Editor's Note
Cold civil wars are defined by recurrence. Programs are struck down, declared finished, and quietly rebuilt under new names. Institutional defeats become occasions for refinement. The deeper conflict is not over any single policy, but over the moral architecture that governs our ruling class. Until that architecture is confronted, each apparent victory will prove provisional.
This essay, from Jesse Merriam, treats affirmative action as a case study in how the destructive regime survives — and why the struggle is more enduring than many on the Right wish to admit.
In 2023, the Supreme Court banned race-conscious admissions in Students for Fair Admissions v. Harvard. The assumption thereafter was that affirmative action would collapse. It did not.
Helen Andrews, in her recent Compact essay "America Needs a Better Meritocracy," notes that Asian enrollment at Harvard jumped from 26 percent in the class of 2025 to 41 percent in the class of 2029, white enrollment dropped from 47 percent to approximately 31 percent, and black and Latino shares remained a roughly stable "critical mass," with relatively small fluctuations. Based on these numbers, Andrews concludes that "Harvard did not stop discriminating by race; it simply stopped doing so against Asians."
Unlike many commentators who prematurely declared affirmative action dead after SFFA, Andrews recognizes that racial preferences have survived the Supreme Court's ruling.
But even Andrews' more sophisticated reading misses something crucial. What Harvard's enrollment data reveals is not merely that discrimination persisted, but that the institution executed precisely the adaptation strategy outlined during the SFFA litigation itself. During oral argument, plaintiffs' counsel (i.e., the group challenging Harvard's affirmative action program) explicitly suggested that Harvard could manipulate socioeconomic and experiential factors to reduce white enrollment while maintaining an "absolute floor" of ten percent black enrollment. After the SFFA Court prohibited explicit racial preferences, Harvard preserved the outcomes it viewed as legitimacy-essential — black and Latino representation — by using white enrollment as the adjustment variable, exactly as the litigation had mapped out.
The admissions mechanism changed; the moral imperative for diversity did not. As I have explained in various essays, this is exactly what anyone who understands our constitutional regime should have expected from the SFFA decision. This dynamic of adaptation rather than elimination is precisely what makes affirmative action so difficult to dislodge. In this sense, Andrews shows a greater understanding of American constitutional law than many of the nation's leading scholars.
Consider, for example, how at the conclusion of the Court's most recent term, Yale Law School professor Jed Rubenfeld declared that "the era of affirmative action really has come to an end." As I explained in a lengthy law review article tracing the lineage of such proclamations about "the end of affirmative action," scholars and commentators have repeatedly announced affirmative action's death — only to watch its supposed corpse spring back to life, time and again.
In each instance, the obituary followed a familiar pattern. After a Supreme Court decision appeared to deliver a decisive blow — invalidating a racial preference in a particular area of law — the ruling was widely hailed as the end of the affirmative action regime. Yet, soon after the decision was issued, it became clear that affirmative action had not been eliminated or even meaningfully weakened.
This persistence gives rise to a striking paradox. For two generations, the Supreme Court — in cases ranging from government contracting to public employment to university admissions — has repeatedly invalidated affirmative action programs as a matter of constitutional law. Over that period, the public has consistently opposed affirmative action as a matter of policy, and several state governments have formally banned the use of racial preferences in their public colleges and universities. Despite this sustained legal, social, and political resistance, affirmative action has managed not merely to survive but to strengthen — migrating, rebranding, and embedding itself ever more deeply within institutional practices.
This pattern has created what I have called "the affirmative action regime." By this, I mean not a single program but a system of racial preferences and diversity-driven policy decisions that, over several decades, have come to pervade large swaths of American public and private life.
It is a regime, then, and not a policy, as it governs despite resistance. Indeed, few — if any — programs in modern American public life have shown this degree of resilience in the face of such broad and sustained opposition.
To explain why affirmative action has survived, and why even the Trump administration's current offensive will likely fall short, this essay proceeds in two parts. The first is retrospective and theoretical, focusing on why past efforts to end affirmative action have failed. The second is prospective and political, examining what the Trump administration is now attempting and why these efforts, despite their unprecedented scope, remain constrained by the constitutional morality that sustains the regime.
