Affirmative Action: Towards a Coherent Debate

Coleman Hughes1Coleman Hughes is a fellow at the Manhattan Institute and a contributing editor at City Journal, where his writing focuses on race, public policy, and applied ethics. His writing has been featured in the New York Times, the Wall Street Journal, National Review, Quillette, The City Journal, and The Spectator. Hughes has appeared on many podcasts—including The Rubin Report, Making Sense with Sam Harris, and The Glenn Show—and also hosts his own, Conversations with Coleman. In June 2019, he testified before the U.S. Congress about slavery reparations.

* * *

A part of the series, Affirmative Action at a Crossroads.

This November, the citizens of California will vote on a proposition to remove the following words from their state constitution: “The state shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting.” These words, taken almost verbatim from the Civil Rights Act of 1964, were originally added in 1996 as part of Proposition 209—a successful effort to end racial preferences at state-funded institutions.

Although the Civil Rights Act of 1964 and Prop. 209 share similar language, the public attitude towards them could not be more different. The former is universally praised, while the latter has provoked bitter debate, falling along partisan lines, since its inception. To its supporters, Prop. 209 is an antiracist document in the same vein as the Civil Rights Act. To its critics, it is a reactionary document that has cut off the path to opportunity for millions of Blacks and Hispanics—in other words, it represents the forces that opposed the civil-rights movement.

Why is Prop. 209 the site of bitter debate despite being textually similar to a document that is universally praised? The reason stems from a fundamental disagreement over what constitutes “racism”—a disagreement that only became explicit in the years after the Civil Rights Act was passed. The disagreement is this: To some, “racism” refers to discrimination against an individual on the basis of race. To others, it refers to the end result of a process that yields unequal outcomes—where equality is understood to mean group-level parity between races. This is more than a semantic disagreement over the meaning of a word. It is a clash between competing visions for the rules that should govern multiethnic societies.

In this Essay, I will call the first view—the view that opposes racial preferences—“pro-equality.” I will call the second view, which supports racial preferences, “pro-equity.” (I have tried to choose labels that have a neutral or positive valence.)

The aim of this Essay is not to argue for any particular side (though I have published opinions on the subject). My more modest aim here is to clarify the terms of the debate. The purpose of this Essay is twofold: (1) to distinguish coherent positions on racial preferences from incoherent ones, and (2) to examine the strengths and weaknesses of each coherent position.

To begin, a coherent position on racial preferences (whether in favor or against) must accept that there is a public policy trade-off between equality and equity—between treating individuals without regard to race on the one hand and achieving equal results at the group level on the other. Getting more equality necessarily means getting less equity, and vice versa. For example, making the student body at Harvard University resemble the overall population (more equity) will require more racial discrimination against individual applicants (less equality). Conversely, ending all racial discrimination against applicants (more equality) will result in a student population that diverges more from the general population (less equity).

To deny that this trade-off exists would be to sidestep the core issue. To insist that this trade-off does not exist in every sector of American society, while true, is irrelevant for current purposes. One can hope for a day when this trade-off will no longer exist, a day when equality and equity go effortlessly hand in hand. One can even come up with policy proposals that would hasten the arrival of that day. But neither hopes nor policy proposals eliminate the trade-off as it exists today—or eliminate the need for an ethical analysis of it.

The moment we acknowledge the trade-off between equality and equity, we can eliminate one position on racial preferences as incoherent from the start: the view that we should maximize equality and equity at the same time, or put differently, that admissions officers and firms should fully prioritize treating applicants equally while also fully prioritizing group-level parity. However balanced this view may sound, it would only make sense if there were no trade-off between equity and equality to negotiate.

That leaves three coherent positions, each of which negotiates the trade-off in a different way. First, a person might be completely pro-equality—that is, they might believe that ending discrimination against individuals is categorically more important than achieving group-level parity. Second, a person might be completely pro-equity, categorically prioritizing equal outcomes over ending discrimination against individuals. Third, a person could take a middle path by saying that neither equity nor equality is categorically more important than the other. Rather than try to maximize either equity or equality, this third view settles for having less of each in exchange for having some of both.

Each position has its strengths and weaknesses. The pro-equality view is strong in that it completely opposes racial discrimination against individuals, and therefore gels effortlessly with the core principle of the civil-rights movement: judging a person by the content of their character rather than the color of their skin. As a result, a person who adopts the pro-equality view can never be put in the awkward position of defending racial discrimination in some cases (e.g., against Asian-American university applicants) while condemning it in others (e.g., against Black Americans).

The weakness of the pro-equality position is that it must accept any level of group-level disparity, no matter how starkly unequal, so long as that disparity was arrived at by means of a race-blind process. Consider a hypothetical: If a race-blind admissions process at Harvard were to yield a student body that was only 1 percent Black, someone with the pro-equality position could not fault Harvard admissions, or ask them to fix it. To be sure, they could blame systemic racism for this disparity and seek to address its root causes. But they would have to accept Harvard’s policies as not just acceptable, but morally required. Any discomfort at the near absence of Black faces on campus would have to be considered a price worth paying. Someone with the pro-equity position, by contrast, can never be put in the awkward position of seeming to justify the underrepresentation, or absence, of Black people from any particular sector of society.

