Harvard Law Review Harvard Law Review Harvard Law Review

Affirmative Action

The Harvard Plan That Failed Asian Americans

In November 2014, Students for Fair Admissions (SFFA) filed a complaint against Harvard College in federal district court.1×1. Complaint, Students for Fair Admissions, Inc. v. President & Fellows of Harvard Coll., 308 F.R.D. 39 (D. Mass. 2015) (No. 14-cv-14176) [hereinafter SFFA Complaint]. On the same day, SFFA also filed a complaint against the University of North Carolina, although it did not specifically focus on Asian Americans. Complaint, Students for Fair Admissions Inc. v. Univ. of N.C., 319 F.R.D. 490 (M.D.N.C. 2017) (No. 1:14-cv-954). SFFA claims that Harvard discriminates against Asian Americans by holding them to higher admissions standards than any other racial group, including whites.2×2. See SFFA Complaint, supra note 1, at 43–53. Because Harvard is an institution that accepts federal funds, it “violates Title VI when it engages in racial or ethnic discrimination [prohibited by] the Equal Protection Clause of the Fourteenth Amendment.”3×3. Id. at 94. This Note assumes that the Title VI of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000d to 2000d-7 (2012), and equal protection analyses are coextensive — private institutions that receive federal funding, like Harvard University, are subject to the same requirements as public institutions like the University of Michigan. See Gratz v. Bollinger, 539 U.S. 244, 276 n.23 (2003) (“We have explained that discrimination that violates the Equal Protection Clause of the Fourteenth Amendment committed by an institution that accepts federal funds also constitutes a violation of Title VI.”). SFFA argues that Harvard’s race-based admissions program is impermissible under Supreme Court precedent. Beyond that, SFFA urges “the outright prohibition of racial preferences in university admissions — period.”4×4. SFFA Complaint, supra note 1, at 6. Ironically, the allegedly discriminatory “Harvard Plan” is the very one that Justice Powell held up as a model in Regents of the University of California v. Bakke5×5. 438 U.S. 265, 316–17 (1978) (Powell, J.). and the admissions program that American universities have emulated for decades.6×6. See Susan Welch & John Gruhl, Affirmative Action and Minority Enrollments in Medical and Law Schools 63 (1998) (describing university responses to Bakke).

For the last thirty years, Asian admissions has been a hot topic.7×7. This Note uses the terms “Asian” and “Asian American” interchangeably. Furthermore, this Note discusses Asian Americans as a whole but recognizes that different subgroups may be affected differently by affirmative action policies. Rumors, and sometimes concrete evidence, of racial discrimination and ceiling quotas fuel the controversy. The Princeton Review advises applicants that “an Asian-sounding surname” may be a disadvantage.8×8. The Princeton Review, Cracking College Admissions 174 (John Katzman et al. eds., 2d ed. 2004). Asian parents, many of whom immigrated to the United States for their children’s education, have staged protests outside of the Supreme Court.9×9. Emil Guillermo, As SCOTUS Hears Affirmative Action Arguments, Asian American Advocates Weigh In, NBC News (Dec. 10, 2015, 3:09 PM), https://www.nbcnews.com/news/asian-america/scotus-hears-affirmative-action-arguments-asian-american-advocates-weigh-n477806 [https://perma.cc/3LDF-GW3d]. And Pacific Islanders — traditionally classified with Asians under the label “Asian Pacific American” — have become so concerned about an admissions handicap that they have withdrawn from the Asian category to group themselves with Native Hawaiians.10×10. See Daniel Golden, The Price of Admission 204 (2006). Within the Asian American community, the topic of affirmative action can be divisive. While a vocal minority wants completely race-blind admissions, many more support affirmative action programs for underrepresented minorities.11×11. Nancy Leong, The Misuse of Asian Americans in the Affirmative Action Debate, 64 UCLA L. Rev. Discourse 90, 93 (2016). Their objection is to “negative action,” or unequal treatment in comparison to the white majority.12×12. See William C. Kidder, Essay, Negative Action Versus Affirmative Action: Asian Pacific Americans Are Still Caught in the Crossfire, 11 Mich. J. Race & L. 605, 605–06 (2006).

This Note proceeds in three Parts. Part I considers the arguments that SFFA and other Asian American organizations have made, and provides a brief history of previous investigations and legal actions. Part II analyzes the Supreme Court’s diversity rationale and argues that its development has been harmful for Asian students. Under the Court’s affirmative action decisions, litigants like SFFA face an uphill battle. Part III returns to where it all started: with Justice Powell in Bakke. Through an examination of Justice Powell’s notes and correspondence, this Note argues that his endorsement of the Harvard Plan was shortsighted and unwise. Even in 1978, the dangers of the diversity rationale should have been apparent.

I. Asian American Challenges to Admissions Policies

A. The Evidence

Asian American groups have made similar types of arguments in past and current complaints alleging discriminatory admissions policies. Due to the inherent difficulty of proving racial discrimination, their arguments rely heavily on statistical evidence. Admissions data in the aggregate, they say, show that universities hold Asians to higher standards than all other groups. Doing so is the equivalent of applying race-based penalties to Asian applicants.

First, Asian Americans challenge the notion that they are overrepresented. The question of parity depends on the baseline for comparison: although Asians are overrepresented in relation to the general population, they tend to be the most underrepresented group when compared to the applicant pool. Studies show that Asians have the lowest acceptance rates of all racial groups.13×13. See, e.g., Asian Am. Coal. for Educ., Complaint for Unlawful Discrimination Against Asian-American Applicants in the College Admissions Process 12–13 (2016) [hereinafter AACE Complaint], http://asianamericanforeducation.org/wp-content/uploads/2016/05/Complaint_Yale_Brown_Dartmouth_Full.pdf [https://perma.cc/R3BN-TPXF]. In a comprehensive study of ten academically selective institutions, Professor Thomas Espenshade and sociologist Alexandria Walton Radford found that both public and private institutions admitted Asians at the lowest rates. Thomas J. Espenshade & Alexandria Walton Radford, No Longer Separate, Not Yet Equal 80 tbl.3.3 (2009).

Second, complainants assert that the low rates of acceptance are especially egregious given that Asian Americans tend to be better qualified than the average applicant. This is demonstrable at least with regard to quantifiable academic qualifications. At the top range of SAT score-senders — from which selective institutions draw the majority of their students — around 50% are Asian.14×14. See SFFA Complaint, supra note 1, at 47. Similar patterns exist for other conventional indicators of academic merit, including the National Merit Scholarship, Intel Science Talent Search, and various national awards for high school students.15×15. Ron Unz, The Myth of American Meritocracy: How Corrupt Are Ivy League Admissions?, Am. Conservative (Nov. 28, 2012), http://theamericanconservative.com/articles/the-myth-of-american-meritocracy/ [https://perma.cc/B35H-83NF].

Put another way, Asians must perform better than all other groups to have the same chance of admission. One study showed that in order to be admitted to certain selective institutions, Asian applicants needed to score — on the 1600 point scale of the “old SAT” — 140 points higher than whites, 270 points higher than Hispanics, and 450 points higher than African Americans if other factors are held equal.16×16. Espenshade & Radford, supra note 13, at 92 tbl.3.5. The study controlled for other factors, including but not limited to, social class, being a recruited athlete, and legacy status. See id. It is possible that Asians are underqualified with regard to nonacademic factors, but the few studies that have researched the issue, including one by the U.S. Department of Education, have found no correlation between race and extracurricular achievements. See, e.g., Shawn Ho, A Critique of the Motivations Behind Negative Action Against Asian Americans in U.S. Universities: The Model Victims, 5 Colum. J. Race & L. 79, 93 (2015).

