Vol. 45 No. 16 · 10 August 2023

The Race-Neutral Delusion

Randall Kennedy on the end of affirmative action

4660 words

On​ 29 June, in Students for Fair Admissions v. President and Fellows of Harvard College and the companion case Students for Fair Admissions v. University of North Carolina, the Supreme Court of the United States outlawed racial ‘affirmative action’ as it has been practised at institutions of higher education in America since the 1970s.

Affirmative action gives a boost to certain categories of applicant for scarce places at selective institutions. What constitutes a ‘boost’ will vary. It might mean advertising educational opportunities to isolated communities which, historically, have sent few students to elite institutions. It might mean using race as a tiebreaker, giving the nod to a candidate of colour over a white candidate who is similarly accomplished. It might mean selecting a candidate of colour over a white candidate who is considerably better qualified according to established criteria. The litigation involving Harvard revealed, for example, that applicants with the same academic index – a measure comprising test scores and grades – had strikingly different prospects of admission depending on race. Generally speaking, it took considerably better test scores and grades to win admission if you were white or Asian than if you were Black or Latino. This is the scenario that has caused the most controversy, in which an elite institution aids candidates of colour by putting not just a thumb on the scale but a whole fist, leading to situations in which it is clear that, typically, Black and Latino students are less strong academically when they arrive at college than their white and Asian peers.

Colleges and universities evaluate applicants according to multiple criteria, including gender, place of origin, socioeconomic disadvantage, athletic prowess, legacy status (whether the applicant is the child of an alumnus), and in some cases the amount of money the family has donated to the institution over the years. But it is the use of race as a dimension of preference which, because of tendencies deeply ingrained in American culture, has met with the most antagonism. One such tendency is white supremacism, in particular anti-Black racism. Many Americans habitually oppose any policy that they perceive as helping Blacks, especially if whites are believed to bear the costs. A second tendency is to abhor the idea of a society which, on a racial basis, promotes certain groups over others. People who feel this way may have the intuition that barring race as a factor in assessment is a simple, effective way of avoiding an illicit racial hierarchy. A third tendency – the one that issues in affirmative action – is to believe that race must be taken into account to assist subordinated racial minorities, at least for the time being, in order to redress past wrongs, counterbalance present racial inequities, better integrate marginalised groups into the American mainstream, and bring about beneficial diversities in teaching, learning and decision-making.

Affirmative action is practised by only a small proportion of colleges and universities in the US: generally, those which attract many more applicants than they have available places. Most schools are happy to accept anyone who can meet minimal entrance requirements and pay tuition. The most prestigious private and public institutions, though, are crucibles of competition. In last year’s round at Harvard, there were more than sixty thousand applicants for undergraduate spots; fewer than two thousand were admitted. At the University of North Carolina, 43,500 applicants competed for 4200 places. A similarly intense winnowing takes place at Yale, Princeton and leading public institutions such as Berkeley and the University of Michigan.

One reason so much attention is focused on the admissions policies at these schools is that they exercise an outsized influence in American society, whether the field is journalism, popular culture, medicine, law, finance or government. All of the Supreme Court justices who decided the affirmative action cases went to colleges and law schools that practised affirmative action. All but one received their law degrees from either Yale or Harvard; the outlier, Amy Coney Barrett, attended law school at Notre Dame.

For several decades, the Supreme Court had permitted institutions of higher education to use race as a factor in selecting candidates for admission. However, it mandated that race could be taken into account only in service of a pedagogical mission: to create a body of students from a wide range of backgrounds so that all can benefit from a collective diversity of skills, perspectives and experiences. Educational institutions were permitted to take factors into account that would help them establish settings in which the rural student could learn from the urban, the lover of jazz from the lover of classical, the white from the Black, and so on. Under this diversity rationale for affirmative action, any candidate’s race could theoretically count in their favour. The whiteness of an applicant to a predominantly Black institution might be a plus just as the Blackness of an applicant to a predominantly white institution might be a plus.