The pro-equity position has its own strengths and weaknesses. A strength of the pro-equity position is that it offers a ready solution—a solution that can be implemented directly by institutions themselves—to the problem of racially unequal results. Faced with inequity, the pro-equality position may gesture to long-term public policy solutions, but it cannot give anyone at a non-government institution—a college dean or a CEO, for instance—anything to do directly and immediately. The pro-equity person, on the other hand, has a direct and an empowering message to such people: there is something concrete and measurable you can do right now—namely, increase the number of Black people you allow into your organization.

The weakness of the pro-equity position is that it must justify racial discrimination against some individuals in order to fight racial discrimination against others. On its face, this can seem like a contradiction. At minimum, the position requires a revision of the core principle of the civil-rights movement: the idea that racial discrimination is inherently evil. The pro-equity position must revise that principle and instead say that racial discrimination is contingently evil—contingent, that is, on whether such discrimination is increasing equity or reducing it. Insofar as racial discrimination is increasing equity, then not only is it not evil, it is morally required. By contrast, a pro-equality person can never be put in the awkward position of having to disagree with the principle that undergirded the civil-rights movement.

A second weakness of the pro-equity position is that its core principle—racially equal results—is applied selectively. Nobody complains about the overrepresentation of Black Americans in desirable sectors of society—such as the NBA—or their underrepresentation in undesirable metrics, such as suicide. As a result, it is open to question whether the equity principle is actually held earnestly or whether it is merely a trojan horse for a principle of Black ethnic chauvinism.

The middle-ground position—which accepts a less-than-total commitment to both equality and equity in exchange for having some degree of commitment to each—has its own strengths and weaknesses. It’s strong in that it compromises between what might be viewed as two extreme positions. As a result, it is a realistic and diplomatic option for any organization that includes both pro-equity and pro-equality individuals. Neither side can complain that its values are being completely ignored. Thus, the middle-ground position represents the spirit of compromise that allows ideologically diverse institutions to function.

Yet what the middle-ground position gains in pragmatism it loses in persuasiveness. The reason is that it dodges the most important question: What is “racism”? Is it racially discriminating against individuals? Or is it the net result of processes that yield unequal results? That is the fundamental question motivating the entire disagreement, to which the middle-ground position replies with a confused shrug. With a concept as emotionally and morally charged as racism, it may be seen as a cop-out to split the difference between positions that are fundamentally at odds with one another. To an unsympathetic observer, the middle-ground position amounts to saying something like this: “racism refers to racial discrimination against individuals but also refers to processes that yield unequal results, even though equal results can only be achieved by discriminating, which we have already agreed is racist.” At minimum, that stance is confusing; at most, it is unintelligible.

It is hopeless to expect that the affirmative-action debate will be settled anytime soon. But as the citizens of California re-evaluate Prop. 209, we can hold out hope that the debate will be held honestly, with parties on all sides being forced to acknowledge, and defend, the weak spots in their positions.

* * *

Coleman Hughes is a fellow at the Manhattan Institute and a contributing editor at City Journal, where his writing focuses on race, public policy, and applied ethics. His writing has been featured in the New York Times, the Wall Street Journal, National Review, Quillette, The City Journal, and The Spectator. Hughes has appeared on many podcasts—including The Rubin Report, Making Sense with Sam Harris, and The Glenn Show—and also hosts his own, Conversations with Coleman. In June 2019, he testified before the U.S. Congress about slavery reparations.

* * *

Click here to return to the Affirmative Action at a Crossroads series main page.

Authors

9 thoughts on “Affirmative Action: Towards a Coherent Debate

  1. Sorry but it is not down to people who believe racism is between individuals or people who believe in equality of outcome. This is incredibly disingenuous and you start your essay off by misrepresenting what you spend the rest of your essay railing against. You would be hard pressed to find anyone fighting racism in this country that truly believes in equality of outcome, that is a made up phenomena created by right wing think tanks to tell their exclusively republican subscribers what ‘the left’ thinks. You clearly have no participated in meaningful debate with anyone anti-racist and instead insist on telling us what they mean. You can do much better than this, you have a strong academic background but haven’t put it to use yet on this topic. I think you should stick to philosophy.

    1. Hm, nobody? What about Ibram X. Kendi? Quote (“How to be an Antiracist”, p. 17-19):

      “Racism is a marriage of racist policies and racist ideas that produces and normalizes racist inequities. […] Racial inequity is when two or more racial groups are not standing on approximately equal footing. […] A racist policy is any measure that produces or sustains racial inequity between racial groups. An antiracist policy is any measure that produces or sustains racial equity between racial groups […]. “Racial discrimination” is an immediate and visible manifestation of an underlying racial policy. When someone discriminates against a person in a racial group, they are carrying out a policy or taking advantage of the lack of a protective policy. We all have the power to discriminate. Only a few of us have the power to make policy. Focusing on “racial discrimination” takes our eyes off the central agents of racism: racist policy and racist policymakers, or what I call racist power. […] [I]f racial discrimination is defined as treating, considering, or making a distinction in favor or against an individual based on that person’s race, then racial discrimination is not inherently racist. The defining question is whether the discrimination is creating equity or inequity. If discrimination is creating equity, then it is antiracist. If discrimination is creating inequity, then it is racist” (Any typos are mine)

      Care to explain how Mr. Hughes summary of this position is a straw man?