Third, Asian Americans point to the stability of racial demographics at universities. While the number of Asian applicants has increased manifold in the last few decades, the number of admitted students has stagnated, with some variation.17×17. See, e.g., SFFA Complaint, supra note 1, at 69; AACE Complaint, supra note 13, at 15–16. By comparison, institutions that do not employ race-based affirmative action programs, such as Caltech, have seen a dramatic increase in Asian enrollment “commensurate with the increase of college-age Asian-Americans.”18×18. AACE Complaint, supra note 13, at 15. This, according to complainants, is evidence of informal quotas and racial balancing — similar to the de facto ceiling quotas imposed on Jewish students in the early twentieth century.19×19. See, e.g., Jerome Karabel, The Chosen 110–26 (2005); Stephen Steinberg, How Jewish Quotas Began, Commentary (Sept. 1, 1971), https://www.commentarymagazine.com/articles/how-jewish-quotas-began/ [https://perma.cc/HNS4-YM7F].

B. Legal Challenges by Asian Americans

Asian American groups have made variants of these arguments since the early 1980s and have filed multiple complaints against and urged investigations into a number of universities.

At Stanford, the Committee on Undergraduate Admissions and Financial Aid, after an exhaustive internal investigation, conceded negative action against Asian applicants.20×20. Dana Y. Takagi, The Retreat from Race 39–40 (1992). Its 1986 report stated: “No factor we considered can explain completely the discrepancy in admission rates between Asian Americans and whites.”21×21. Id. Subconscious bias by admissions officers was likely the culprit, it concluded, but the Committee “elected not to investigate the bias because ‘the analysis required would be formidable.’”22×22. Id. at 40. A similar episode took place at Brown, where an internal committee found that “Asian American applicants have been treated unfairly in the admissions process.”23×23. Id. at 29. Like Stanford, Brown maintained that the unequal treatment was unintentional and due to cultural bias. Karabel, supra note 19, at 501. On the other hand, internal investigations at Cornell, Princeton, and Harvard did not find discrimination against Asian applicants.24×24. Takagi, supra note 20, at 9.

In 1988, the U.S. Department of Education launched two high-profile civil rights investigations into Harvard and UCLA. After two years of review, the Office of Civil Rights (OCR) cleared Harvard but found that UCLA had discriminated against Asian applicants.25×25. Id. at 164. OCR determined that UCLA’s graduate math program had not complied with Title VI because it had rejected Asian students whose qualifications were comparable to admitted white students. Per the OCR order, UCLA made “belated admissions offers” to the rejected students.26×26. Id. At Harvard too, OCR found that Asian students were admitted at significantly lower rates than similarly qualified white students. But Harvard’s preference for legacy applicants and recruited athletes explained the disparity. The report concluded: “OCR finds that Harvard’s use of preferences for children of alumni, while disproportionately benefiting white applicants, does not violate Title VI of the Civil Rights Act of 1964.”27×27. Karabel, supra note 19, at 505.

Now is an important moment for Asian Americans and university admissions. In addition to the SFFA lawsuit against Harvard and a potentially related Department of Justice investigation,28×28. William S. Flanagan & Michael E. Xie, Justice Department Actively Investigating Harvard Admissions, Harv. Crimson (Oct. 11, 2017), http://www.thecrimson.com/article/2017/10/11/foia-confirms-doj-investigation/ [https://perma.cc/GGC8-EUG6]. Asian Americans have filed federal civil rights complaints against, among others, Yale University, Columbia University, and the University of Chicago.29×29. AACE Complaint, supra note 13, at 12.

II. The Law of Affirmative Action

Current law on affirmative action is based on Justice Powell’s lone opinion in Bakke,30×30. Allan Bakke, a white male, was twice rejected by the Medical School of the University of California at Davis. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 276–77 (1978). Each year, the medical school accepted a class of one hundred students but reserved sixteen spots exclusively for minorities. Id. at 279. which shifted the justification for racial preferences from remedying societal discrimination to the attainment of diversity. This shift has had a largely negative impact on Asian applicants.

A. The Supreme Court Decisions

Affirmative action in higher education grew out of the civil rights movement. After centuries of discrimination and segregation, minorities in America did not — and could not — rush into universities the moment they finally opened their doors.31×31. See Colin S. Diver, From Equality to Diversity: The Detour from Brown to Grutter, 2004 U. Ill. L. Rev. 691, 695–96. University administrators, troubled by their overwhelmingly white student bodies, established affirmative action programs to assist minority groups that had been disadvantaged by past and present discrimination.32×32. Id. at 695. These early programs included Asian Americans along with other minorities.33×33. See, e.g., Sharon S. Lee, The De-Minoritization of Asian Americans: A Historical Examination of the Representations of Asian Americans in Affirmative Action Admissions Policies at the University of California, 15 Asian Am. L.J. 129, 132–33 (2008).

Race-based preferences quickly became controversial, largely due to white students’ objections to “reverse discrimination.”34×34. See Bakke, 438 U.S. at 288 & n.25 (Powell, J.). In 1978, the issue came before the Supreme Court in Bakke. Justice Powell’s pivotal opinion recognized “the attainment of a diverse student body” as a compelling state interest under the Equal Protection Clause.35×35. Id. at 311; see also id. at 311–12, 314. The Court divided 4–4–1. Chief Justice Burger and Justices Stewart, Rehnquist, and Stevens concluded that the special admissions program was racially discriminatory and violated Title VI. Id. at 421 (Stevens, J., concurring in the judgment in part and dissenting in part). Thus, they declined to address the constitutional issue. Id. at 411–12. Justices Brennan, White, Marshall, and Blackmun defended the Davis program as permissible under both Title VI and the Fourteenth Amendment. In addition, they endorsed the remedial rationale. Id. at 325–26 (Brennan, J., concurring in the judgment in part and dissenting in part). The diversity rationale invoked First Amendment values, particularly the “freedom of a university to make its own judgments as to education[,] includ[ing] the selection of its student body.”36×36. Id. at 312 (Powell, J.). In endorsing diversity, Justice Powell rejected the other rationales offered by the University of California, including its predominant one of “remedying the effects of ‘societal discrimination.’”37×37. Id. at 307. As Justice Powell explained, because Davis Medical School was a new institution with no history of racial discrimination as documented by a legislative, adjudicative, or administrative finding, the remedial rationale was inappropriate under these circumstances. See id. at 307–10. This reasoning did not foreclose the use of the remedial rationale by institutions that had previously engaged in racial discrimination. See id. at 307. The concept of general societal discrimination against minorities — without “particularized findings of past discrimination”38×38. Id. at 295 n.34.  — was simply too “amorphous” to justify the burden on innocent white applicants like Allan Bakke.39×39. Id. at 307.

Although Justice Powell was the lone Justice to favor the diversity rationale at the time, his opinion was typically viewed as controlling, and the Supreme Court later endorsed it in Grutter v. Bollinger40×40. 539 U.S. 306 (2003). and Fisher v. University of Texas at Austin (Fisher II).41×41. 136 S. Ct. 2198 (2016). Today, diversity remains the primary compelling interest that can justify race-based admissions programs in higher education.

