The Cause of a Disparity Does Not Design its Remedy

Quit looking for causes of disparities and focus on solutions

THE CAUSE OF A DISPARITY DOES NOT DESIGN ITS REMEDY

Quit looking for causes of disparities and focus on solutions

Dave Gilbert and Jake Mackey

The reason we haven’t been able to have an honest discussion about solutions to the problem of racial inequality in wealth, educational attainment, and so on is that people on the left and people on the right get hung up in their disagreements over the source of disparities. The left tends to blame disparities on racism, past or present, and the right tends to blame them on culture, behavior, or most ominously, genetics. This essay proposes that we simply stop worrying about the origin of disparities and focus on solutions. There’s a principled reason to quit obsessing about origins, namely, the origin of a disparity does not design the remedy for it.

Consider the recent Supreme Court decision in Louisiana v. Callais, where the court came out against so-called racial gerrymandering. In the wake of the decision, Tennessee Republicans approved a new congressional map that broke up Memphis’s majority-black, Democratic-held district. The resulting protest imagery was blunt. Tennessee State Rep. Justin Jones (D) burned a paper replica of the Confederate flag, casting the map not merely as partisan hardball but as racial regression. But partisan gerrymandering is legal, both parties practice it, and the Supreme Court has placed it beyond the reach of the federal courts. The protest’s underlying claim is that the same tactic changes character when the seat it eliminates is a majority-black district once protected by Section 2 of the Voting Rights Act, so that with Callais having loosened that protection, ordinary partisan map-drawing now reads as racial exclusion rather than the hardball both parties accept everywhere else.

Many on the left have long argued that persistent racial disparities—in political representation, corporate leadership, higher education, home ownership, incarceration, or health outcomes—reflect the continuing effects of racism, white supremacy, and historical discrimination, and that explanations tracing those disparities to anything else are suspect for that reason. So explanations that locate a disparity in aspects of black culture—statistically common black preferences, proclivities, habits, and even dysfunction—do not escape the charge of racism, because they are read as a displaced theory of inherent black inferiority. The charge holds even when their defenders explicitly reject hereditarianism, the claim that group differences are written into biology and passed down genetically.

Ibram X. Kendi gives the most influential version of this binary in How to Be an Antiracist. For Kendi there is no neutral category: a policy is either antiracist, in that it reduces racial inequity, or racist, in that it produces or sustains it, and a facially race-neutral policy that still yields unequal outcomes falls on the racist side by definition, whatever its intent. The same logic pulls the cultural explanation back toward racism, because once black disadvantage is attributed to “black culture,” the next question is where that culture came from, and the antiracist answer is that racism, exclusion, segregation, and inherited deprivation produced it. Either way the framework forecloses the middle ground where a policy might be neither racist nor antiracist—where unequal outcomes follow from a defensible rule that was not designed to produce them and isn’t best remedied by reversing it.

Thomas Sowell’s Black Rednecks and White Liberals shows what that foreclosed middle ground actually contains. Sowell, a conservative, argues that the self-destructive features of urban black culture are real and cultural rather than genetic, and he traces them to the “cracker culture” that Scots-Irish and border-British migrants carried into the antebellum South and that black Southerners absorbed over generations of proximity. His causal story runs straight through slavery and the white South, the kind of origin Kendi’s framework would accept—yet Sowell draws from it a remedy that points inward, toward change within black communities, rather than toward undoing a racist policy. The causal account and the remedial conclusion come apart, which is the possibility Kendi’s binary rules out in advance.

The binary rules it out because it treats the remedy as following automatically from the cause. The remedy it licenses is explicitly race-conscious, as Kendi writes in How to Be an Antiracist:

The only remedy to negative racist discrimination that produces inequity is positive antiracist discrimination that produces equity. The only remedy to past negative racist discrimination that has produced inequity is present positive antiracist discrimination that produces equity. The only remedy to present negative racist discrimination toward inequity is future positive antiracist discrimination toward equity.

For Kendi and many on the left, opposing a race-conscious remedy is not a competing view about how to close the gap but itself a form of racism. But the inference from cause to remedy does not hold. Even if every racial disparity at issue is, in some important sense, downstream from historical racism or structural discrimination, the question of the best way to remediate it remains. Even if past or present racism is the cause of disparity, that fact alone does not settle what the law may do to particular individuals and institutions in the name of correcting group disparity—nor, regardless of what the law permits, what would actually work to close the gap. A disparity may tell us where to look. It does not, by itself, design the remedy.