      1. Shocked to see anyone have an issue with anything Kendi wrote here even with you parsing his sentences together. This is not an example of what you are proposing, my first comment was clear of the false assertion Coleman Hughes made in this piece and you have not provided a rebuttal that suits your argument against mine. If you have a comment on the actual piece and not my comment, well feel free to leave it just like I did.

  2. Thanks for this. It is clarifying in a messy area. Too often the Manhattan Institute buries the most revealing stories under the weight of its wealthy funder.

  3. Thanks very much for this clarifying piece. Clearing up your assumptions and definitions is very helpful to understand how you view the debate. It also opens them up to attack more easily then hiding them behind postmodern verbiage. Too few are willing to risk that in this debate.

    Here is an aspect I think is missing from your overview:

    Policies may satisfy the equality-criterion on the surface and yet discriminate against specific racial groups by taking aim at proxy-properties that are disproportionately common among those groups. One example is the infamous difference in the severity of punishment for the possession of crack and cocaine. Voter-ID laws might be considered to be another one. Likewise, one might view university entrance tests which increasingly focus on “extracurriculars” as discriminatory against Asian applicants.

    In fact I think it is rather trivial to come up with obviously racially discriminatory policies that pass the equality test.

    I want to use this to steel man the middleground position. Let’s associate both the equality criterion and the equity criterion with a test. A policy fails the equality test if it discriminates by race on an individual base. A policy fails the equity test (Andrew Sullivan calls this the “Kendi test”) if it maintains or increases disparity between the racial groups.

    I think a more tenable version of the middle ground position is one that uses the equality test but in addition applies the following test: a policy fails the discriminatory proxy test if it is based on a criterion that is disproportionately distributed among racial groups and that disparity is at least partially due to racial discrimination and is difficult to change on the individual level.

    For example, the SATs arguably aim at an applicant’s IQ. There are presently IQ differences between racial groups which the majority of scholars views as the consequence of environmental differences caused by systemic racism. So SATs fail the first criterion. Secondly, IQ is a property that is known to be largely immutable among adults. So SATs also fail the second criterion.

    Obviously, such an assessment can be very difficult and open to reasonable disagreement. Is it fair to consider an applicant’s criminal record when members of her racial group more frequently have a criminal record than the average? But the difficulty of such assessments does not mean that they should not be attempted.

    Conversely, one might come up with a middle-ground position that combines the equity test with some other test geared at equality. I will not elaborate this here but suspect that it would likewise yield a coherent middle ground position.

  4. As I understand, ‘racism’ means ‘systematic prioritization, in everyday relations and in institutions, of one particular group over another/others’; here, historically, it’s been whites over pretty much everyone else. Importantly, racism has served the purpose of dividing working people, so that they may worry more amongst themselves about racial hierarchies than the fact that they are all, financially speaking, in the same boat—they are, more or less, slaves in an unjust economic system. So, it’s not simply ‘discrimination’—it is a steady building up of societies that reward some groups and alienate others. Poverty is most highly correlated with certain of these groups; but it’s the enemy of all.

    The subsets most impacted—most likely, statistically, to suffer from the effective illness of poverty—are Blacks and indigenous people. Affirmative action is an attempt to open to these groups institutions whose evaluative and admissions processes have historically favored achievement/educational attainment of a certain kind—of a certain socioeconomic status that might be used as a proxy for whiteness.

    I don’t know that the idea that we may, at some point, have to consider choosing between a poor white person and poor Black person….is a reason to do away with affirmative action policies, all. Certainly, there will be cases in which we must consider personal circumstances other than race, and work things out with parties involved, and perhaps compromise. It needn’t be strictly algorithmic. The disjunction presented here….looks like a case of letting some ideal obscure the good.

    1. By which I meant, ultimately: I don’t think that the above presents a helpful framing of the issue at hand, or of the choices involved.

      1. As in….let’s not mess with affirmative action. It helps people, and is a net benefit to society. I have yet to read any of Mr. Hughes’s op-eds presenting good evidence of the opposite effect. Doing away with affirmative action would be a really short-sighted, and frankly regressive, move.

  5. I think the fundamental problem is that nobody wants to be told they are an affirmative action hire.

    “Well Tyrone, for political/social/cultural/market reasons our institution has decided that having X% of your kind around achieves certain self interested goals for our organization, so just sit there taking minutes at the meeting and please try not to f*** up anyones computer code while you enjoy this unearned sinecure. If you get into too much trouble just have Lee Kim check your math. See you at happy hour.”

    Telling the people running K-12 or pre-K or social workers or whoever that its their fault doesn’t do much either. We’ve tried everything and the IQ scores are stuck where they are.

Leave a Reply