1. The Ends: Defining Diversity. —

For the purposes of this analysis, the Court’s diversity rationale stands out in three ways: deference to the university, inseverability from proportionate representation, and departure from the original antidiscrimination goals of affirmative action programs.

The Supreme Court has never sought to define diversity with much precision: How much diversity is sufficient? How big of a role can race play in admissions? Is racial diversity equally important in engineering versus the liberal arts? What does student body diversity actually look like? Instead, the Court has left these questions to university administrators on the assumption that they are “complex educational judgments in an area that lies primarily within the expertise of the university.”42×42. Grutter, 539 U.S. at 328. Another, less explicit assumption holds that affirmative action programs are “benign” racial classifications that help minorities rather than “invidious” classifications that harm them.43×43. See Richard H. Sander, A Systemic Analysis of Affirmative Action in American Law Schools, 57 Stan. L. Rev. 367, 481 (2004). Despite the Court’s insistence that it applies strict scrutiny to all racial classifications,44×44. See, e.g., Fisher II, 136 S. Ct. at 2208; Grutter, 539 U.S. at 326. it scrutinizes university affirmative action programs with considerably less rigor.45×45. See Randall Kennedy, For Discrimination 201 (2013). Professor John Jeffries, a former clerk for Justice Powell, notes: “Powell applied strict scrutiny with unexpected pliancy” in Bakke. John C. Jeffries, Jr., Bakke Revisited, 2003 Sup. Ct. Rev. 1, 7.

Deference to the university is substantial, so much so that some commentators have argued that university administrators have a “blank check” to craft race-based admissions programs.46×46. K.G. Jan Pillai, The Defacing Reconstruction of Powellian Diversity, 31 T. Marshall L. Rev. 1, 37 (2005) (describing Justice O’Connor’s Grutter opinion). The Court defers both to the university’s conclusion that “diversity is essential to its educational mission” and to its definition of student body diversity.47×47. Grutter, 539 U.S. at 328. It also presumes the university’s good faith, “absent ‘a showing to the contrary.’”48×48. Id. at 329 (quoting Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 319 (1978) (Powell, J.)); see also Tom Campbell, Separation of Powers in Practice 124 (2004) (arguing that this presumption of good faith is “path-breaking”). Consequently, the Supreme Court “impose[s] no formal evidentiary requirement” on the university to justify its diversity interest.49×49. Angelo N. Ancheta, Contextual Strict Scrutiny and Race-Conscious Policy Making, 36 Loy. U. Chi. L.J. 21, 33 (2004).

A legitimate interest in diversity is often defined in opposition to racial balancing, which the Court rejects as “patently unconstitutional.”50×50. Grutter, 539 U.S. at 330; see also Bakke, 438 U.S. at 305 (Powell, J.). But the difference between the two is too subtle to be meaningful. Racial balancing values “race for its own sake,”51×51. Grutter, 539 U.S. at 355 (Thomas, J., concurring in part and dissenting in part). while diversity is compelling for the educational benefits that come with it.52×52. Id. at 329 (majority opinion). These benefits can be vague and abstract: “cross-racial understanding” and “break[ing] down racial stereotypes,” as well as the promotion of “learning outcomes” and “better prepar[ing] students for an increasingly diverse work force.”53×53. Id. at 330. Because the Court presumes the good faith of admissions officers who purport to pursue educational benefits, it is difficult to imagine any admissions goal that could not be easily reframed from unconstitutional to constitutional.54×54. During admissions season, the University of Michigan Law School Director of Admissions reviewed “daily reports” tracking the racial composition of admitted students. Id. at 318. He testified that he did not do this for racial balancing but for the attainment of diversity and its educational benefits. The Court accepted this explanation in Grutter. Id.

The Supreme Court’s decisions have approved an interpretation of diversity that relies on at least some notion of proportionate representation.55×55. See Chan Hee Chu, Note, When Proportionality Equals Diversity: Asian Americans and Affirmative Action, 23 Asian Am. L.J. 99, 106–08 (2016). Universities aim to increase the number of underrepresented minorities because, by definition, they belong to groups that have disproportionately low numbers on university campuses. On its face, the admissions goal of “critical mass,” first approved by the Court in Grutter, seems to cut against the idea of proportionate representation. Critical mass means “‘meaningful representation,’ . . . a number that encourages underrepresented minority students to participate in the classroom and not feel isolated.”56×56. Grutter, 539 U.S. at 318. This might suggest that critical mass is a threshold number that is similar for every group — whether a student feels isolated on campus has little to do with her racial group’s population in the state or country.

But data do not support this theory. As Chief Justice Rehnquist noted in his Grutter dissent, each group’s apparent critical mass varies dramatically and correlates with population size.57×57. Id. at 383–85 (Rehnquist, C.J., dissenting). Justice Alito, dissenting in Fisher II, made the same observation. The University of Texas enrolled more Hispanics (19.9%) than Asians (18.6%),58×58. Fisher II, 136 S. Ct. 2198, 2224 n.2, 2236 (2016) (Alito, J., dissenting). and its own research “showed that more classes lacked Asian-American students than lacked Hispanic students.”59×59. Id. at 2219. Yet of the two groups, the university considered only Hispanics to be “underrepresented” and below critical mass.60×60. Id. at 2219–20. This distinction appears to turn on Texas’s state demographics: 3.8% Asian and 37.6% Hispanic or Latino.61×61. Id. at 2224 n.2.

Finally, the diversity rationale shifts focus away from the harmful impact of racism on minorities. In this respect it diverges the most from the remedial rationale. For the Court and many Americans, the diversity rationale is appealing precisely because it is more easily viewed as “colorblind.”62×62. But see David A. Strauss, The Myth of Colorblindness, 1986 Sup. Ct. Rev. 99, 99. At least in theory, anyone can contribute to and benefit from diversity. Justice Powell emphasized this point in Bakke — speaking approvingly of the Harvard Plan, which considered race as one diversity factor among others, he noted that a “farm boy from Idaho” can just as readily contribute to diversity as “a black student [who] can usually bring something that a white person cannot offer.”63×63. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 316 (1978) (Powell, J.). This made the Harvard Plan a “facially nondiscriminatory admissions policy.”64×64. Id. at 318.

Fisher II has taken this “de-racialized” notion of affirmative action to its logical extreme. Justice Kennedy’s majority opinion, in upholding the University of Texas admissions policy, asserted that race “does not operate as a mechanical plus factor for underrepresented minorities.”65×65. Fisher II, 136 S. Ct. at 2207 (citing Fisher v. Univ. of Tex. at Austin, 645 F. Supp. 2d 587, 606 (W.D. Tex. 2009)). This was demonstrated by the district court’s finding that “the consideration of race, within the full context of the entire application, may be beneficial to any UT Austin applicant — including whites and Asian-Americans.”66×66. Id. (quoting Fisher, 645 F. Supp. 2d at 606). But see Reva B. Siegel, Equality Talk: Antisubordination and Anticlassification Values in Constitutional Struggles over Brown, 117 Harv. L. Rev. 1470, 1538–39 (2004) (arguing that Justice O’Connor’s dicta in Grutter came closest to embodying the antidiscrimination values of the remedial rationale). This vision of diversity, albeit comforting in its colorblindness, disregards historical and social context.67×67. Selena Dong, Note, “Too Many Asians”: The Challenge of Fighting Discrimination Against Asian-Americans and Preserving Affirmative Action, 47 Stan. L. Rev. 1027, 1046 (1995). While the remedial rationale recognizes that lack of diversity is a consequence of racism,68×68. Trina Jones, The Diversity Rationale: A Problematic Solution, 1 Stan. J. C.R. & C.L. 171, 179 (2005). the diversity rationale promotes racial representation as an end in itself and allows even members of the white majority to benefit from the consideration of race.