The diversity rationale was initially posited by the Harvard law school professor Archibald Cox. It was picked up in 1978 by the conservative Justice Lewis F. Powell in the Supreme Court’s first plenary encounter with affirmative action in Regents of the University of California v. Bakke. Powell, in a pivotal decision, embraced the diversity rationale while rejecting all other justifications for racial affirmative action. His opinion was ratified 25 years later in Grutter v. Bollinger, a case arising from admissions policies at the University of Michigan. Colleges and universities adopted the notion of ‘diversity’ enthusiastically in as much as it permitted them to continue with racial affirmative action, albeit at the cost of having to drop all reference to reparative justice, integration and other grounds that some considered more substantial than the need for diversity.

‘Diversity’ has been a triumph of labelling. Before 1978 it was an obscure hypothesis, yet over the past forty years it has become a key word in American culture, embraced not only by leading universities but by major corporations and arbiters of taste, high and low. One explanation for the ascendancy of ‘diversity’ is that it provided a ground for boosting racial minority candidates while obviating any need to revisit America’s history of racial injustice; it pointed no accusatory fingers.

The diversity rationale was also alluring because it shifted the emphasis away from the perceived deficiencies of racial minority candidates and onto their strengths. Most justifications for affirmative action accentuated the comparative weakness of racial minority candidates by defending their selection even if they were, according to conventional criteria, inferior to their competitors and even if choosing them meant sacrificing some degree of quality. On the diversity rationale, by contrast, racial minority candidates may be considered better qualified than their apparently superior white peers once the value of minority perspective, experience and voice is taken into account. Where other justifications for affirmative action seek to justify exceptions to meritocratic selection, the diversity rationale is consistent with meritocracy: it is predicated on the notion that racial minority status can be a valuable credential. It also makes the presence of racial minorities on campus (or on an editorial board or in a cabinet or on a jury) not merely a necessary expiation of past sins, but a positive good.

The diversity rationale had champions in high places, but it also had influential detractors. Some dismissed it as a pedagogical theory without solid foundation. Others believed that educational authorities were opportunistically pursuing the old reparationist agenda under cover of diversity. The late Lino Graglia, an ultra-conservative law professor, complained that throughout academia, Justice Powell’s opinion had been ‘taken as little more than an invitation to fraud’. Sanford Levinson, a progressive law professor, said that he had become increasingly dismayed by ‘the costs to intellectual honesty of the felt need to shoehorn one’s arguments into the language of “diversity”’.

In 2014, Students for Fair Admissions, a group created by Edward Blum, a conservative lawyer dedicated to eradicating positive racial discrimination, sued Harvard and the University of North Carolina, alleging that, in their zeal to increase their intake of Black and Latino students, the schools had discriminated against whites and Asians. Lower courts ruled in favour of the defendants. But when in 2021 the Supreme Court docketed the cases for further review, it was widely assumed that the justices would reverse the lower courts and invalidate affirmative action. In the past, with the aid of conservative justices, the court had, somewhat surprisingly, given the policy a reprieve. But this time affirmative action faced a reactionary court that had shown itself, as in its overruling of Roe v. Wade last year, willing to be aggressive in securing objectives long coveted on the right.

The court’s thinking is detailed in a majority opinion by Chief Justice John Roberts, concurring opinions by Clarence Thomas, Neil Gorsuch and Brett Kavanaugh, and dissents by Sonia Sotomayor and Ketanji Brown Jackson.* Roberts held that the admissions programmes at Harvard and UNC ran afoul of the equal protection clause of the Fourteenth Amendment to the US Constitution, which requires states to provide to all persons the ‘equal protection of the laws’. He acknowledged the benefits of diversity cited by the universities: that it helps create a setting conducive to training future leaders, to encouraging the robust exchange of ideas, to facilitating adaptation to an increasingly plural society, to producing new knowledge, and to breaking down stereotypes and fostering empathy. These are, Roberts said, ‘commendable goals’. But they could not be credited as ‘compelling’ because they are altogether too vague to be suitably reviewed by courts. ‘How is a court to know,’ he asked, ‘whether leaders have been adequately “trained”, whether the exchange of ideas is “robust”, or whether “new knowledge” is being developed?’ Roberts insisted that whenever racial selectivity is in question it is essential that courts be able to assess purported benefits with some measurable degree of clarity since the drawing of racial distinctions is so singularly important and volatile a matter in American life and law.

