Strengths of African American, Asian American, Pacific Islander, Latinx, and Native American Students (Angela M. Locks, Rocío Mendoza & Deborah Faye Carter, eds. 2023). ↑
78. . Indeed, it can be argued that the distinction between race and experience with race is illusory. See Benjamin Eidelson and Deborah Hellman, Unreflective Disequilibrium: Race-Conscious Admissions After SFFA, 4 Am. J.L. & Equal. 295, 298 (2024) (“Often, understanding how a person’s race has affected them, or how it figures in their present self-conception, requires also appreciating that it is their race—or even their being Black, being white, or the like—rather than something else that is playing this role.”). ↑
79. . This Article is not legal advice! ↑
80. . Act of May 20, 1997, ch. 155, § 51.803, 75th Leg., Reg. Sess., 1997 Tex. Gen. Laws 304, 304, 306 (codified at Tex. Educ. Code Ann. § 51.803) (requiring automatic undergraduate admission at all state institutions of higher education for applicants who graduated with a grade point average in the top ten percent of the applicant’s high school graduating class). ↑
81. . See Hopwood v. Texas, 78 F.3d 932, 944 (5th Cir. 1996) (holding that approval of race as a plus factor by Justice Lewis Powell in his controlling opinion in Bakke never was the law and had, in any event, been superseded by cases mandating color-blindness). ↑
82. . See Gratz v. Bollinger, 539 U.S. 244, 297–98 (2003) (Souter, J., dissenting) (arguing that percentage plans like Texas’s allow states to achieve “racially diverse results without saying directly what they are doing or why they are doing it”); Grutter v. Bollinger, 539 U.S. 306, 340 (2003) (stating that percentage plans “may preclude the university from conducting the individualized assessments necessary to assemble a student body that is not just racially diverse, but diverse along all the qualities valued by the university”). ↑
83. . Fisher v. Univ. of Tex. at Austin, 570 U.S. 297, 306, 314 (2013) (holding that the appeals court had applied an insufficiently rigorous version of strict scrutiny to the university’s supplemental race-conscious admissions program). ↑
84. . Fisher v. Univ. of Tex. at Austin, 579 U.S. 365, 369, 388 (2016) (finding that the consideration of race in the holistic admissions plan at the University of Texas at Austin (for candidates not qualifying under the percentage plan) did not violate equal protection). ↑
85. . See Halley Potter, School Segregation in U.S. Metro Areas, Century Found. (May 17, 2022), https://tcf.org/content/report/school-segregation-in-u-s-metro-areas/ [https://perma.cc
    /HV63-3W8T] (charting levels of school segregation by region and type); Sean F. Reardon & Ann Owens, 60 Years After Brown: Trends and Consequences of School Segregation, 40 Ann. Rev. Socio. 199, 203 (2014) (showing nationwide trends in school segregation, especially in the South where black–white segregation returned to pre-1968 levels in the 1980s and has been relatively stable since); Timothy Mattison, The State of School Segregation in Texas and the Factors Associated with It, Tex. Educ. Rev., Spring 2020, at 18, 28–29 (finding evidence that segregation of Black and Latinx students still exists across Texas public and public charter schools). ↑
86. . See Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 315–16 (1978) (plurality opinion) (finding that multitrack admissions programs “with a prescribed number of seats set aside for each identifiable category of applicants” were not a necessary means to the end of genuine educational diversity). ↑
87. . See Freeman v. Pitts, 503 U.S. 467, 494 (1992) (“Once the racial imbalance due to the de jure violation has been remedied, the school district is under no duty to remedy imbalance that is caused by demographic factors.”); see also Everett v. Pitt Cnty. Bd. of Educ., 788 F.3d 132, 145–47 (4th Cir. 2015) (affirming a district court finding that a school board had “eliminated the vestiges of its past discrimination” and had no subsequent duty to implement desegregation efforts to mitigate the effect of demographic shifts). ↑
88. . 466 U.S. 429 (1984). ↑
89. . Id. at 430–31. ↑