Four problems arise when disparity is treated as dispositive. First, a boundary problem: disparity may identify a broad group, but it does not tell the state which individuals legitimately belong to that group, or which subgroups or ancestral lines should count. Second, a cost-allocation problem: a preference for one group often burdens individuals in other groups. Third, a conflict-of-interests problem: the same disparity can point toward competing remedies. Fourth, a causal problem: race may be visible in the data while standing in for something more direct.

There are two general approaches to designing remedies. One relies on the liberal tradition and begins from equal treatment. It seeks to remove barriers that prevent individuals from pursuing their own goals, and where race must be considered, it favors narrow remedies tied to specific wrongs. On this view, removing barriers is the best way to move toward fairer, and perhaps more proportional, outcomes over time. A second approach is less trusting of that tradition and is often associated with critical race theory and antiracism. It treats persistent disparity as a problem to be addressed directly, using broad racial classifications to close gaps in proportional representation across institutions.

Vitolo v. Guzman is useful because it illustrates both the doctrinal disagreement between these approaches and the boundary problem that follows from the second one. The case involved the Restaurant Revitalization Fund, a COVID-era small-business relief program administered by the Small Business Administration. For the first 21 days, the program gave priority to restaurants owned by women, veterans, and owners the SBA treated as socially and economically disadvantaged under racial and ethnic presumptions. We discussed the case in a Journal of Free Black Thought article titled “Why Did Critical Race Theory Emerge from Legal Studies?” because Vitolo gave a concrete example of the collision between equity arguments and equal-protection limits.

Judge Bernice Bouie Donald’s dissent gives the strongest legal version of the broader remedial view. She looked to “centuries of intentional discrimination and oppression,” and to the federal government’s history in housing, banking, farm loans, and other domains. On that view, the government is not merely correcting a specific agency’s specific wrong with a narrow remedy. It is using present policy to address the accumulated moral debt of American history.

Judge Amul Thapar’s majority opinion rejected that move. If the problem was lack of access to capital, the government could prioritize businesses that actually lacked it. If the problem was exclusion from prior COVID relief, it could prioritize businesses that had missed out. If the problem was pandemic injury, it could prioritize businesses that suffered the greatest losses. The government did not have to sort restaurant owners into racial and ethnic categories to address those injuries.

That is the doctrinal divide. Thapar’s reasoning is more consistent with current equal-protection doctrine than is Donald’s. Racial classifications trigger strict scrutiny; the wrong must be concrete and traceable to the actor adopting the remedy; race-neutral alternatives must be seriously considered; and the remedy must help the people actually harmed through the least discriminatory means available. Those limits are not merely procedural. They reflect the current constitutional settlement: race-conscious remedies are permitted only under narrow conditions, while broader efforts to equalize group outcomes require a different account of what equal protection allows.

Callais brings that divide into voting-rights law. The Court’s majority and its supporters saw Louisiana’s second majority-black district as race-predominant line-drawing, or racial gerrymandering, that Section 2 did not require and equal protection could not tolerate. On that view, equal protection protects the individual voter from being sorted by race, and Section 2 cannot require the state to subordinate that individual right to a racial target. The dissenting justices and their supporters began from a different premise. In their view, equal protection is not only an individual right against racial sorting. It also protects the political power of a racial group.

Problem 1: The boundary problem

Vitolo shows what happens when a broad claim about historical injury has to be converted into eligibility rules for public benefits. Thapar emphasized how arbitrary the resulting ethnic lines appeared. The program preferred Pakistanis over Afghans, Japanese over Iraqis, and Hispanics over Middle Easterners. That was the boundary problem in concrete form. Once the government uses proportional disparity or historical disadvantage to allocate public benefits by group, it has to explain why one adjacent group is inside the line and another is outside it. If black-owned businesses were underrepresented in prior relief, that would matter. But comparable claims from excluded groups would still leave the state with the same question: where should the eligibility line be drawn, and why?

The same problem appears inside the categories themselves. Should the same presumption apply to a business owned by an ADOS black American whose family lived under Jim Crow and a business owned by a recent black immigrant from an upper-middle-class Nigerian family? Maybe the answer is yes. Or maybe no. The disparity does not tell us.

An ancestry-based remedy does not escape the problem. The state still has to decide who qualifies, how far back the inquiry goes, what records count, how uncertainty is handled, and who falls outside the line. Some Americans who identify as white may have black enslaved ancestors. Many Americans have mixed ancestry. Federal racial classification now largely relies on self-identification rather than a scientific racial taxonomy, and many Americans change their racial identification over time.