A Three-Part Framework Explaining Conservative Failure
To understand why the movement against affirmative action has repeatedly stalled, we need a more general account of what enables political movements to effect durable change through judicial power. Over the last decade, I have argued that legal liberalism and legal conservatism operate according to fundamentally different movement logics — differences that help explain why liberals have been far more successful at translating court victories into lasting political transformation.
Three distinctions matter most: support structures, political nexus, and public morality.
The first concerns what judicial-politics scholars call support structures: the network of organizations, resources, and professional incentives that sustain a legal movement over time. The legal left possesses an extensive infrastructure devoted to strategic litigation, including organizations such as the ACLU, the NAACP Legal Defense Fund, Lambda Legal, and Planned Parenthood. These organizations file cases as part of long-term litigation campaigns designed to reshape judicial doctrine, institutional practices, and public norms.
Legal conservatism lacks anything comparable. The Federalist Society — the movement's central institutional node — does not litigate at all. Libertarian-oriented firms such as the Institute for Justice and the Pacific Legal Foundation pursue deregulatory agendas largely orthogonal to conservative politics. Alliance Defending Freedom comes closest to functioning as a conservative analogue to the ACLU, but its mission is substantively narrow and its institutional reach far more limited. The result is a movement that can win doctrinal victories without possessing the organizational capacity to translate those victories into durable governance.
The second distinction concerns political nexus. Courts do not operate in isolation. Judicial decisions are shaped by, and depend upon, the broader political environment — not only in how opinions are written, but also in whether rulings are implemented, expanded, or quietly neutralized. Successful legal movements must therefore work in concert with political actors willing and able to give judicial decisions practical force.
Here again, legal conservatism has been at a disadvantage. For decades, the conservative legal movement has centered itself on originalism — a theory of interpretation rather than a political program. Many leading figures within that movement have explicitly distanced themselves from partisan politics and, in some cases, actively opposed dominant factions within the Republican Party. Legal liberalism, by contrast, has often worked in close coordination with the Democratic Party to advance shared policy objectives through both judicial and administrative governance.
The third and most important distinction concerns public morality. Most social change does not occur through judicial decrees alone. It occurs through voluntary private action — through universities, corporations, nonprofits, and cultural institutions responding to what they perceive as moral obligation. A successful legal movement must therefore align itself with the governing moral ethos of the constitutional order.
On this dimension, legal liberalism enjoys a decisive advantage.
In various works over the last several years — most notably in my 2023 booklet How We Got Our Antiracist Constitution: Canonizing Brown v. Board of Education in Courts and Minds — I have argued that the civil-rights revolution did not merely reshape constitutional doctrine but entrenched a governing constitutional morality, one that persists regardless of, and often in tension with, formal law. That morality is organized around two mutually reinforcing imperatives and is enforced not primarily through judicial holdings but through institutional practice.
On the affirmative side, diversity is treated as a supreme moral good, operationalized by universities, accrediting bodies, federal agencies, corporate compliance offices, and human-resources departments, such that any institution unable to demonstrate it faces scrutiny and sanction. On the negative side, discrimination is treated as the supreme evil, such that any legal rule that obstructs the eradication of discrimination must likewise yield.
Once this constitutional morality is entrenched, opposition becomes largely internal rather than external. Legal conservatives have thus been forced to argue within these two axes rather than against them. In doing so, they have surrendered much of what would otherwise be a fundamental critique of affirmative action.
A Debate Over How, Not Whether, to Practice Affirmative Action
This dynamic helps explain why affirmative action and DEI programs have persisted despite decades of opposition. In my 2023 law review article, "Why DEI Will Not Die," I applied this three-part framework to show how institutional weakness, political disengagement, and moral asymmetry combined to insulate affirmative action from effective challenge.
For many years, the only legal organization consistently filing lawsuits against affirmative action was the Center for Individual Rights, a small firm with limited resources. At the political level, national Republicans largely avoided the issue, constrained by alliances with business interests and the military, both of which had embraced diversity programs. At the moral level, opposition to affirmative action was forced to speak the language of diversity itself.