In previous affirmative action rulings, a narrowly divided court had given educational authorities considerable deference. On this occasion, Roberts’s posture was different. ‘Universities may define their missions as they see fit,’ he writes, but ‘the constitution defines ours. Courts may not license separating students on the basis of race without an exceedingly persuasive justification that is measurable and concrete enough to permit judicial review.’ For Roberts, the diversity rationale fell far short of a ‘persuasive justification’. Its vagueness, in his view, shrouded attendant vices. One was open-endedness. Justice Powell had preferred the diversity rationale to others which, he feared, might become permanent sources of justification for racial selectivity. Justice Sandra Day O’Connor, writing in Grutter in 2003, anticipated that in 25 years’ time there would no longer be any need for ‘the use of racial preferences’ in creating diverse student bodies: race-based affirmative action, she wrote, ‘must have reasonable durational limits’. In this latest judgment, Roberts was announcing that time had run out, that it was time for affirmative action to end, that the moment had come for ‘the perilous remedy of racial preferences’ to cease.

Roberts also criticised the Harvard and UNC admissions programmes for imprecision and arbitrariness in their definition of ‘race’ and classification of groups and individuals. This has always been a difficulty in the drawing up of laws pertaining to race in the US, but courts have often chosen to ignore the complexities and contradictions in the administration of racial distinctions, relying on common sense instead. Now, all of a sudden, Roberts is more exacting. Complaining that the racial categories used by Harvard and UNC are ‘opaque’, unclear, even capricious, he scolds the institutions for being inattentive to the differences between South Asian and East Asian students, for leaving the category ‘Hispanic’ undefined, and for offering virtually no guidance when it comes to classifying students from the Middle East.

A further problem, according to Roberts, is that for all the talk of the benefits of variety, the diversity rationale homogenises people. He scoffs at those who take the position, voiced by Harvard, that ‘a Black student can usually bring something that a white person cannot offer.’ That proposition, Roberts writes, is itself a ‘pernicious stereotype’.

According to the Supreme Court, by using race at all in their admissions programmes, universities become complicit with an age-old moral, social and legal evil, in that classifying people by race ‘demeans the dignity and worth’ of individuals, judging them by ‘ancestry’ instead of by their ‘own merit and essential qualities’. For Roberts, any racial selectivity in admissions is a dangerous enterprise, and illicit unless it can be justified and effectuated in a clear, measurable, narrowly tailored fashion. In the judgment of the court, racial affirmative action in civilian higher education no longer meets that standard, if it ever did.

The majority opinion contains two wrinkles. First, in a footnote, it puts aside the question of racial affirmative action at military academies ‘in light of the potentially distinct interests’ that these institutions may have. These ‘distinct interests’ might include the belief that it is imperative – for the sake of morale and legitimacy, and therefore as a matter of military preparedness – to ensure the presence of racial minorities in an officer corps overseeing ranks that are substantially filled with racial minority service-members. The court did not approve of racial affirmative action in the military; rather, it chose not to confront the issue.

Second, at the end of his opinion, Roberts assures admissions officers that ‘nothing in this opinion should be construed as prohibiting universities from considering an applicant’s discussion of how race affected his or her life, be it through discrimination, inspiration, or otherwise.’ He seems to be saying that it is permissible for applicants to discuss race and for institutions to consider whatever the applicants say, so long as the aim is strictly limited to assessing a particular candidate’s personal qualities and not the fulfilment of some broader institutional mission having to do with race relations. Wary that defenders of affirmative action might look for loopholes, Roberts warns sternly that ‘universities may not simply establish through application essays or other means the regime we hold unlawful today.’