90. . Id. at 433. ↑
91. . 539 U.S. 306 (2003). ↑
92. . Id. at 361–62 (Thomas, J., concurring in part and dissenting in part). ↑
93. . See Pers. Adm’r of Mass. v. Feeney, 442 U.S. 256, 279 (1979) (“‘Discriminatory purpose,’ however, implies more than intent as volition or intent as awareness of consequences. It implies that the decisionmaker . . . selected or reaffirmed a particular course of action at least in part ‘because of,’ not merely ‘in spite of,’ its adverse effects upon an identifiable group.” (citation omitted)). ↑
94. . See Institute of Education Sciences, Characteristics of Degree-Granting Postsecondary Institutions 2 (Aug. 2023), https://nces.ed.gov/programs/coe/indicator/csa/postsecondary-institutions [https://perma.cc/TP8R-M8PR] (showing that 92% of two-year institutions and 25% of four-year institutions had open admissions policies for fall 2021). ↑
95. . Grutter, 539 U.S. at 358–61 (Thomas, J., concurring in part and dissenting in part). ↑
96. . See id. at 340 (majority opinion) (rejecting the district court’s view that University of Michigan Law School should have considered an admissions lottery). ↑
97. . From 1972 until 2017, the Netherlands used a lottery for medical school admissions; that practice was recently reintroduced. See Olle ten Cate, Rationales for a Lottery Among the Qualified to Select Medical Trainees: Decades of Dutch Experience, 13 J. Graduate Med. Educ. 612, 612 (2021). ↑
98. . Laurence H. Tribe, The Mystery of Motive, Private and Public: Some Notes Inspired by the Problems of Hate Crime and Animal Sacrifice, 1993 Sup. Ct. Rev. 1, 23 (1993). ↑
99. . The effect only operates on average. Many Asian American and white applicants with high school grades and test scores only barely above the minimum threshold have a better chance of admission under the lottery than they would have had under the pre-SFFA approach or the
    2023–2024 race-blind approach. ↑
100. . I say “almost nothing” rather than “nothing” because the duration of the status quo ante can, in some exceptional circumstances, be relevant to the buildup of reasonable reliance interests that make the status quo ante the relevant baseline. I discuss these exceptions infra subpart III(C). ↑
101. . See Grutter, 539 U.S. at 357 (Thomas, J., concurring in part and dissenting in part) (stating that “there is no pressing public necessity in maintaining a public law school at all”). ↑
102. . 426 U.S. 229 (1976). ↑
103. . Id. at 242 (“Disproportionate impact is not irrelevant, but it is not the sole touchstone of an invidious racial discrimination forbidden by the Constitution. Standing alone, it does not trigger the rule, that racial classifications are to be subjected to the strictest scrutiny and are justifiable only by the weightiest of considerations.”) (citation omitted). ↑
104. . See, e.g., Charles R. Lawrence III, The Id, the Ego, and Equal Protection: Reckoning with Unconscious Racism, 39 Stan. L. Rev. 317, 323 (1987) (critiquing the discriminatory intent standard because it ignores the reality of unconscious bias and the effect of historic racism on the “collective unconscious”); David A. Strauss, Discriminatory Intent and the Taming of Brown, 56 U. Chi. L. Rev. 935, 955 (1989) (arguing that the discriminatory intent standard articulated in Washington v. Davis tamed the decision in Brown by excluding considerations of “stigma, subordination, or second-class citizenship” and ignoring the impact of state action on “established institutions”); William M. Wiecek, Structural Racism and the Law in America Today: An Introduction, 100 Ky. L.J. 1, 7 (2011) (calling Washington v. Davis “the single most important decision of the United States Supreme Court for understanding the failure (or refusal) of the Justices to recognize structural racism”). ↑
105. . Current law places the burden on the defendant to “demonstrate that the challenged practice is job related for the position in question and consistent with business necessity.” 42 U.S.C. § 2000e-2(k)(1)(A)(i). That language was added to Title VII by the Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071, 1071 (1991), essentially codifying the Supreme Court’s interpretation of the pre-amendment Act in Griggs v. Duke Power Co., 401 U.S. 424, 431, 436 (1971). ↑
106. . Davis, 426 U.S. at 245–48. ↑
107. . Id. at 248. ↑
108. . Legacy and donor preferences provide financial support to the university by encouraging alumni loyalty and donations. Preferences for faculty children help the university recruit and retain faculty. These are race-neutral aims even though they have a disparate racial impact. ↑
109. . Legacy preferences are in decline but can still be found, including at some selective state universities. See Audry Williams June & Brian O’Leary, Which Colleges Consider Legacies in Their Admissions? Not Just the Ones You’re Thinking, Chron. Higher Educ. (Dec. 12, 2023), https://www.chronicle.com/article/which-colleges-consider-legacies-in-their-admissions-not-just-the-ones-youre-thinking [https://perma.cc/KQ2Y-VHPW] (listing the close to 600 colleges that consider legacy status in their admissions process). For example, the included table indicates that as of December 2023, the University of Virginia (UVA) gave legacy preferences, which might or might not be accurate. Id. For the 2023–2024 admissions cycle, UVA replaced a checkbox for legacy status with an optional question inviting applicants to discuss any “personal or historic connection” to the university. Nick Anderson, University of Virginia Will Limit “Legacy” Factor in Admissions, Wash. Post (Aug. 2, 2023), https://www.washingtonpost.com
     /education/2023/08/01/uva-legacy-admissions-college-application/ [https://perma.cc/FPA9-2VAV]. The question goes on to state that the kinds of relationships that can establish such a connection “might include, but are not limited to, being a child of someone who graduated from or works for UVA, a descendant of ancestors who labored at UVA, or a participant in UVA programs.” Id. Reduction in the role that legacy status plays thus appears to be paired with an effort to increase or maintain racial diversity through race-neutral means. I reserve comment on the fact that the university founded by Thomas Jefferson uses a euphemism to describe persons who performed involuntary labor at UVA by virtue of their enslavement (perhaps even by Jefferson himself). ↑
110. . For a potent critique of meritocracy, see generally Michael J. Sandel, The Tyranny of Merit (2020). ↑
111. . 572 U.S. 291 (2014). ↑
112. . Id. at 298–99, 314 (plurality opinion). ↑
113. . Id. at 298 (“[T]he university revised its undergraduate admissions process, but the revision still allowed limited use of race-based preferences.”). ↑
114. . Id. at 298; id. at 315 (Roberts, C.J., concurring); id. at 316 (Scalia, J., concurring in the judgment); id. at 332 (Breyer, J., concurring in the judgment); id. at 337 (Sotomayor, J., dissenting). ↑
115. . Justice Kennedy’s plurality opinion differentiated the circumstances in Schuette from those in Washington v. Seattle School District No. 1, 458 U.S. 457 (1982), in which the court invalidated a voter-initiated ban on busing to desegregate schools because the busing plan inured “primarily to the benefit of the minority.” Id. at 472. Justice Kennedy also distinguished Reitman v. Mulkey, 387 U.S. 369 (1967), which held that a voter measure prohibiting any state legislative interference with an owner’s prerogative to decline to sell or rent residential property on any basis was invalid under equal protection because it prevented action against race discrimination, id. at 379, 381, and Hunter v. Erickson, 393 U.S. 385 (1969), in which the Supreme Court struck down a voter-enacted city charter amendment requiring antidiscrimination housing ordinances to be approved by referendum. Id. at 386, 391. Justice Kennedy interpreted this line of cases to establish the principle that “when hurt or injury is inflicted on racial minorities by the encouragement or command of laws or other state action, the Constitution requires redress by the courts.” Schuette, 572 U.S. at 313. Justices Scalia and Thomas would have overruled the line of cases Justice Kennedy distinguished. Id. at 322 (Scalia, J., concurring in the judgment) (“Patently atextual, unadministrable, and contrary to our traditional equal-protection jurisprudence, Hunter and Seattle should be overruled.”). ↑