2. The Means: Symbolic Limits. —

Despite its deference to university administrators, the Court imposes constitutional limits on the means by which universities can consider race in admissions. These limits, however, are mainly symbolic and have almost no effect on admissions results.

Notably, racial quotas are impermissible.69×69. Bakke, 438 U.S. at 316 (Powell, J.); see also Fisher II, 136 S. Ct. at 2203; Grutter v. Bollinger, 539 U.S. 306, 335 (2003). This means that a university cannot — for any racial group — set aside “a fixed number or percentage which must be attained, or which cannot be exceeded.”70×70. Grutter, 539 U.S. at 335 (quoting Local 28 of the Sheet Metal Workers’ Int’l Ass’n v. EEOC, 478 U.S. 421, 495 (1986) (O’Connor, J., concurring in part and dissenting in part)). By implication, it also means that all applicants must compete within the same pool: no group can be insulated from competition with other groups through “separate admissions tracks.”71×71. Id. at 334 (citing Bakke, 438 U.S. at 315–16 (Powell, J.)). Numerical goals, on the other hand, are perfectly permissible. Goals are not quotas because they do not involve rigid numbers; they are flexible and can vary from year to year.72×72. Id. at 335. As the dissenting Justices and commentators have pointed out, again and again, the line between quotas and goals is a thin one.73×73. See, e.g., id. at 389 (Kennedy, J., dissenting) (“[T]he concept of critical mass is a delusion used by the Law School . . . to achieve numerical goals indistinguishable from quotas.”). In practice, universities may “‘sculpt’ the class with race and gender percentages in mind.”74×74. Sara Harberson, The Truth About “Holistic” College Admissions, L.A. Times (June 9, 2015, 5:00 AM), http://www.latimes.com/opinion/op-ed/la-oe-harberson-asian-american-admission-rates-20150609-story.html [https://perma.cc/2493-C2V5]; see also Richard Sander, Why Strict Scrutiny Requires Transparency: The Practical Effects of Bakke, Gratz, and Grutter, in New Directions in Judicial Politics 277, 278 (Kevin T. McGuire ed., 2012). To avoid the appearance of quotas, admissions officers have been told to “vary the numbers a bit more” and produce yearly fluctuations.75×75. Sander, supra note 43, at 408.

Universities must also engage in “individualized consideration” of each applicant.76×76. Grutter, 539 U.S. at 334. Race cannot be the only, or predominant, factor in the decision to admit or reject a student. Nor can the admissions program mechanically award “bonuses” to members of certain racial groups.77×77. Id. at 337. What it can do is award race-based “bonuses” in a flexible, nonmechanical way. Thus, the University of Michigan undergraduate admissions program was unconstitutional because it gave all underrepresented minorities twenty extra points, but the University of Michigan Law School admissions program was constitutional because it reviewed each applicant holistically before giving some applicants an undefined number of bonus points on the basis of race.78×78. Id.; Gratz v. Bollinger, 539 U.S. 244, 268–75 (2003). Less important is how much weight race may carry. On average, the law school program assigned more weight to race than did the undergraduate program. See Sander, supra note 74, at 290.

B. Diversity and Asian Americans

In admissions, the Supreme Court’s primary effect was to encourage universities “to go underground.”79×79. Sander, supra note 43, at 383. After Bakke, admissions officers “got the message”: they could continue to apply race-based preferences, as long as they abided by the Court’s formalist guidelines.80×80. Sander, supra note 74, at 286; see also Bernard Schwartz, Behind Bakke 155 (1988). In a report prepared by their lawyers, the American Council on Education and the Association of American Law Schools advised schools to “emulate features of the Harvard plan: They should not have quotas or set-asides. They should not have separate committees or procedures to evaluate minority and nonminority candidates, as Davis had. And they should articulate their policies, including their goal of diversity.”81×81. Welch & Gruhl, supra note 6, at 63. This wouldn’t be difficult to implement, some lawyers noted, because the difference between a permissible diversity plan and an impermissible quota was “nothing more than a smirk and a wink.”82×82. Id. To many commentators, this focus on form over function is evidence that the diversity rationale is merely an “elaborate pretext” for the remedial rationale — under the cover of diversity, universities continue to operate affirmative action programs in order to remedy the effects of racial discrimination.83×83. Kennedy, supra note 45, at 203; see, e.g., Kenneth L. Karst & Harold W. Horowitz, The Bakke Opinions and Equal Protection Doctrine, 14 Harv. C.R.-C.L. L. Rev. 7, 15–17 (1979).

This Note argues that the Asian experience defies this narrative. If universities were primarily motivated by remedial goals, Asians — a minority group that has been subject to past and present discrimination — should receive either race-based preferences or, at the very least, treatment identical to that of whites. While minority “underrepresentation may signal that discrimination is present,” minority overrepresentation does not support the inference that minorities enjoy institutional advantages unavailable to whites, or that they discriminate against the white majority.84×84. Lee, supra note 33, at 150. See generally John Hart Ely, The Constitutionality of Reverse Racial Discrimination, 41 U. Chi. L. Rev. 723 (1974). Any “Asian handicap” would be unjustifiable under the remedial rationale.

Instead, universities appear to have taken the diversity rationale seriously and have adopted a vision of diversity that is driven by the notion of representation. Because Asians are an overrepresented minority group, this interpretation of diversity works to their detriment.

1. Diversity in Practice. —

When diversity is inseverable from proportionate representation, the worst position for an applicant is to be a member of an “overrepresented” group. By most accounts, Asians are the most overrepresented racial group in selective institutions. Asians constitute 5.7% of the U.S. population,85×85. QuickFacts: United States, U.S. Census Bureau, https://www.census.gov/quickfacts/table/PST045216/00 [https://perma.cc/Y54X-4ZU2]. but they’re currently 22.2% of the freshman class at Harvard,86×86. Harvard Admitted Students Profile, Harv. U., https://college.harvard.edu/admissions/admissions-statistics [https://perma.cc/EB99-D3RG]. 21% at Stanford,87×87. Undergraduate Student Profile, Fall 2016, Stan., http://facts.stanford.edu/academics/undergraduate-profile (last updated June 26, 2017) [https://perma.cc/GC7S-6E3P]. 42% at Caltech,88×88. Fall Enrollment 2016–17, Caltech, https://www.registrar.caltech.edu/academics/enrollment [https://perma.cc/FWV4-P4AG]. and 42.3% at Berkeley.89×89. UC Berkeley Fall Enrollment Data, UC Berkeley, http://opa.berkeley.edu/uc-berkeley-fall-enrollment-data [https://perma.cc/5QRE-A6BH]. To a certain extent, these Asian students have displaced white students, who have become underrepresented. This effect is most evident at universities that do not employ race-based admissions policies: compared to 61.3% of the national population,90×90. QuickFacts: United States, supra note 85. white students comprise only 29% of Caltech’s freshman class,91×91. Fall Enrollment 2016–17, supra note 88. and 24.2% of Berkeley’s.92×92. UC Berkeley Fall Enrollment Data, supra note 89. It should be noted that both Caltech and Berkeley are located in California, where Asians are a higher percentage of the state population — 14.8%. QuickFacts: California, U.S. Census Bureau, https://www.census.gov/quickfacts/table/PST045216/06 [https://perma.cc/JJ8Z-KLE7]. At universities with race-based programs, white students are slightly below 50% at Harvard,93×93. Harvard College does not release admissions statistics on white students but provides that 51% of its newly admitted students belong to minority groups. Harvard Admitted Students Profile, supra note 86. and 36% at Stanford.94×94. Undergraduate Student Profile, Fall 2016, supra note 87.

