Holistic Affirmative Action

The debate over affirmative action is one that has managed to consistently permeate the American political landscape. Particularly, the U.S. Supreme Court’s decision in Students for Fair Admissions, Inc. v. President & Fellows of Harvard College (2023) to ban race-based discrimination in college admissions has reinvigorated debate over the moral permissibility of affirmative action policies. Affirmative action has remained such a contentious issue primarily because its implementation suggests principles that appear incompatible with conventional understandings of fairness, such as by enacting deliberate discrimination. Proponents of affirmative action tend to argue that it is simply “leveling the playing field” and ensuring equal access to goods that certain groups have an unfair advantage in accessing. However, critics are concerned that, in many cases, affirmative action can affront valuable meritocratic principles and disproportionately benefit the least disadvantaged members of a group. In this paper, I will explore which aspects of affirmative action are necessary to build towards a just society, and why our current conception of affirmative action fails to uphold them consistent with broader principles of fairness. I will argue that a form of affirmative action, called “Holistic Selection,” which contextualizes candidates’ life experiences when evaluating their applications, provides crucial support for disadvantaged individuals without violating the principles of a fair society.

First, we must define affirmative action. As it is used colloquially, the term “affirmative action” lumps together various policies with a common goal (giving preferential treatment to disadvantaged groups). However, since every affirmative action policy has a different execution and thus effect, they are philosophically distinct and, as such, cannot be evaluated collectively. Robert Taylor (2009) develops a useful categorization system for classifying affirmative action policies that are operationally similar. The first category, Formal Equality of Opportunity, requires the elimination of legal or private discrimination against disadvantaged groups. The second category, Aggressive Formal Equality of Opportunity, expands on the first, requiring sensitivity training and outreach efforts to ensure the impartiality of the selection process. The third category, Compensating Support, provides tangible social support or resources (e.g. special training programs, financial backing) to disadvantaged groups so that they can compete more effectively for positions. The fourth category, Soft Quotas, implements compensatory discrimination to aid disadvantaged groups, and the fifth, Hard Quotas, requires a proportion of the selected candidates to be members of a given disadvantaged group. The first three categories, which I will refer to as Preferential Support, aim to equalize opportunities by reducing disadvantages, effectively removes “the weights from the legs of participants in a race” (Taylor 492). The last two categories, which I will refer to as Preferential Selection, aim to equalize outcomes by altering competitive processes to favor members of disadvantaged groups.

While Preferential Support does not directly alter the outcomes of a given applicant, Preferential Selection intentionally rejects procedural neutrality. This influences discriminative processes in ways that may not be fair. However, even though Preferential Support avoids the objections commonly raised against Preferential Selection, it does not seem sufficient for a just society. Preferential Support aims to eliminate disadvantages by reducing undue burdens, but it does not remedy the burdens that candidates have already had to overcome. In order to determine whether we need something beyond Preferential Support, we must evaluate the three major arguments made by proponents of affirmative action: the compensation argument, the diversity argument, and the equality of opportunity argument. 

According to the compensation argument, members of a historically disadvantaged group are owed compensation for the unjust treatment of their forebears. This argument is quite clear in the individual case: if A steals B’s sewing machine, A is obligated to compensate B for the value of the sewing machine. Likewise, a group that is unjustly harmed is entitled to compensation for the harm done. However, Pojman (1998) argues that it is not clear whether B is owed the full fruits of the stolen good. Say A uses the sewing machine to build a million-dollar quilt business. Afterwards, A returns the sewing machine to B, but it is not clear whether B has any claim over the quilt business. Intuitively, B is owed at least the opportunity cost of not having the sewing machine, but who knows what B could have earned if they had access to the sewing machine. It was A that transformed the sewing machine into the successful quilt business, but it would not have been possible without B’s sewing machine. So, the compensation argument establishes that disadvantaged groups are entitled to compensation for systemic injustice, but the extent of the owed compensation is ultimately unclear.

The diversity argument is more pragmatic: studies have shown that a diverse environment increases productivity (Hunt et al. 2023), belonging and motivation (McKinsey 2025), and trust (Bergmann 1998). Since these factors are components that contribute to wellbeing, we have a prima facie imperative to increase them. Thus, it is valuable to cultivate a diverse environment, which means we must provide methods for members of disadvantaged groups to enter homogeneous environments. Additionally, by increasing the representation of disadvantaged groups in selective, traditionally homogeneous environments, we are securing role models for other members of that group, which is essential for eliminating disadvantages (Bengtson 2025). However, the diversity argument does not specify the method for which diversity should be achieved, or how aggressive our approach need be.