Theworst thing about Roberts’s opinion, as I see it, is its insistence that constitutional law recognise no distinction between malign and benign racial discrimination, between derogatory and positive racial distinctions. Roberts has long asserted that actions predicated on a racial basis are presumptively ‘invidious’ in all contexts. In other words, all racial selectivity is bad, toxic, illicit. Some exceptions may be permissible – in an emergency, perhaps, to avoid some catastrophe. But in general, according to Roberts, any sorting of people that includes even a partial racial element is wrong.

This view ought to be rejected in favour of a position that recognises the difference between friendly and unfriendly racial discrimination. To echo Justice John Paul Stevens, who in a case in 1995 pointed out the difference between a ‘No Trespassing’ sign and a welcome mat, there is certainly a moral distinction – and there should be a legal distinction – between a sign that says ‘Blacks get out!’ and one that says ‘Blacks most welcome!’ Both are racially selective. Both invoke racial classifications. But while the former signals a wish to exclude, the latter encourages inclusion. The legal system surely ought to distinguish between these states of consciousness and their corollary social conduct. Roberts’s unwillingness to do so is ominous. The threat, as we look forward, is that his simplistic, symmetrical, formalist, sociologically unnuanced conception of ‘discrimination’ will be deployed to undo any effort aimed at helping Blacks or any other racial group, on the grounds that doing so is an illicit discrimination against anyone outside the group.

This kind of thinking has deep roots in American history. President Andrew Johnson vetoed the Civil Rights Act of 1866 because in his view it discriminated against whites and privileged Blacks. The Act bestowed citizenship on all persons born in the US and endowed all persons with the same rights as whites with regard to contracting, owning property, suing or being sued, or serving as a witness. The law was proposed because the Supreme Court had ruled in Dred Scott v. Sandford (1857) that African Americans, free or enslaved, were racially ineligible for federal citizenship, and because many states had barred African Americans and other racial minorities from enjoying even the most rudimentary civil rights. Johnson opposed the citizenship provision because it would immediately make citizens of native-born Blacks while immigrants born in Europe had to wait several years to qualify for citizenship via naturalisation (a process that was open only to white people). The proposed law, he wrote in his veto message, amounted to ‘a discrimination against large numbers of intelligent, worthy and patriotic foreigners … in favour of the negro’. He also opposed the provision giving federal protection to civil rights, complaining that it afforded ‘discriminating protection to coloured persons’. The Civil Rights Act would, he wrote, ‘establish for the security of the coloured race safeguards which go infinitely beyond any that the General Government has ever provided for the white race. In fact, the distinction of race and colour is by the bill made to operate in favour of the coloured and against the white race.’ Johnson similarly opposed the Fourteenth Amendment, which decreed that states offer to all persons the equal protection of the laws. Although Johnson’s opposition was overcome, his reaction was by no means idiosyncratic. It was preceded and followed by many decisions animated at least in part by an anxious perception that something must be amiss insofar as racial minorities, Blacks in particular, were being helped to get ahead.

Some left-liberal detractors of racial affirmative action argue that the Supreme Court ruling will make room for the creation of admissions policies that are more attentive to redistribution along class lines. They march under the banner of ‘class, not race’. Their most prominent spokesman is Richard Kahlenberg, who testified as an expert witness for the plaintiffs in Students for Fair Admissions. He argues that the affirmative action regime now invalidated by the court did little to boost candidates from poor households while giving undue assistance to racial minorities from households higher on the socioeconomic ladder. A well-designed ‘class, not race’ affirmative action programme would, in Kahlenberg’s view, do three things at once. It would improve educational opportunities for candidates unfairly burdened by socioeconomic inequities (for example, inferior primary and secondary schooling; parents who didn’t attend university; isolation from networks rich in social capital). It would result in a substantial amount of racial diversity since racial minorities, Blacks in particular, are disproportionately represented in the ranks of the poor. And it would produce these benefits while avoiding the backlash that constrained racial affirmative action from the outset, and has now brought about its end.