At times, Asian overrepresentation provokes anxiety and backlash. Non-Asian classmates have complained that Asians destroy the class curve, mocked Asian students for being test-taking machines, and given universities racist nicknames — MIT was “Made in Taiwan” and UCLA was “United Caucasians Lost Among Asians.”95×95. Frank H. Wu, Neither Black nor White: Asian Americans and Affirmative Action, 15 B.C. Third World L.J. 225, 239 (1995).

University administrators, though far less cruel, have echoed the sentiment that Asians are overrepresented on campuses. Some did so explicitly: former dean Henry Rosovsky of Harvard referred to Asians as “no doubt the most over-represented group in the university.”96×96. Fox Butterfield, Harvard’s “Core” Dean Glances Back, N.Y. Times (June 2, 1984), http://www.nytimes.com/1984/06/02/us/harvard-s-core-dean-glances-back.html [https://perma.cc/25PR-EQ26]. Judges have also made this observation. As recently as Fisher I,97×97. 133 S. Ct. 2411 (2013). the Texas district court judge noted that “Asian-Americans . . . are largely overrepresented compared to their percentage of Texas’ population.”98×98. Fisher v. Univ. of Tex. at Austin, 645 F. Supp. 2d 587, 606 (W.D. Tex. 2009), aff’d, 631 F.3d 213 (5th Cir. 2011), vacated, 133 S. Ct. 2411.

If proportionate representation is important to the attainment of diversity, then Asian overrepresentation can threaten diversity. David Gardner, former President of the University of California, certainly thought so — he concluded that Asian overrepresentation had “an adverse effect on the [UC] system’s attempts to increase Hispanic and black enrollment.”99×99. UC Chief Wants a Better Ethnic Mix, The Tribune (Oakland, Cal.), Dec. 12, 1986, at A-16. Most university officials did not publicly announce their concern over Asian overrepresentation, although some expressed their views behind closed doors. Princeton Professor Uwe Reinhardt recalled a conversation with the university administration, during which a representative told him: “[I]t’s useful to have different cultures represented here. You wouldn’t want half the campus to be Chinese.”100×100. Golden, supra note 10, at 205. Reinhardt’s response, which I echo here, was “why not?” Id.

Universities’ first response to Asian overrepresentation was to exclude them from affirmative action programs. Initially, these programs aimed to assist all minorities that had suffered discrimination and generally included Asians along with other groups. By the time the Supreme Court heard Bakke, however, Asians were already well on their way to overrepresentation in certain fields. As a result, Justice Powell deemed their inclusion in special admissions “especially curious in light of the substantial numbers of Asians admitted through the regular admissions process.”101×101. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 309 n.45 (1978) (Powell, J.). Three years prior, Berkeley’s law school had already excluded Asians from special admissions.102×102. Sharon S. Lee, Over-Represented and De-Minoritized: The Racialization of Asian Americans in Higher Education, 2 InterActions, no. 2, 2006, at 1, 6. And in 1984, the Berkeley undergraduate admissions office also formally eliminated affirmative action for Asians.103×103. John Aubrey Douglass, Anatomy of Conflict: The Making and Unmaking of Affirmative Action at the University of California, in Color Lines 118, 128 (John David Skrentny ed., 2001). For years, Harvard refused to recognize Asians as a minority group altogether — they were barred from minority recruiting campaigns as well as Freshman Minority Orientation.104×104. Karabel, supra note 19, at 499.

As discussed above, considerable statistical and circumstantial evidence suggests that universities may also engage in “negative action” against Asian applicants. Indeed, some schools have conceded that their admissions policies discriminated against Asians.105×105. See supra p. 607. Admissions officers are not necessarily driven by racial animus; rather, they may “sincerely, if mistakenly, believe[] that curtailing the admission of Asian Americans would serve various pedagogical and social goals.”106×106. Jerry Kang, Negative Action Against Asian Americans: The Internal Instability of Dworkin’s Defense of Affirmative Action, 31 Harv. C.R.-C.L. L. Rev. 1, 19 (1996). Some may think that having too many Asians causes educational harm to institutions and their students. Commenting on the Fisher litigation in 2012, a former Ivy League admissions officer worried that without race-based preferences, “our elite campuses would look like UCLA and Berkeley . . . . That wouldn’t be good for Asians or for anyone else.”107×107. Ethan Bronner, Asian-Americans in the Argument, N.Y. Times (Nov. 1, 2012), http://www.nytimes.com/2012/11/04/education/edlife/affirmative-action-a-complicated-issue-for-asian-americans.html [https://perma.cc/YQJ6-44JQ].

The Court-approved diversity rationale provides universities with the flexibility to curtail Asian overrepresentation with very little accountability. Professor Alan Dershowitz has argued that because the “‘diversity-discretion’ model . . . lacks real substantive content, [it] is inherently capable of manipulation for good or evil results.”108×108. Alan M. Dershowitz & Laura Hanft, Affirmative Action and the Harvard College Diversity-Discretion Model: Paradigm or Pretext?, 1 Cardozo L. Rev. 379, 404 (1979). The Harvard Plan lauded by Justice Powell was, in fact, created for the purpose of limiting the number of overrepresented, high-scoring Jewish students in the 1920s. To tackle what he called Harvard’s “Jewish problem,” then-President Lowell came up with a diversity-based “admissions system capable of manipulating a variety of factors, such as personality, character, geography, and genealogy, in order to produce the desired ethnic balance in an entering class.”109×109. Id. at 401. Modern admissions officers are free to consider these same factors and weigh them however they see fit, provided that their methods are not too mechanical.

Even if admissions officers do not expressly aim to cap Asian enrollment, they are nevertheless subject to implicit racial biases. In particular, implicit biases influence how university administrators conceptualize diversity and the ways in which students can contribute to it. Motivated by numerical goals and unchecked by the law, officers may subconsciously hold different groups to different standards or carry out within-group comparisons. A former admissions official, for example, admitted that “there’s an expectation that Asian Americans will be the highest test scorers and at the top of their class; anything less can become an easy reason for a denial.”110×110. Harberson, supra note 74. Having high scores is only the bare minimum; Asian applicants must then distinguish themselves from all the other Asians. Marilee Jones, the former dean of admissions at MIT, was unusually candid about the rejection of an Asian American student: “[I]t’s possible that Henry Park looked like a thousand other Korean kids with the exact same profile of grades and activities and temperament . . . yet another textureless math grind.”111×111. Golden, supra note 10, at 201. Biases are even more dangerous when it comes to subjective, unquantifiable factors — things like personality, character, and Harvard’s “What sort of human being are you now?”112×112. What We Look For, Harv. U., https://college.harvard.edu/admissions/application-process/what-we-look [https://perma.cc/EU9V-GSBA].