Finally, the equality of opportunity argument poses that everyone should have “equal access to the goods which society distributes competitively” (Sher 2002). Since everyone deserves equal moral consideration, it would be unfair for a certain group to have diminished access to a good because of their group membership, especially if that membership is due to factors beyond the individual’s agency. Opportunity is one of those goods. Since people are not able to achieve agency without opportunity, depriving someone of equal access to opportunity is wrong. Since this unequal access to opportunity limits people’s agency, the government must implement policies that provide equal opportunity to members of groups whose access to that opportunity is systemically inhibited. This argument gives us a clear extent to which the government should implement affirmative action: it must ensure that everyone has equal access to opportunities.

However, the systemic denial of opportunity has a cascading effect whereby the denial of an early opportunity also denies the opportunities that would follow. For example, a poor high school education that does not prepare someone for a college education also denies the opportunity of working a job in, say, high finance and all the opportunities that job entails. Thus, it seems that simply rectifying current barriers through Preferential Support is not enough. While these measures ensure that opportunities are distributed equally in the future, they do not address the accumulated deficits caused by the denial of past opportunities. If a person has been deprived of the ability to cultivate the social and educational capital required to be a competitive candidate for an opportunity, merely offering the social support to now develop that capital does not put them on equal footing with another candidate who was able to do this years earlier. Thus, the equality of opportunity argument entails that Preferential Support is not sufficient for a just society. However, implementing Preferential Selection comes at a significant ethical cost. It is frequently objected to on the grounds of two main arguments: the merit objection and the mismatch objection.

The merit objection is comprised of three sub-arguments, each focusing on a different location of Preferential Selection interventions. The reverse discrimination argument focuses on the act of selecting someone. It claims that, in employing Preferential Selection on a competitive opportunity, we are inherently discriminating against someone. By selecting a candidate on the basis of their group membership, we are selecting against another candidate on the basis of their non-membership. Thus, this amounts to a discrimination that is objectionable, since this deems individuals as less worthy merely because of a group membership. Hence, Preferential Selection violates the rights of those who are discriminated against and thus should not be implemented. The principle objection, which focuses on the ideological effect of selection, claims that, by incorporating the group membership of a candidate into the decision-making process, Preferential Selection violates the meritocratic principles that our society is built upon. Under our meritocratic principles, a candidate has a claim to an opportunity if they have “earned it” through their qualifications. Thus, it would wrong the most qualified candidate to select another who is less qualified. The upending of these meritocratic principles would lead to “mediocrity and incompetence” (Pojman 1998), which would lead to worse outcomes for everyone. Thus, implementing Preferential Selection policies would not only be unfair to qualified candidates, but also cause harm to all members of society. Finally, the stigma objection focuses on the social aspect of selection. Since Preferential Selection may select candidates with fewer on-paper qualifications, it carries a social stigma that unfairly burdens recipients. This stigma is two-fold; members of non-disadvantaged groups who lost out on an opportunity may resent recipients of Preferential Selection (based on meritocratic principles), and members of disadvantaged groups may damage their own self-esteem wondering if they deserve their position or if they only got it because of Preferential Selection. Thus, it seems that implementing Preferential Selection may harm members of disadvantaged groups by attaching a social stigma to them. 

The mismatch objection, instead, focuses on who receives the benefits and bears the costs of Preferential Selection. It argues that Preferential Selection offers the most advantage to the best-off members of a disadvantaged group: the people who need it the least. An already-competitive member of a disadvantaged group will receive much larger benefits from Preferential Selection than a member who is not even in the place to apply. Additionally, the mismatch objection argues that it will be the worst-off in the advantaged group who will bear the cost of Preferential Selection. For example, class-agnostic Preferential Selection could place an affluent member of a disadvantaged group in a better competitive position than a poor member of an advantaged group with equal qualifications, even if the affluent member had access to tutors and faced fewer disadvantages. Hence, the mismatch objection establishes that Preferential Selection does not fairly distribute opportunities.

Thus, we need something beyond Preferential Support, but we cannot accept Preferential Selection. I propose a new framing of affirmative action called Holistic Selection, which uses crucially different language to avoid the objections of Preferential Selection while still expanding upon the provision of equal opportunities beyond what Preferential Support would. Holistic Selection takes into consideration the life experiences of an individual candidate when evaluating their application. Holistic hiring methods don’t aim to select the most qualified candidate, the candidate with the strongest credentials—rather, they seek the most apt candidate, the candidate who would succeed the most at the role. Examining a candidate from a qualifications lens is purely backwards looking, whereas evaluating a candidate’s aptitude involves predicting their future performance. It seems clear that a rational institution would prefer the most apt candidate anyway, since they would provide the most value to the institution, and yet, most companies primarily screen candidates based on their qualifications. The reason institutions use qualifications to select candidates is because predicting someone’s future performance involves capturing a holistic picture of the candidate, a non-trivial task. Historically, selection pools were comprised of people with similar life experiences who have overcome very similar obstacles—the structural exclusion of women and minorities led to selection pools made of almost exclusively young, white, unmarried men (Clinton White House 2025)—in which case qualifications are a very accurate proxy for aptitude. As a result, there was virtually no incentive to invest in developing robust methods to evaluate a candidate’s aptitude. However, as Civil Rights legislature enforced the expansion of selection pools to contain people with very different life experiences who have encountered very different obstacles, qualifications become a worse proxy. In recent years, it has become clear that strict reliance on qualifications for selection is a vestigial practice that is institutionally inefficient and wrongs disadvantaged candidates. But can Holistic Selection withstand the objections against changing the selection method?

Holistic Selection is complemented by the idea of merit, with the subtle reexamination of when a candidate deserves an opportunity. Have they “earned” the opportunity through their qualifications, or by becoming the most apt candidate? Clearly, a high schooler with straight A’s has more claim for admission to a prestigious college than a high schooler with many B’s if they had access to the same resources. However, it would be reasonable to think that the high schooler with straight A’s could be a worse candidate than the high schooler with many B’s if the A-student had access to tutors while the B-student had to work a full-time job to support their family. It is certainly conceivable that the B-student, having exerted more effort to achieve their qualifications, has demonstrated their ability to succeed in similar, future situations to a greater degree than the A-student, and thus holds the claim to being selected as the best candidate under meritocratic principles. Even though the B-student has fewer credentials (i.e. is less qualified), they are more apt. Thus, merely analyzing a candidate’s credentials is not enough to uphold meritocratic principles; it is also important to consider a candidate’s life experiences. From this perspective, we can easily resolve the merit objection. Holistic Selection doesn’t discriminate on the basis of group membership, forgo meritocratic principles, or generate a stigma. 

Additionally, Holistic Selection is immune to the mismatch objection. Since it evaluates people on an individual basis, it will consider the fact that the member of a disadvantaged group was aided by their affluence and that the member of an advantaged group struggled with poverty. By considering each individual’s burdens, Holistic Selection distributes opportunities fairly.

It would be unfair to distribute opportunities on the basis of qualifications, since this would ignore the different disadvantages and obstacles individuals have had to overcome to achieve similar qualifications. We can attempt to remedy this through Preferential Support initiatives, but that would not address the accumulated deficits in opportunities caused by systemic inequality. Preferential Selection attempts to approximate life experiences by using minority group membership. It assumes all members of a systemically oppressed or disadvantaged group have experienced similar obstacles and as such, treats them similarly. However, this approach is problematic and can be objected to in terms of merit and mismatch. By changing our conception of merit and implementing Holistic Selection, we can strive for a fairer distribution of opportunity. 

Works Cited

Bengtson, Andreas. 2024. “Affirmative Action without Competition.” American Journal of Political Science, April. https://doi.org/10.1111/ajps.12864.

Brown, Elizabeth R., Curtis E. Phills, Joshua Kahn, and Sadana Mukundan. 2025. “Feeling a Sense of Belonging Is Associated with More Motivation within Organizations That Value Diversity and Equity.” Scientific Reports 15 (1). https://doi.org/10.1038/s41598-025-04456-9.

Clinton White House. “Affirmative Action: History and Rationale.” Archives.gov, 2025, clintonwhitehouse4.archives.gov/WH/EOP/OP/html/aa/aa02.html.

McKinsey. 2023. “Diversity Matters Even More: The Case for Holistic Impact.” McKinsey & Company. December 5, 2023. https://www.mckinsey.com/featured-insights/diversity-and-inclusion/diversity-matters-even-more-the-case-for-holistic-impact.

Pojman, Louis P. 1998. “The Case against Affirmative Action.” International Journal of Applied Philosophy 12 (1): 97–115. https://doi.org/10.5840/ijap199812111.

Sher, George. 2002. “Justifying Reverse Discrimination in Employment.” In The Affirmative Action Debate, edited by Stephen Cahn, 58-67. New York: Routledge.

Taylor, Robert S. 2009. “Rawlsian Affirmative Action.” Ethics 119 (3): 476–506. https://doi.org/10.1086/598170.