The result is that contemporary disputes over affirmative action are rarely debates about whether racial preferences should exist at all. They are debates about how affirmative action should be practiced — namely, which racial proxies to employ, which metrics to prioritize, and which mechanisms best obscure the role of race while preserving desired outcomes.
This was illustrated vividly within the Students for Fair Admissions v. Harvard litigation, in which none of the parties was willing to defend a color-blind, merit-based admissions system grounded in academic criteria alone. The reason was obvious. Based on Harvard's own admissions data, a purely academic top-ten-percent admissions system would have yielded black enrollment under one percent (the precise percentage in the year in question was 0.76 percent).
Such a result is morally intolerable within a constitutional order that treats diversity as a foundational good. Accordingly, even the plaintiffs challenging Harvard's admissions practices were careful to emphasize that alternative criteria could be used to "boost underrepresented minority representation" and reduce white enrollment. During oral argument, plaintiffs' counsel suggested that Harvard could manipulate socioeconomic and experiential factors to guarantee an "absolute floor" of ten percent black enrollment, and likely more, using Harvard's sophisticated admissions data. The implication was unmistakable: even the challengers resisted race-blind meritocracy as politically — and morally — unacceptable.
The problem with Harvard's program, in other words, was not that it sought racially calibrated outcomes, but that it pursued them too openly. The litigation instead pressed institutions to embed racial preferences more subtly within nominally race-neutral frameworks.
This is what constitutional governance looks like when it is structured around the moral imperative of diversity. Whether one purports to oppose or defend affirmative action, the shared commitment is not to color-blind meritocracy but to preserving racially calibrated elite institutions — where underrepresentation is treated as a problem only for some groups, and the dispute concerns how to maintain that calibration without drawing undue attention to how it is achieved.
The Aftermath of SFFA
This dynamic explains why the Supreme Court's SFFA decision, despite invalidating Harvard's and UNC's explicit policies, stopped well short of banning affirmative action outright — and why it was immediately followed by the familiar round of affirmative-action autopsies. In response, I warned that these autopsies would once again prove premature and predicted that the decision would do little to alter racial outcomes in elite admissions.
Based on the first year of data, that prediction appears to have been correct. A recent study by the New York Times of 59 colleges found a very modest (roughly one percent) drop after the SFFA ruling in Hispanic and black enrollment. Likewise, a City Journal study found that at highly selective colleges "black enrollment … remained virtually unchanged." The same was true in law school enrollment, where black enrollment decreased by only .13 percent and Hispanic enrollment remained "almost identical to the prior year."
Observers expressed genuine surprise. Writing in The Atlantic, Rose Horowitch noted that, following the SFFA decision, many selective institutions had seemingly increased their affirmative action efforts, concluding that "things have not gone the way anyone expected." But this, of course, is exactly what I expected. It wasn't because I had some special interpretation of the SFFA ruling, or some special insight into how university admissions work. Rather, it was because I understood what controls our constitutional order.
It is not the Constitution, and it is not even the Supreme Court. It is our civil-rights morality.
The Harvard enrollment data that opened this essay illustrates the mechanism. After SFFA foreclosed explicit racial preferences, Harvard did not abandon its commitment to maintaining certain racial outcomes. It preserved a critical mass of black and Latino representation by dramatically restructuring its incoming class — substantially increasing Asian enrollment and lowering white enrollment. But this was not the elimination of affirmative action: It was adaptation through substitution. And it was entirely predictable — perhaps even inevitable — as the working out of civil-rights constitutionalism in an increasingly diverse social order.
Our Unslayable Ghoul
The persistence of affirmative action calls to mind a famous passage from Justice Scalia's concurrence in a 1993 Supreme Court case. In it, Scalia described how a repeatedly discredited legal doctrine would inevitably reappear: "like some ghoul in a late-night horror movie that repeatedly sits up in its grave and shuffles abroad, after being repeatedly killed and buried."
Affirmative action occupies a similar place in our constitutional order. No matter how often it is invalidated, condemned, or banned, it returns — reshaped, renamed, and newly justified — because it is sustained by the moral logic of the civil-rights regime itself. Perhaps only a political actor willing to override institutional norms and moral conventions could finally confront this ghoul. Could that actor be President Trump?
In a second installment, I will take up that question.