The​ ‘class, not race’ position is mistaken. First, the laudable aim of improving educational opportunities for students from poor families is by no means precluded by a continuing effort to address the distinct problems wrought by racial wrongs, historical and contemporary. There are compelling reasons to give poor whites a boost in the competition for elite educational resources. But variations on those same reasons also support the continuation of sensibly designed racial affirmative action. Second, ‘class, not race’ advocates assert that socioeconomic affirmative action will produce an acceptable amount of racial diversity. Whether or not one agrees will depend on the criteria used to assess socioeconomic class and on what degree of racial minority presence is considered ‘acceptable’. It is virtually beyond dispute, however, that a selection scheme focused wholly on class, leaving race out of consideration, will diminish the number of Black and Latino students attaining admission to elite institutions. Poor whites constitute a large reservoir of competitors who will often be better prepared and have better credentials than their Black peers, including Blacks on higher rungs of the socioeconomic ladder.

Some proponents of ‘class, not race’ seem to believe that the conservatives who bitterly resisted racial affirmative action for so many years will now happily support ‘race neutral’ affirmative action for the disadvantaged. That is naive. Some conservatives might support race-blind help for the needy, but many of them are merely opportunistic in their use of egalitarian arguments against racial affirmative action. Otherwise they evince no sympathy whatever for the poor. Worse, the most fervent opponents of racial affirmative action will now come after socioeconomic affirmative action programmes, alleging that in reality they are racial affirmative action in disguise. Depending on the ideological winds blowing through the federal judiciary, these opponents may gain strong legal support. If all racial discrimination is equally objectionable, they might argue, and if it is unconstitutional for officials purposefully to disfavour racial minorities by indirect means, it must also be unconstitutional to purposefully favour racial minorities by indirect means. Under the conception of ‘discrimination’ that the Roberts court often embraces, a programme aiming to admit more poor students is permissible, but a programme aiming to increase the numbers of Black students by increasing the number of poor students is not. Currently, the elite Thomas Jefferson High School for Science and Technology in Virginia is ensnared in litigation over its selection process. A district court ruled in favour of plaintiffs who claimed that nominally ‘race neutral’ changes the school had made to its selection scheme were prompted by the ‘race conscious’ aim of admitting more Black students (even if that entailed lowering the number of Asian Americans admitted). The court of appeals reversed the ruling, saying that the plaintiffs had failed to establish that school officials had acted with discriminatory intent. That resolution, however, is tenuous; judges gripped by Roberts’s conception of discrimination may well strike down such ‘race conscious, race neutral’ selection schemes.

It wouldn’t be surprising if, in the wake of the Supreme Court’s ruling, some leaders of educational institutions did opt to deploy ‘race neutral’ workarounds, with the unstated aim of admitting as many racial minority candidates as prudently possible. This would be an ironic analogue to the resistance mounted against desegregation in the wake of Brown v. Board of Education of Topeka (1954). Richard Rothstein, a co-author of Just Action: How to Challenge Segregation Enacted under the Colour of Law, calls for defiance and urges colleges and universities to pay whatever costs follow. Justice Sotomayor, in her dissent in the Harvard case, doesn’t expressly call for defiance but comes close to doing so, writing that ‘universities can and should continue to use all available tools to meet society’s needs for diversity in education.’ Mocking what she refers to as her colleagues’ ‘unjustified exercise of power’, Sotomayor anticipates that their ruling ‘will serve only to highlight the court’s own impotence in the face of an America whose cries for equality resound’. Enemies of racial affirmative action are, however, experienced, well-resourced and fuelled by ideological conviction. It doesn’t hurt their cause that many Americans disapprove of what they perceive as an arrogant and mendacious higher education establishment. Right-wing jurists antagonistic to ‘woke’ academia will be on the lookout for dodges in admissions policies; Students for Fair Admissions recently sent a letter to 150 institutions warning them against seeking to evade the Supreme Court’s rulings. Many Trump-appointed judges would like nothing better than to rip into a liberal elite college for ‘discriminating’ against whites (or Asians) and then lying about it.