116. . Schuette, 572 U.S. at 338 (Sotomayor, J., dissenting). ↑
117. . Students for Fair Admissions v. President & Fellows of Harvard Coll., 143 S. Ct. 2141, 2156 n.2 (2023) (“[D]iscrimination that violates the Equal Protection Clause of the Fourteenth Amendment committed by an institution that accepts federal funds also constitutes a violation of Title VI.”). ↑
118. . Justice Gorsuch, joined by Justice Thomas, concurred in the majority opinion in SFFA, even as he called for reading Title VI independently from equal protection. Id. at 2219–20 (Gorsuch, J., concurring). His grounds for distinguishing the statute from the constitutional obligation differ from those expounded here. ↑
119. . See Washington v. Davis, 426 U.S. 229, 239 (1976) (“[O]ur cases have not embraced the proposition that a law or other official act, without regard to whether it reflects a racially discriminatory purpose, is unconstitutional solely because it has a racially disproportionate impact.”). ↑
120. . 42 U.S.C. § 2000e-2(k)(1)(A)(i). ↑
121. . See, e.g., 28 C.F.R. § 42.104(b)(2) (2024) (prohibiting employment “criteria or methods of administration” that “have the effect of” discriminating on the basis of “race, color, or national origin”). The Department of Transportation has a similar regulation, 49 C.F.R. § 21.5(b)(2) (2024), mirroring the Department of Justice prohibition. ↑
122. . See Alexander v. Sandoval, 532 U.S. 275, 278, 293 (2001) (holding that there is no private right of action to enforce disparate impact regulations under Title VI). ↑
123. . Connecticut v. Teal, 457 U.S. 440, 442–43, 443 n.4 (1982). ↑
124. . Mandala v. NTT Data, Inc., 975 F.3d 202, 210 (2d Cir. 2020) (quoting Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 650 (1989)). The Civil Rights Act of 1991 rejected the holding of Wards Cove but not its discussion of how to demonstrate disparate impact. See Lanning v. Se. Pa. Transp. Auth., 181 F.3d 478, 487–88 (3d Cir. 1999) (showing that the Civil Rights Act of 1991 maintained the Wards Cove disparate impact language); see also Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 308 (1977) (holding that to demonstrate disparate impact, the “proper comparison was between the racial composition of [the defendant school district’s] teaching staff and the racial composition of the qualified public school teacher population in the relevant labor market”). ↑
125. . 431 U.S. 324 (1977). ↑
126. . See id. at 366–67 (holding that victims of “gross and pervasive discrimination” might be deterred from applying, and therefore should not be denied opportunity for relief). ↑
127. . 433 U.S. 321 (1977). ↑
128. . Id. at 330. ↑
129. . U.S. Dep’t of Just., Title VI Legal Manual, § VII(c)(1)(c)(iii)(a) (2021), https://www.justice.gov/crt/fcs/T6Manual7 [https://perma.cc/F24B-WNGE]. ↑
130. . Lucas v. S.C. Coastal Council, 505 U.S. 1003, 1029 (1992). ↑
131. . Richard H. Fallon, Jr., John F. Manning, Daniel J. Meltzer & David L. Shapiro, Hart and Wechsler’s The Federal Courts and The Federal System 523 (7th ed., 2015) (mentioning also Supreme Court review of state court judgments, the Full Faith and Credit Clause and implementing legislation, the Federal Arbitration Act, and “various federal constitutional provisions [that] protect against criminal punishment except in accordance with previously enacted state laws”). ↑
132. . See supra note 40 and accompanying text. ↑
133. . See, e.g., Flood v. Kuhn, 407 U.S. 258, 282–84 (1972) (reaffirming an earlier holding in Federal Baseball Club v. National League, 259 U.S. 200, 209 (1922), that baseball was outside the scope of the Sherman Act, not because it was correct, but because the “aberration is an established one” exhibiting strong private, societal, and public reliance); Dickerson v. United States, 530 U.S. 428, 443 (2000) (reaffirming the Miranda rule because it “has become embedded in routine police practice to the point where the warnings have become part of our national culture” (citing Mitchell v. United States, 526 U.S. 314, 331–32 (1999) (Scalia J., dissenting))); General Electric Co. v. Gilbert, 429 U.S. 125, 142–45 (1976) (holding that Title VII prohibitions against workplace sex discrimination did not apply to excluding pregnancy from workers’ health and disability insurance in part because of reliance on prior EEOC guidance to employers and a consistent Department of Labor interpretation of the Equal Pay Act). ↑
134. . See Amber Dance, Has the Pandemic Put an End to the SAT and ACT?, Smithsonian Mag. (July 15, 2021), https://www.smithsonianmag.com/innovation/has-pandemic-put-end-to-sat-act-180978167/ [https://perma.cc/9XSD-D8Z9] (stating that more than 600 schools, including liberal arts colleges and Ivy League institutions, switched to test-optional policies for the 2020–2021 admissions cycle); Darrell Lovell & Daniel Mallinson, How Test-Optional College Admissions Expanded During the COVID-19 Pandemic, Urb. Inst. (Dec. 16, 2021), https://www.urban.org/research/publication/how-test-optional-college-admissions-expanded-during-covid-19-pandemic [https://perma.cc/A7YW-F7NS] (showing that an increased number of selective universities deployed test-optional admissions policies during the pandemic). ↑
135. . Office of the President, University of California Board of Regents Unanimously Approved Changes to Standardized Testing Requirement for Undergraduates (May 21, 2020), https://www.universityofcalifornia.edu/press-room/university-california-board-regents-unanimously-approved-changes-standardized-testing [https://perma.cc/6ZSY-5TTE]. ↑
136. . Teresa Watanabe, UC Slams the Door on Standardized Admissions Tests, Nixing Any SAT Alternative, L.A. Times (Nov. 18, 2021, 7:52 PM), https://www.latimes.com/california/story/2021-11-18/uc-slams-door-on-sat-and-all-standardized-admissions-tests?utm_id=42788&sfmc_id
     =2973663. [https://perma.cc/CZ3K-2SQL] (reporting that the Board of Regents of the University of California could not find any alternative exam that would avoid biased results, and that UC would continue “test-free admissions now and into the future”). ↑
137. . In its litigation against Harvard, SFFA argued that Harvard’s (race-conscious) admissions process, in fact, had just this aim. See Brief for Petitioner at 72–75, Students for Fair Admissions v. President & Fellows of Harvard Coll., 143 S. Ct. 2141 (2023) (Nos. 20-1199, 21-707) (arguing that Harvard’s personality rating was intended as an “anti-Asian penalty”). ↑
138. . See id. at 28–30 (offering evidence that Harvard consistently awarded Asian American applicants lower personal ratings that led to a statistically significant admissions penalty). ↑
139. . Brief for Respondent at 3–4, Students for Fair Admission, 143 S. Ct 2141 (No. 20-1199). ↑
140. . Tribe, supra note 98, at 29 n.71. ↑
141. . I say “knowingly” rather than “purposefully” because I assume that the white Jackson authorities held racist views that led them to oppose Blacks swimming in the same pools as whites but were indifferent to whether Black residents of Jackson swam in their own pools. ↑
142. . Of course, the Palmer Court itself thought otherwise. It distinguished Griffin v. County School Board of Prince Edward County, 377 U.S. 218 (1964)—in which the Court invalidated a Virginia County’s use of newly granted state authority to close its public schools and fund private segregated schools rather than integrate the public ones, id. at 231—on the ground that in Griffin, the government was implicated in the racially disparate alternative educational opportunities in a way that was not true of the private racially segregated swimming pools. See Palmer v. Thompson, 403 U.S. 217, 222 (1971) (arguing that, by contrast with Griffin, “there is nothing here to show the city is directly or indirectly involved in the funding or operation of” formerly public pools that had been handed over to segregated operators). ↑
143. . Kennedy, supra note 34, at 180; see also id. at 198 (“[T]he city did not treat everyone the same. It did not racially stigmatize whites but did racially stigmatize blacks.”). Professor Kennedy also observed: “In addition to that stigmatic harm, discrimination at the pool prompted some youngsters to frequent unsupervised locales at which they faced higher risks of danger, leading to tragic injuries or even death.” Id. at 180–81. Like the unequal opportunities for alternative swimming venues discussed by Professor Tribe, this harm does not readily generalize to other contexts, and thus I do not discuss it further. ↑
144. . Id. at 205. ↑
145. . Id. ↑
146. . 163 U.S. 537 (1896). See Kennedy, supra note , at 198–200 (discussing the similar efforts to avoid desegregation in railcars and pools). ↑
147. . Plessy, 163 U.S. at 551. ↑
148. . See Kennedy, supra note , at 199–200 (“[T]he Court in Palmer, like the Court in Plessy, rejected the plaintiffs’ contention, concluding that they were mistaken, that what they took to be a denigrating stigmata was actually a reasonable policy undertaken in good faith for the benefit of all.”). ↑
149. . Compare Plessy, 163 U.S. at 540 (describing Louisiana statute “providing for separate railway carriages for the white and colored races”), with Palmer v. Thompson, 403 U.S. 217, 218 (1971) (describing facially race-neutral action by Jackson, Mississippi). ↑
150. . See Plessy, 163 U.S. at 540 (explaining that the statute “enacts ‘that all railway companies carrying passengers . . . shall provide equal but separate accommodations for the white, and colored races’”). ↑
151. . See, e.g., Matthew D. Adler, Expressive Theories of Law: A Skeptical Overview, 148 U. Pa. L. Rev. 1363, 1446–47 (2000) (arguing that government expression by itself does not necessarily alter anyone’s “standing as a full member of the political community” and that therefore the “linguistic meaning” of government action “is neither necessary nor sufficient to cause status harm to” anyone); Steven D. Smith, Expressivist Jurisprudence and the Depletion of Meaning, 60 Md. L. Rev. 506, 562–65, 576 (2001) (critiquing the notion of law’s “objective” meaning and questioning the normative grounds for expressivism). But see Elizabeth S. Anderson & Richard H. Pildes, Expressive Theories of Law: A General Restatement, 148 U. Pa. L. Rev. 1503, 1542–45 (2000) (offering a general defense of expressive theories and specific arguments relevant to the equal protection context); Michael C. Dorf, Same-Sex Marriage, Second-Class Citizenship, and Law’s Social Meanings, 97 Va. L. Rev. 1267, 1298 (2011) (summarizing various doctrines that support the conclusion that extant case law “sometimes forbids the government from acting or speaking in ways that connote the second-class citizenship of various persons”). ↑
152. . There are reasons to reject this proposition. See Tyler Rose Clemons, Coercive Ideology, 83 Md. L. Rev. 1121, 1186–89 (2024) (arguing that official expressions of white supremacy during Jim Crow served as a means of coercing behavior, not merely as an imposition of stigma). ↑
153. . Allen v. Wright, 468 U.S. 737, 755 (1984) (quoting Heckler v. Mathews, 465 U.S. 728, 740 (1984)). ↑
154. . To be more precise, Palmer’s bottom line has never been repudiated. As noted above, its statement that intent is irrelevant to the evaluation of a facially race-neutral policy was repudiated in Washington v. Davis. See supra notes 37– and accompanying text. ↑
155. . See supra note 40. ↑
156. . See, e.g., Swann v. Charlotte-Mecklenburg Bd. of Educ., 402 U.S. 1, 15 (1971) (“Once a right and a violation have been shown, the scope of a district court’s equitable powers to remedy past wrongs is broad, for breadth and flexibility are inherent in equitable remedies.”); Davis v. Bd. of Sch. Comm’rs of Mobile Cnty., 402 U.S. 33, 37 (1971). As the unanimous Court in Mobile County explained:“[N]eighborhood school zoning,” whether based strictly on home-to-school distance or on “unified geographic zones,” is not the only constitutionally permissible remedy; nor is it per se adequate to meet the remedial responsibilities of local boards. Having once found a violation, the district judge or school authorities should make every effort to achieve the greatest possible degree of actual desegregation, taking into account the practicalities of the situation.

     Id. at 37. ↑

157. . See Green v. Cnty. Sch. Bd. of New Kent Cnty., 391 U.S. 430, 439, 441–42 (1968) (rejecting a formerly de jure segregated school district’s adoption of a facially race-neutral “freedom-of-choice” student assignment plan without finding it necessary to assess whether the “plan might of itself might be unconstitutional”). ↑
158. . See, e.g., Freeman v. Pitts, 503 U.S. 467, 494 (1992) (“Once the racial imbalance due to the de jure violation has been remedied, the school district is under no duty to remedy imbalance that is caused by demographic factors.”). ↑
159. . Cf. Neil H. Buchanan & Michael C. Dorf, Justice Delayed: Government Officials’ Authority to Wind Down Constitutional Violations, 103 B.U. L. Rev. 2065, 2090 (2023) (explaining that the related question of how much time government officials should have to phase out unconstitutional and otherwise unlawful programs is “necessarily context dependent” and thus governed by “a standard rather than a rule”). ↑
160. . See Green, 391 U.S. at 437–38 (holding that under Brown I and Brown II, school boards were “clearly charged with the affirmative duty to take whatever steps might be necessary to convert to a unitary system in which racial discrimination would be eliminated root and branch”); cf. United States v. Fordice, 505 U.S. 717, 743 (1992) (stating that a state “may not leave in place policies rooted in its prior officially segregated system that serve to maintain the racial identifiability of its universities if those policies can practicably be eliminated without eroding sound educational policies”). ↑
161. . Cf. S. Camden Citizens in Action v. N.J. Dep’t. of Env’t Prot., No. Civ.A. 01–702, 2006 WL 1097498, at *22–23 (D.N.J. Mar. 31, 2006) (explaining that the disparate racial impact from a permitting decision for a polluting plant would be unlawful if it resulted from intentional discrimination). ↑
162. . See Michael C. Dorf, Even a Dog: A Response to Professor Fallon, 130 Harv. L. Rev. F. 86, 88 (2016) (using the same example to illustrate the same problem). ↑
163. . See Richard H. Fallon, Jr., Constitutionally Forbidden Legislative Intent, 130 Harv. L. Rev. 523, 541 (2016) (describing “objective” legislative intent as “conceptually distinct from the actual psychological intentions or motivations of the legislators who voted to enact a statute”). ↑
164. . See, e.g., Dan B. Dobbs & Caprice L. Roberts, Law of Remedies § 2.5 (3d ed. 2018) (“[E]quitable relief must be denied unless the legal remedy is inadequate and denial will not cause irreparable harm to plaintiff.”). ↑
165. . See, e.g., Brown v. Bd. of Educ. of Topeka, 98 F. Supp. 797, 797 (D. Kan. 1951) (describing the remedies sought by plaintiffs, which included injunctive and declaratory relief but not damages). ↑
166. . The issue arises somewhat routinely in cases involving challenges to admissions policies because, by the time a case reaches the Supreme Court, rejected applicants will typically have completed their education elsewhere. For example, Abigail Fisher’s case against the University of Texas made its way onto the Supreme Court’s plenary docket twice, in 2013 and again in 2016, even though “she graduated from another university in May 2012, thus rendering her claims for injunctive and declaratory relief moot.” Fisher v. Univ. of Tex. at Austin, 758 F.3d 633, 639 (5th Cir. 2014). Nonetheless, presumably because Fisher also sought monetary relief in the form of a refund of her original $100 application fee, the Supreme Court apparently concluded that it had the authority to adjudicate her claims each time. Cf. id. at 640 (explaining that, although the university contested plaintiff’s standing in the Supreme Court, the Justices decided the merits without addressing standing and thereby precluded consideration of standing by the appeals court on remand); see also Doe v. Kamehameha Schs., 470 F.3d 827, 834 n.5 (9th Cir. 2006) (en banc) (“Plaintiff now seeks only damages because he is no longer a high school student. This case is not moot, however, because if the Kamehameha Schools’ admissions policy were unlawful, Plaintiff has a possible claim for money damages.”). ↑
167. . See Wechsler, supra note 2, at 15 (distinguishing the ad hoc invocation of principles characteristic of politics from judicial decisionmaking, which “must be genuinely principled, . . . reaching judgment on analysis and reasons quite transcending the immediate result that is achieved.”) ↑
168. . Compare Peter Westen, The Empty Idea of Equality, 95 Harv. L. Rev. 537, 542 (1982) (arguing “that statements of equality logically entail (and necessarily collapse into) simpler statements of rights”), with Kent Greenawalt, How Empty Is the Idea of Equality?, 83 Colum. L. Rev. 1167, 1168 (1983) (contesting Westen’s call to banish equality while acknowledging that “claims about substantive principles of equality . . . call[] forth competing views about relevant criteria”). ↑