2. The Prospects of Litigation. —

Even if litigants can prove that admissions policies effectively penalize Asians in comparison to white students, it is unlikely that they would win under Supreme Court doctrine. Absent a “smoking gun” demonstrating bad faith or particularly egregious discrimination, almost all affirmative action programs are insulated from legal challenge.113×113. Schwartz, supra note 80, at 157.

No sensible university imposes a formal quota, and formalism drives the Court’s analysis.114×114. See Marcia G. Synnott, The Evolving Diversity Rationale in University Admissions: From Regents v. Bakke to the University of Michigan Cases, 90 Cornell L. Rev. 463, 498 (2005). As defined by the Supreme Court, quotas are rigid rather than flexible, and specific numbers rather than ranges.115×115. See supra section II.A.2, pp. 612–13. According to a former associate dean of admissions at the University of Virginia, admissions committees “may have a goal, and a goal is not a quota. That is where you get into semantics.”116×116. Theodore R. Delwiche, “Harvard Not Fair” Seeks Rejected Applicants for Race-Based Affirmative Action Suit, Harv. Crimson (Apr. 24, 2014, 12:28 PM), http://www.thecrimson.com/article/2014/4/24/race-admissions-lawsuit-website/?page=single [https://perma.cc/N893-AE78]. As long as a university does not announce a specific percentage cap on Asians and slightly varies its racial makeup year by year, it does not employ formal quotas. And for a plaintiff to show that an affirmative action program uses the “functional equivalent of a quota,” she must prove that the university acted in bad faith.117×117. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 318 (1978) (Powell, J.); see id. at 318–19. To do this is extremely difficult: “[T]he admissions officers themselves [must] acknowledge their own ‘bad faith.’”118×118. Schwartz, supra note 80, at 157.

Likewise, universities know better than to mechanically apply an admissions boost to all members of a particular group. For appearance’s sake, they probably don’t explicitly give race-based “minus points” to any applicant. Of course, not receiving any points translates to “minus points” when compared to the “bonus points” awarded to others. Fisher II made clear that underrepresented minorities are not the only applicants who can receive “bonus points” on the basis of race — anyone can, “including whites and Asian-Americans.”119×119. Fisher II, 136 S. Ct. 2198, 2207 (2016) (quoting Fisher v. Univ. of Tex. at Austin, 645 F. Supp. 2d 587, 606 (W.D. Tex. 2009)). Given white underrepresentation in some institutions, there is no reason why whiteness cannot theoretically warrant the award of “bonus points” to reach critical mass.

Statistical evidence suggests that, in the aggregate, white students have received “bonus points” in comparison to Asians.120×120. See supra section I.A, pp. 605–06. This doesn’t mean that white students categorically receive preferential treatment, or that universities mechanically award all white applicants the same number of “bonus points.” After all, the Fisher II Court said that Asians can benefit from race-based considerations as well. Even if universities privilege white students over Asians most of the time, they do so without explicitly violating any of the Court’s commands. The general rule may be an Asian penalty, but the existence of some exceptions proves that individualized consideration is preserved. The Court used this logic when it pointed to a handful of outliers as adequate evidence that the Michigan Law School admissions program was nonmechanical.121×121. See Grutter v. Bollinger, 539 U.S. 306, 338 (2003); id. at 382–83 (Rehnquist, C.J., dissenting).

Asian litigants would face better prospects if courts were to apply more rigorous scrutiny, but doing so may involve a significant departure from the Supreme Court’s past affirmative action cases. Despite courts’ rhetoric about applying strict scrutiny to all racial classifications, they probably relax scrutiny when white plaintiffs allege discrimination. Asian American litigants are different from Allan Bakke or Abigail Fisher — they are minority group members and more sympathetic plaintiffs. Under “real” strict scrutiny, courts may find that admissions policies have functionally established “separate tracks” for different racial groups and insulated them from competition with each other.122×122. See Kermit Roosevelt III, The Ironies of Affirmative Action, 17 U. Pa. J. Const. L. 729, 729–30 (2015) (arguing that “under real strict scrutiny, almost all race-based affirmative action programs are likely unconstitutional”). They may also find that universities are engaged in “racial balancing” because there is no compelling reason a majority-white campus provides more educational benefits than a majority-Asian one. Depending on the information uncovered during discovery — and given how blunt admissions officers can apparently be — they may even find bad faith.

One consideration that counsels against “real” strict scrutiny is that groups like SFFA challenge affirmative action policies as a whole, rather than just “negative action” compared to white students. Doing so pits the plaintiffs against other minority groups, who would suffer the most harm from an outright prohibition of race-based consideration.123×123. See, e.g., William G. Bowen & Derek Bok, The Shape of the River, at xli (2d prtg. 2000). This is unfortunate, especially since most Asian Americans continue to support affirmative action programs and multiple Asian affinity groups filed amicus briefs supporting the University of Texas in Fisher II.124×124. Over 160 Asian American and Pacific Islander Groups File Amicus Briefs in U.S. Supreme Court in Support of Affirmative Action, Asian Am. Legal Def. & Educ. Fund (Nov. 3, 2015), http://aaldef.org/press-releases/press-release/over-160-asian-american-and-pacific-islander-groups-file-amicus-briefs-in-support-of-affirmative-act.html [https://perma.cc/46WQ-THV8]. Mee Moua, former president of Asian Americans Advancing Justice, spoke out against the SFFA action: “[T]he faction of Asian Americans publicly opposed to affirmative action is misguided and doesn’t understand that what they’re opposed to is ‘negative action.’”125×125. Alia Wong, Asian Americans and the Future of Affirmative Action, The Atlantic (June 28, 2016), https://www.theatlantic.com/education/archive/2016/06/asian-americans-and-the-future-of-affirmative-action/489023/ [https://perma.cc/M63D-XF5M].

The problem is that unlike the remedial rationale, which limits race-based preferences to racial minorities, the diversity rationale does not allow for the easy disaggregation of affirmative and negative action.126×126. Chu, supra note 55, at 99. If diversity derives meaning from proportionality, negative action against overrepresented groups is the flipside of affirmative action for underrepresented minorities.

III. Bakke’s Legacy

At the time of Bakke, Asians did not have a significant enough presence in America. But the existence of overrepresented minorities was not unknown to the Justices — they discussed the Jewish American experience in higher education, which was detailed in multiple amicus briefs. Knowing as he did that elite institutions had imposed Jewish quotas in the recent past, Justice Powell should have acted with more caution before endorsing the Harvard Plan.

A. The Justices on Asians

The Justices did not devote much time to discussing Asian Americans, probably because Asians were barely 1% of the population in 1978.127×127. Frank Hobbs & Nicole Stoops, U.S. Census Bureau, U.S. Dep’t of Commerce, CENSR-4, Demographic Trends in the 20th Century 77 fig.3-4 (2002), https://www.census.gov/prod/2002pubs/censr-4.pdf [https://perma.cc/Q3DM-44CU]. When the subject came up in conference, it was because Asians brought to light the pesky problem of line drawing. Asians “faced almost hysterical prejudice for decades after first arrival in this country . . . [and] face a quieter, subtler form of prejudice today.”128×128. Memorandum from Bob Comfort, Law Clerk, Supreme Court of the United States, to Justice Lewis F. Powell, Jr., Assoc. Justice, Supreme Court of the United States 37 (Aug. 29, 1977) (on file with the Harvard Law School Library) [hereinafter Comfort Memo]. Yet giving Asians preference in admissions — as Davis did — could lead down a slippery slope of countless ethnic groups demanding special treatment. This point particularly troubled Justice Powell, who noted that there is no “principled basis for inclusion or exclusion of groups on the basis of race or ethnic origin.”129×129. Lewis F. Powell, Jr., Preconference Notes, Regents of the University of California v. Bakke 1 (Oct. 13, 1977) (on file with the Harvard Law School Library). If Asians were included in the special program, then “[w]hy not Italians, Irish, Greeks, etc.”?130×130. Id.

If anything, the inclusion of Asians in special admissions ultimately cut against the validity of the Davis program. For Justice Powell, the fact that sufficiently represented Asians received preferential treatment “suggested that the Davis program was not a narrow response to social necessity but a political solution to interest-group politics.”131×131. John C. Jeffries, Jr., Justice Lewis F. Powell, Jr. 472–73 (1994).

Justice Powell and his clerks viewed the Asian question as a problem for Justice Brennan’s remedial argument — one that Brennan refused to address adequately. In his opinion, Justice Powell included two footnotes that dealt with the uniqueness of the Asian situation. Footnote 36 attacked Justice Brennan’s societal discrimination rationale as incomplete, because “nothing is said about Asians” and why discrimination does not appear to inhibit their academic performance.132×132. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 296 n.36 (1978) (Powell, J.). Footnote 45 questioned why Asian applicants, many of whom were admitted through regular admissions, were eligible for special admissions.133×133. Id. at 309 n.45.

B. The Justices on Jews

The status of another minority group — Jewish Americans — captured the attention of the Court, but the Justices missed the full import of the lessons of that group’s history in higher education admissions.

By the 1970s, it was widely accepted that American universities had, for decades, employed various means to limit the enrollment of overrepresented Jewish students.134×134. See sources cited supra note 19. Although Jewish quotas generally ended after World War II, many Jews remained understandably suspicious of affirmative action programs.135×135. See Steinberg, supra note 19 (noting that “many Jews are alarmed by the introduction of unofficial quotas favorable to black applicants”). Seven Jewish groups submitted amicus briefs in support of Allan Bakke.136×136. See Allan P. Sindler, Bakke, DeFunis, and Minority Admissions 243 (1978). Anxious about a potential return to racially discriminatory admissions, they directly compared affirmative action programs to Jewish quotas in the past:

[A]fter only three or four decades of nondiscriminatory admissions, in which creed, color, and ethnic origins have been rejected as appropriate criteria for university admissions, the universities, which for centuries set the style in excluding or restricting Jewish students and those of various other religious, racial, and ethnic minorities, may again be able to do so, again in the name of enlightenment and diversity, if the decision below is not affirmed.137×137. Brief Amici Curiae of Anti-Defamation League of B’nai B’rith et al. at 5–6, Bakke, 438 U.S. 265 (No. 76-811), 1977 WL 188002 [hereinafter ADL Bakke Brief].

Prominent Jewish scholars — Professor Alexander Bickel was perhaps the most influential — also denounced race-based admissions.138×138. See, e.g., Alexander M. Bickel, The Morality of Consent 132–34 (1975); Leonard Fein, The War Inside the Jews: A Painful Breakdown of the Liberal Consensus, New Republic, Oct. 15, 1977, at 16. Bickel had coauthored the Anti-Defamation League’s amicus brief in DeFunis v. Odegaard,139×139. 416 U.S. 312 (1974) (per curiam). The Court held that the case was moot and did not reach the merits. Id. at 319–20. from which its Bakke brief heavily borrowed.140×140. Compare ADL Bakke Brief, supra note 137, with Brief of the Anti-Defamation League of B’nai B’rith Amicus Curiae, DeFunis, 416 U.S. 312 (No. 73-235). He vehemently opposed the use of racial quotas, arguing that it “derogates the human dignity and individuality of all to whom it is applied.”141×141. Bickel, supra note 138, at 133.

The criticism had a big impact on the Justices. Justice Blackmun, writing to the conference, envisioned a Jewish consensus against affirmative action. He believed that “Alex Bickel’s elegant and shining words”142×142. Memorandum from Justice Harry A. Blackmun, Assoc. Justice, Supreme Court of the United States, to the Conference 11 (May 1, 1978), reprinted in Schwartz, supra note 80, app. E, at 257. represented “the ‘accepted’ Jewish approach” and observed that “nearly all the responsible Jewish organizations who have filed amicus briefs here are one side of the case.”143×143. Id. at 12.

For the liberal Justices and Justice Powell, affirmative action programs were fundamentally different from Jewish quotas, although there was some disagreement on the distinguishing principle. Justice Marshall emphasized that the Davis quota was a “quota to get someone in” — “not a quota to get someone out.”144×144. Lewis F. Powell, Jr., Conference Notes, Regents of the University of California v. Bakke 5 (Dec. 9, 1977) (on file with the Harvard Law School Library). According to Justices Brennan and Blackmun, stigma was the governing principle. The Davis program was permissible because it did not stereotype Bakke “as an incompetent, or pin[] [him] with a badge of inferiority because he is white.”145×145. Memorandum from Justice William J. Brennan, Jr., Assoc. Justice, Supreme Court of the United States, to the Conference 9 (Nov. 23, 1977), reprinted in Schwartz, supra note 80, app. D, at 235.

From the very beginning, Justice Powell was averse to strict numerical quotas. As he articulated in conference, Davis’s “colossal blunder . . . was to pick a number.”146×146. William J. Brennan, Jr., Conference Notes, Regents of the University of California v. Bakke (Dec. 9, 1977), as reprinted in Lee Epstein & Jack Knight, Piercing the Veil: William J. Brennan’s Account of Regents of the University of California v. Bakke, 19 Yale L. & Pol’y Rev. 341, 353 (2001). That Jewish quotas were so notorious probably made the concept of quotas seem even more distasteful. Justice Powell’s clerk persuaded him that the race-as-one-factor Harvard Plan of the 1970s differed from the anti-Semitic Harvard Plan of the 1920s: “The fact of opening the whole class distinguishes this approach from the closed quotas of President Lowell . . . .”147×147. Comfort Memo, supra note 128, at 60.

Notably, Justice Powell took issue with Justice Brennan’s focus on stigma, which he believed would allow for Jewish quotas. An earlier draft of Justice Powell’s opinion included a controversial footnote:

[L]imiting the concept of stigma to the imposition of a badge of inferiority would inhibit appropriate scrutiny of classifications such as the quotas imposed upon admission of Jews to some educational institutions in the early part of this century, which were based upon the belief that by virtue of superior ability that group would come to dominate such institutions.148×148. Lewis F. Powell, Jr., 5th Draft Opinion, Regents of the University of California v. Bakke 25 n.34 (drft. n.d.) (on file with the Harvard Law School Library).

It cited Stephen Steinberg’s well-known article, “How Jewish Quotas Began,”149×149. Id.; see Steinberg, supra note 19. as evidence that universities had imposed Jewish quotas because of their “perceived capability of dominating the universities.”150×150. Lewis F. Powell, Jr., Memo to File, Regents of the University of California v. Bakke (June 29, 1978) (on file with the Harvard Law School Library) [hereinafter Powell Memo].

Offended by Justice Powell’s suggestion that he would condone Jewish quotas, Justice Brennan hit back with a retort: “My clerks and I were particularly mad about this because [Powell’s] First Amendment approach quite clearly left the door open for approval of Jewish quotas, so that [Powell] was really ‘calling the kettle black.’”151×151. Epstein & Knight, supra note 146, at 369 (reprinting Justice Brennan’s notes on Bakke).

Justice Brennan’s indignation persuaded Powell to reconsider. Justice Powell recorded in his notes: “Bill Brennan called me, following circulation of my fifth draft, and was quite upset by this portion of footnote 34. He characterized it as ‘personally offensive,’ saying that his respect and admiration for our Jewish citizens was widely known.”152×152. Powell Memo, supra note 150. As a result, Justice Powell agreed to remove the reference to Jewish quotas.153×153. Justice Powell further noted that “[n]either I nor any of my clerks thought there was any reason to request this change. But I would certainly honor a request like this from Bill.” Id. True to his promise, the footnote was substantially revised in the final draft; it argued that Justice Brennan’s opinion offered “no principle for deciding whether preferential classifications reflect a benign remedial purpose or a malevolent stigmatic classification.”154×154. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 295 n.34 (1978) (Powell, J.). For his part, Justice Brennan never mentioned Jewish quotas in his opinion either.

C. Justice Powell’s Diversity

Justice Powell thought that Jewish quotas were despicable, and there is little doubt that he did not intend to endorse an admissions plan that could allow for something similar. His clerk had persuaded him that the modern Harvard Plan was distinguishable from President Lowell’s Harvard Plan: it was open rather than closed, and it treated race positively rather than negatively. The problem, however, is that it was essentially the same plan. As discussed above, the open/closed and positive/negative distinctions are formalistic and functionally meaning-less.155×155. See supra pp. 610–13. And even Harvard’s past Jewish quotas weren’t explicitly closed — the whole class was supposedly open to all, while the admissions office used multifactor considerations to indirectly lower the number of Jewish students.156×156. See Karabel, supra note 19, at 130. The Harvard Plan discussed in Bakke was just as discretionary and “inherently capable of gross abuse” as Lowell’s plan.157×157. Dershowitz & Hanft, supra note 108, at 385 (emphasis omitted). It even considered many of the same factors as the original one.

The difference was in the officers administering the plan — and their facially benign and progressive goals — not in the plan itself. Justice Powell simply placed too much trust in admissions officers at Harvard and elsewhere: he presumed they would act in good faith. Perhaps he believed that in 1978, after Brown and the civil rights movement, no university would again use racial quotas to keep minorities out of American institutions. Perhaps he also believed that, notwithstanding the experience of Jewish students, a group’s level of representation was a good enough proxy for its treatment by society and that white students would never become underrepresented on university campuses.158×158. Justice Powell appeared to have considered the implications of the diversity rationale in an earlier draft: “[A]pproval of such a policy does not presage a sub rosa return to the restrictive educational quotas of the earlier half of this century. The ethnic preference considered here would be one shown, in good faith, to members of a group which otherwise would be insignificantly represented or totally unrepresented.” Lewis F. Powell, Jr., 5th Typescript Draft, Regents of the University of California v. Bakke 44 (drft. Nov. 3, 1977) (on file with the Harvard Law School Library) (citations omitted). This language was ultimately removed.

In hindsight, Justice Brennan had the stronger argument. Justice Powell’s accusation that Brennan’s conception of stigma would condone Jewish quotas was based on a false premise. Lowell did not impose Jewish quotas because he believed that Jews were academically superior; he did so because he thought they were socially inferior and didn’t belong at Harvard. The Steinberg article (mis)cited by Justice Powell was quite clear about this: proponents of Jewish quotas “pointed to objectionable traits of Jews . . . [and] the desirable traits of the university that were presumably endangered.”159×159. Steinberg, supra note 19. University officials frequently complained that Jewish students worked doggedly hard and were unfairly competitive, and that their uncouth immigrant manners diminished the gentility and social standing of elite institutions.160×160. Id. Similarly, if an admissions officer does not want to admit too many Asians, it is not because she thinks Asians are “too smart” for her institution, but because she thinks they are “textureless math grind[s]” who might threaten diversity or undermine the educational experience.161×161. Golden, supra note 10, at 201.

The past forty years have vindicated Justice Brennan’s critique that Powell’s approach “left the door open” for potential discrimination against overrepresented minorities. Even if we consider only the universities that have admitted to anti-Asian discrimination and the admissions officials who have openly opposed Asian overrepresentation, it is clear that the diversity rationale can be manipulated to condone the exclusion of minorities. Add to this the language of the Fisher II majority, which indicated that both whites and minorities may benefit from racial preferences.162×162. See supra p. 612. By detaching affirmative action from social and historical context and grounding its justification in First Amendment values of institutional autonomy, Justice Powell’s diversity rationale allowed for these unintended consequences.


Justice Harlan’s Plessy v. Ferguson163×163. 163 U.S. 537 (1896). dissent, celebrated for its grand vision of the Colorblind Constitution, also remarked on a group that was neither black nor white:

There is a race so different from our own that we do not permit those belonging to it to become citizens of the United States. Persons belonging to it are, with few exceptions, absolutely excluded from our country. I allude to the Chinese race.164×164. Id. at 561 (Harlan, J., dissenting).

Admittedly, Asian Americans have not experienced the same level of horrific discrimination as some other minority groups have. They have nevertheless been subject to lynching, mass internment, and school segregation.165×165. Dong, supra note 67, at 1047–48.

Almost a century after Plessy, while reviewing Allan Bakke’s “reverse discrimination” claim, Justice Thurgood Marshall wrote to his colleagues: “We are not yet all equals, in large part because of the refusal of the Plessy Court to adopt the principle of color-blindness. It would be the cruelest irony for this Court to adopt the dissent in Plessy now and hold that the University must use color-blind admissions.”166×166. Memorandum from Justice Thurgood Marshall, Assoc. Justice, Supreme Court of the United States, to the Conference 3 (Apr. 13, 1978) (on file with the Harvard Law School Library). For Justice Powell, the tension between colorblind values and the realities of racial discrimination was troubling.167×167. See, e.g., Jeffries, supra note 45, at 6–8. His choice in Bakke was an uncomfortable middle ground, one that spawned a new irony: the diversity rationale brought more harm than benefit to Asian Americans, unequal treatment even in comparison to the white majority. Perhaps from Justice Powell’s perspective, his carefully neutral language masked an “unspoken assumption that the history of racial discrimination in this country inevitably makes race a valid consideration in the diversity formula.”168×168. Karst & Horowitz, supra note 83, at 16. But Justice Powell trusted — unjustifiably — that universities in the future would read between the lines and use the diversity rationale only to include, rather than exclude, racial minorities.