The struggle over racial affirmative action has involved two related but discrete questions which are often conflated. The first is whether affirmative action is permissible. The Supreme Court has determined that as a matter of federal constitutional law it is not. That is a question that should have been answered differently. Instead of imposing a uniform, centralising standard from on high, the court should have displayed judicial restraint, and permitted institutions to make up their own minds whether or not to engage in positive racial discrimination. The court should remove invidious racial discrimination from regular democratic politics. If a decision or policy or law is propelled by an aim to exclude or otherwise injure a group because of its racial identity, the court should declare that action unconstitutional, and hence outside the boundaries of regular political dispute. However, with respect to positive discrimination which adversely affects groups only collaterally, the court ought to be more deferential to democratic politics, more tolerant of local experimentation, and more willing to permit a variety of answers to the question as to whether, as a matter of policy, racial affirmative action should be practised. That is a question about which serious, open-minded people of various ideological stripes disagree. Before the Supreme Court prohibited affirmative action nationally, nine states, including liberal California, had already done away with it.

Racial affirmative action does have major drawbacks. For one thing, it puts a pall over the racial minorities assumed to be its beneficiaries: hence the frequency with which it’s asked whether a given candidate – not just for a place at university, but for a job or, for that matter, a position on the Supreme Court – is unequivocally excellent or merely the best Black person available? What’s more, racial affirmative action in practice has encouraged a destructive denialism. Some of its champions claim that positive discrimination doesn’t involve giving a boost to racial minorities but credits them with a ‘merit’ consisting in nothing more than being racially different from white candidates. Others claim that affirmative action involves no lowering of standards; they then object to the use of various competitive criteria (grades or test scores) as evidence. They may also suppress information that shows dramatic gaps between beneficiaries and non-beneficiaries. Cover-ups of this sort conceal underlying deficiencies that warrant attention. Students who are behind need to be aware of their situation if they are to catch up. Affirmative action double-talk is demoralising and has led to cynicism. Many Black students respond angrily when it is assumed they have benefited from affirmative action. They know that the existence of the policy implies a need for assistance, and although they appreciate the help they also resent the inferences that observers draw.

Affirmative action imposes dilemmas that are avoidable only through delusion. One of its most influential backers, Barack Obama, once said that racial affirmative action programmes, ‘when properly structured, can open up opportunities otherwise closed to qualified minorities without diminishing opportunities for white students’. But how is that possible? If university places are constrained by scarcity, as they are at all selective institutions, special boosts for Blacks and Latinos will necessarily diminish opportunities available to others. That does not mean that affirmative action is invidiously discriminating against others in the sense that it purposefully excludes them because of their race. But it does mean that in an effort to achieve commendable aims, an inevitable consequence of affirmative action will be adversely to affect people who are not Black or Latino.

Affirmative action was, on balance, a useful intervention, which increased the presence of Blacks and Latinos at elite institutions of higher education and, by extension, increased their representation in leadership positions throughout the American establishment. Its termination at the hands of a reactionary Supreme Court is a significant setback. That said, we would do well to remember that the racial affirmative action regime had steep costs and sharp limits. There are good reasons why some progressives tolerated racial affirmative action without feeling much enthusiasm for it, or are even quietly pleased that it has ended, hopeful that something better, more equitable and far-reaching can be put in its place. More disturbing than the loss of affirmative action itself are the circumstances that led to its demise. That it was met with such a fierce and sustained backlash despite its relatively modest scope is an ominous portent. Racial hierarchy in the US remains a huge problem. The Supreme Court’s suppression of affirmative action calls into sharp question whether America has within it the moral and political resources to address satisfactorily that long-festering injustice.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.