California illustrates the point by showing how quickly a moral claim about historical injustice becomes an administrable boundary problem. SB 518, signed in 2025, created the Bureau for Descendants of American Slavery within the Civil Rights Department and requires a Genealogy Division to verify descendant status, defined as direct lineage to a person subjected to American chattel slavery before 1900. That turns the boundary problem into a bureaucracy. Whatever line the state draws across thousands or millions of cases will create new equal-protection and nondiscrimination challenges from people who are excluded. The moment ancestry becomes the basis for public benefits, the state is back in the business of policing descent and litigating the boundary it has drawn.

This is the boundary problem. Disparity may tell us that a broad group is disadvantaged, but it does not tell us which individuals legitimately belong to that group, which subgroups or ancestral lines should count, who should verify membership, or why benefits should go to one adjacent category rather than another.

Problem 2: Cost allocation

Public benefits and admissions are different settings, but both distribute scarce opportunities. A preference for one group imposes costs on members of other groups. The cost-allocation problem is visible in Vitolo. If the preferential program had stood, Afghans, Iraqis, and Middle Easterners would have been pushed behind favored ethnic groups in line for finite relief funds.

Students for Fair Admissions (SFFA) shows the problem manifesting in elite admissions. Asian applicants were discriminated against because of their race. Harvard made race part of the admissions calculus, and Asian applicants were penalized by that calculus. Students who had worked for years to build exceptional academic records could still find themselves burdened by the one thing they could not change: their race. That is an ugly result, and it cannot be made fair by calling it part of a larger pursuit of racial balance.

The issue is whether proportional racial representation should become the governing aim, even when pursuing it means discriminating against other individuals and groups because of their race.

Problem 3: Conflicting protected interests

A third problem appears when protected interests point in opposite directions. Minnesota’s African American Family Preservation and Child Welfare Disproportionality Act grew out of concern that black children and other covered groups are disproportionately represented in foster care. The law requires more demanding “active efforts” before covered children can be removed from their homes, while the ordinary “reasonable efforts” standard applies to children outside the covered categories.

The same protected group can be overrepresented in both state intervention and catastrophic nonintervention. One disparity suggests that black children and other covered groups are removed from their homes too often. Another suggests that black children may be underprotected: between January 2015 and April 2022, black children reportedly accounted for 26.1% of child-maltreatment deaths in Minnesota while making up only 11% of the state’s population. Disparity alone cannot tell us whether the legal standard should make removal harder or easier.

As the first author Dave Gilbert wrote for JFBT, the real question is which interest governs: family preservation or child safety. It is also whether two similarly endangered children should receive different legal protection because one belongs to a disproportionately represented group and the other does not.

Problem 4: The causal problem

A disparity may be real and urgent while still pointing toward the wrong remedy. We co-authored an essay on the widely circulated claim that black newborns were more likely to survive when treated by black physicians rather than white physicians.

That claim was treated as evidence for race-conscious, racially concordant medical care. A later reanalysis found that the apparent racial-concordance effect disappeared once very low birth weight was taken into account. The result was driven by the fact that white physicians were more likely to treat very-low-birth-weight black newborns, often in NICU settings, where those babies were already at much higher risk of dying.

The real disparity was tied to very low birth weight, not the physician’s race. The remedy should have aimed at the conditions producing very low birth weight, not racial matching between doctors and infants. Race may appear in the data because it is standing in for poverty, geography, prenatal care, insurance status, hospital quality, trust, access, or some other material mechanism. If we stop at the racial disparity and immediately design a race-conscious remedy, we may miss the intervention that would actually help.

The first author Dave Gilbert, in a JFBT essay titled “Inequality Has Become Its Own Explanation,” made a parallel point through Lee Jussim’s review of the discrimination literature, including classic work by Kunda and Thagard on how individuating information can displace stereotype-based judgments, as well as field evidence such as Nødtvedt et al.’s Airbnb study. Jussim’s point is that when evaluators have and actually attend to relevant individuating information, they tend to rely far less on demographic stereotypes. The practical remedy is to design decisions so employers have to look at specific, job-relevant evidence about the person in front of them: qualifications, experience, performance, and demonstrated skills.

If the problem is discriminatory decision-making, the best remedy may be to make institutions see people more clearly as individuals, not make them sort people more aggressively by race.

This brings us back to the deeper disagreement. If liberal equal protection is a genuine moral constraint, remedies have to be tied to particular injuries, particular institutions, and particular people. If persistent protected-group disparities are treated as proof that liberal neutrality itself is the problem, the argument moves in a different direction. Richard Delgado and Jean Stefancic describe Critical Race Theory as questioning “the very foundations of the liberal order.” And in Delgado’s “Rodrigo’s Seventh Chronicle,” Rodrigo states the premise more starkly: “Liberal democracy and racial subordination go hand in hand.”

If that is the premise, race-neutral law looks less like a genuine moral constraint than a mask for domination. Not everyone who accepts this explanatory frame endorses the full remedial program that can follow from it. But the stronger version of the argument tends to lead toward much broader remedial politics. Kendi makes that remedial implication explicit when he writes that “the only remedy to racist discrimination is antiracist discrimination.”

In our essay on CRT for JFBT, we discussed Cheryl Harris’s “Whiteness as Property,” where Harris looks to South Africa as a broader model of affirmative action: not merely admissions or employment preferences, but redistribution of property and power, including land and housing, with property rights subordinated to a societal requirement of affirmative action.

If that is where the argument goes, we have arrived at an honest first-principles disagreement. The disagreement is ultimately over whether the law best preserves human dignity by striving to treat individuals equally, under the liberal assumption that a person is a person is a person, or whether it best preserves dignity by recognizing people as members of differing racial and other identity groups, to which it extends discrete, tailored packages of rights.

The recent arc from Students for Fair Admissions to Louisiana v. Callais represents a rejection of the latter principle in favor of the former. Both cases narrow the space in which race-conscious remedies can be defended as temporary departures from ordinary equal-treatment principles. That makes the stopping-rule question unavoidable. If race-conscious law remains justified until anti-black racism has disappeared, then it is no longer merely a temporary exception. It is the triumph of the identity-based principle: permanent race-conscious law as a standing feature of American justice.

The deeper question is what follows once disparity is treated as evidence of continuing racism. Does justice require a comprehensive group-remedial order, in which law keeps reallocating benefits and burdens in pursuit of proportional outcomes? Or does equal protection remain a limiting principle even when the underlying injustice is real? Those questions cannot be answered by pointing to disparity alone. Which disparities trigger the remedy? Which groups and subgroups count? Which institutions are responsible? Who receives the preference? Who bears the cost? What is the stopping rule? And what happens to individual equal protection while the remedy is in place?

History does not design the remedy for us. Even if we stipulate that every disparity at issue is downstream from historical racism (or sexism, or some other constructed injustice), a serious moral and constitutional question remains about how a liberal democracy should respond.

Our view is that equal protection still matters as a moral and constitutional constraint: race-neutral law where possible, narrowly tailored remedies where necessary, and a strong presumption against treating individual citizens as instruments for correcting group disproportions.

Rejecting that broader group-remedial order amounts neither to scientific racism nor to indifference to injustice. It is a different account of justice: one that treats the individual person, not the racial group, as the basic unit of legal concern.

---

Dave Gilbert is a technologist and former visiting assistant professor of communication with interests at the intersection of technology, culture, and identity. Originally from rural Tennessee, he now lives in northern California with his photographer girlfriend and their four cats. Dave finds his greatest joy in bicycling through Big Sur and camping in the Mojave Desert. He is a founding member of Free Black Thought. He shares his views on X here.

Jake Mackey is Associate Professor of Classics at a small liberal arts college in Los Angeles, where he teaches Greek and Latin languages and literatures, their transformative reception by African-American writers, and their enduring relevance for our republic. He is faculty advisor for his college’s student Persuasion club, which provides a rare space for the free exchange of ideas on campus. He grew up between Austin, TX, and a small village in Kerala, in south India. The darker side of his experience in India—growing up in a cult—is captured in this film. He is planning a memoir about the brighter side. He is the author of Belief and Cult: Rethinking Roman Religion (Princeton University Press, 2022) and he is co-editor and co-author with Lawrence Eppard (Director of the Connors Institute and host of the Utterly Moderate Podcast) and Lee Jussim of The Poisoning of the American Mind (George Mason University Press, 2024) and of the forthcoming Virtuous Nonsense: Progressive America’s Epistemic Crisis (Pitchstone Publishing, 2026). He is co-director of the Connors Institute, a member of the Board of Advisors to the Prohuman Foundation, a co-founder of Free Black Thought, and an editor of the Journal of Free Black Thought.