Vol. 45 No. 18 · 21 September 2023

In proposing random selection to solve the problem of entrance to elite institutions, Donald Gillies has company across the globe (Letters, 7 September). The policy was twice adopted by Japan’s Ministry of Education in the 1920s and 1930s in an attempt to tackle the problem of ‘exam hell’ among aspiring secondary school entrants. Advocates argued that after admitting the best and failing the worst applicants, random selection was the fairest and most efficient way to select among the mass of middling candidates. In his recent book The Tyranny of Merit, Michael Sandel makes much the same argument in seeking a cure for the ills caused by contemporary educational competition.

Peter Cave
University of Manchester

Vol. 45 No. 19 · 5 October 2023

Donald Gillies addresses the challenge of selecting applicants for limited places at elite educational institutions (Letters, 7 September). The same problem crops up with regard to a wide range of scarce resources, from places at a childcare centre to tickets for a West End show. Gillies advocates random selection (assuming an established base level of eligibility) and gives the example of admissions to medical school in the Netherlands. The national system of admission to Dutch medical schools by lottery has a long and complex history. It was established in 1972, then abandoned in 2017; legislation has been passed allowing it to start up again this year. The system was never a simple ballot; essentially, it admitted a cohort of outstanding candidates, then balloted any remaining students who met a minimum threshold.

Random selection in education has been used in Britain too. A form of ballot was, for example, implemented for admissions to study medicine at Barts and the London for a period earlier this century. Conall Boyle, in Lotteries for Education: Origins, Experiences, Lessons (2010), surveys ballot systems used to manage oversubscribed school admissions, from Lancashire in the 1980s to schools in Essex, Brighton, London, Hertfordshire and Norfolk. The Lancashire ballot system survived a challenge in the High Court, and the UK Office of the Schools Adjudicator has repeatedly approved admission by ballot.

What little research there has been doesn’t show significant differences in performance or completion by students admitted at random or according to criteria. But a ballot system – even one that allows first for a cohort of ‘outstanding’ candidates – does confound aspirants’ expectations based on the socially reinforced idea of ‘merit’ and, less excusably, on a sense of entitlement derived from family, wealth and heritage. The resentment felt by those who miss out (or by their parents) has led to legal challenges, and has consistently gained enough momentum to bring an end to ballots, as happened in the Netherlands and at Barts and the London.

I use a ballot myself to decide who will be admitted to my oversubscribed law courses. Allowing for threshold competence, I literally put students’ names in a hat, and ask a colleague in a neighbouring office to carry out the draw. Unsuccessful students have never objected.

Simon Rice
University of Sydney

Vol. 45 No. 17 · 7 September 2023

Randall Kennedy, writing about affirmative action in the United States, points out that the practice was used only at elite institutions, such as Harvard, where there are many more applicants than there are places (LRB, 10 August). In the UK, too, most universities can admit the majority of applicants who meet the minimal entrance requirements, but Oxford and Cambridge in particular have many more applicants than they have places to offer. How should this problem be addressed? I would argue that the solution lies in random selection. An institution like Harvard should set minimal entrance requirements. There would of course be many more applicants than available places. The lucky candidates would then be selected by some random process. The great advantage of this method is that it is completely blind to differences in class, gender or ethnic origin. It is also less likely to breed resentment among unsuccessful candidates than other methods, which inevitably arouse suspicions of unfair discrimination of one kind or another. The use of random selection for entrance to elite institutions may strike some as unrealistic, but it has in fact been used very successfully for medical school admissions in the Netherlands for more than thirty years.

Donald Gillies
London SE21

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN


Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences