Thoughts on Bakke and Its Effect on Race-Conscious Decision-Making

By Michael E. Rosman
The University of Chicago Legal Forum

This article focuses on race-conscious decisionmaking (“RCDM”) in admissions processes in higher education. n1 RCDM has become increasingly controversial in recent years, so much so that the citizens of our most populous state outlawed it. n2 This article explores the phenomenon of RCDM’s increasing unpopularity in university admissions, and suggests that part of the reason for that unpopularity is Justice Powell’s diversity rationale in Regents of the University of California v Bakke. n3

The original rationale for RCDM in all walks of society was straightforward. As President Johnson put it in his famous speech at Howard University:

You do not take a person who for years has been hobbled by chains and liberate him, bring him up to a starting line of a race and then say, “you are free to compete with all [*46] the others,” and still justly believe that you have been completely fair. n4

In other words, the long, sordid history of statesponsored discrimination and the legal impediments imposed against African Americans in this country justified providing some extra consideration to AfricanAmerican candidates for various positions.

What happened to this justification? It has been essentially abandoned. The primary argument in support of RCDM, at least in higher education, is no longer Johnson’s race running metaphor, but the ever present, ill-defined “diversity.” n5 Bakke is the cause of this phenomenon. Bakke also, I believe, is one of the causes of the increasing unpopularity of RCDM.

After reporting the survey results showing the increasing unpopularity of RCDM and suggesting possible explanations (Part I), this essay discusses Justice Powell’s opinion in Bakke, its use of the terms “academic freedom” and “diversity,” and the various well-known (and not so well-known) problems that opinion has (Parts II and III). I then employ the small amount of comparative advantage I have in this area, and describe how the inherent problems in Justice Powell’s opinion have affected and skewed [*47] the defenses of RCDM and the litigation involving admissions programs that use RCDM (Part IV). I close (in Part V) with a brief thought or two about the overall problem that Justice Powell created for RCDM in higher education.

I. Faltering Support for RCDM

RCDM has become increasingly unpopular. A poll published during the summer of 2001, conducted by the Washington Post, the Henry J. Kaiser Family Foundation, and Harvard University, asked the following question to 1,709 adults (with an oversampling of minority groups), and reported the following results: n6

In order to give minorities more opportunity, do you believe race or ethnicity should be a factor when deciding who is hired, promoted, or admitted to college, or that hiring, promotions, and college admissions should be based strictly on merit and qualifications other than race or ethnicity?

Race or ethnicity Should be based Don’t know

should be a factor strictly on merit

and qualifications

other than


Total 5% 92% 3%

White 3% 94% 3%

African American 12% 86% 2$

Hispanic 7% 88% 4%

Asian 7% 84% 9%

Indeed, only a modest majority (55%) of respondents favored having employers or colleges make an extra effort to find and recruit qualified minorities, n7 and an overwhelming majority (86%) [*48] opposed using race as a factor when drawing boundaries for U.S. congressional voting districts. n8

The Post did not find these survey results surprising, and focused its attention on white failure to understand African-American circumstances. Thus, the title of its article about the survey was not “Race Preferences Overwhelmingly Unpopular Among All Races,” but rather “Misperceptions Cloud Whites’ View of Blacks.” n9 Indeed, it barely reported RCDM’s widespread unpopularity, saying only (towards the end of the article) that “an overwhelming majority of all whites and blacks continue to reject giving outright preferences to blacks and other minorities in employment or admissions to college.” n10 According to the Post, “‘hard’ preference programs are vanishing fast from the scene, either ended by judges who ruled these programs constituted reverse discrimination or abandoned by their besieged sponsors.” n11

Of course, this is nonsense. “Hard” preferences of the kind identified by the survey question, where race or ethnicity is a “factor,” are quite prevalent, especially in higher education. n12 [*49] Their sponsors are hardly “besieged”; they are openly defending these policies with significant sums of money and casting aspersions on those who disagree. n13 They are the elites, and they resent efforts to go over their heads to the people. n14 This elitism is perhaps one reason why RCDM has become so unpopular, but it is [*50] hardly the only reason or even a primary one. n15 I think there are three leading causes.

First, the Johnson race running metaphor–“You do not take a person who for years has been hobbled by chains and liberate him”–was most effective when RCDM primarily aided African Americans. At the outset it was primarily such a system. n16 Over time, however, whether to increase the political base for RCDM or for some other reason, African Americans themselves have acceded to the expansion of the policy to other ethnicities. But the history of those other groups is simply not sufficiently comparable to the history of African Americans. n17 Even Native Americans, who certainly have had a uniquely difficult history, cannot make the same moral claim as African Americans–who, after all, have experienced both slavery and Jim Crow–for “group justice.” When one then considers Hispanics, Asian Americans, and Asian Indians, as well as “white minorities” like Jews, Italian Americans, and Irish Americans, the distinctions between the groups and their histories become even harder to discern. If an RCDM policy gives preference to Hispanics, but not Jews and Asian Americans, the moral argument based upon past discrimination becomes very difficult to make.

A second reason for the increasing unpopularity of RCDM is the growth of the African-American middle class. Even if it were [*51] confined solely to African Americans, RCDM would have less moral force than it did when President Johnson spoke for the simple reason that the “chains” metaphor is somewhat overstated for many African Americans today. Indeed, no one disputes that African Americans generally have done increasingly well as a group economically. n18 When specific beneficiaries of RCDM seem to the white majority to have lived lives not too much different from their own, the moral case for RCDM is diminished. n19 Proponents of RCDM claim that Clarence Thomas was a beneficiary of RCDM, n20 but the question today is not whether Clarence Thomas should have been a beneficiary of RCDM, but whether his son should as well.

The third reason for RCDM’s failing reputation, is Justice Powell’s decision in Bakke and the Court’s failure to take a similar case in the intervening years. n21 Justice Powell rejected societal [*52] discrimination and increasing minority representation as justifications for RCDM in college admissions, and replaced them with “diversity.” Instead of focusing attention on the effects that Jim Crow might have had on the ability of African Americans to compete for spots at the Davis Medical School, Justice Powell turned the spotlight on the Davis Medical School’s “academic freedom” to consider race in the admissions policy. n22 For a variety of reasons (not the least of which is that no one else on the court mentioned the words “academic freedom”), this has proved to be a most dissatisfying justification.

II. The “Diversity” Rationale Reconsidered

In Bakke, the Court found that the admissions program of the University of California Medical School at Davis, which set aside 16% of the places for incoming students for educationally or economically disadvantaged minorities, was unconstitutional. n23 Five justices, however, concluded that race could be considered in Davis’s admissions process under some circumstances. No single theory, though, explained why that was so. n24

Justice Powell, in an opinion only for himself, applied strict scrutiny to the Davis program. He concluded that “academic freedom,” although not a specifically enumerated Constitutional right, was a “special concern” of the First Amendment and thus a sufficiently compelling interest to meet strict scrutiny. n25 The Regents specifically wanted their institutions to select a group of students who would contribute to a robust exchange of ideas, and argued that “ethnic diversity” was a means of achieving that [*53] goal. n26 While rejecting the argument that Davis’s specific program of reserving spaces for disadvantaged minorities was necessary to achieve the robust exchange of ideas that the Regents allegedly wanted, Justice Powell did state that race and ethnicity could be considered as “plus” factors by universities seeking to achieve that goal. Justice Powell opined that a state interest in a robust exchange of ideas would not justify the consideration of race to achieve the ethnic diversity promoted by UC Davis, but could justify its consideration to achieve a diversity which “encompasses a far broader array of qualifications and characteristics of which racial or ethnic origin is but a single though important element.” n27

The first problem with the “academic freedom” rationale is that no other member of the Court adopted it. Thus, its value as precedent has always been questionable. n28 But aside from that, its internal coherence leaves much to be desired.

Contrary to popular myth, Justice Powell did not hold that diversity itself is a compelling governmental interest. Rather, he held that a school’s academic freedom was a compelling interest, and that a school could pursue a policy of attaining a diverse student body as part of its academic freedom interest in determining who shall be a part of the student body. n29 But almost as soon as Justice Powell granted that freedom as compelling, he took it away. “Ethnic diversity . . . is only one element in a range of factors a university properly may consider in attaining the goal of a heterogeneous student body” because “constitutional [*54] limitations protecting individual rights may not be disregarded.” n30 “Petitioner’s argument that [ethnic diversity] is the only effective means of serving the interests of diversity is seriously flawed.” n31 “Petitioner’s special admissions program, focused solely on ethnic diversity, would hinder rather than further attainment of genuine diversity.” n32

This is an odd species of “academic freedom” indeed. Whose “diversity” does Davis Medical School have the freedom to adopt? Its own? Apparently not. True academic freedom presumably would permit institutions to reserve spots to attain “ethnic diversity” if it deemed ethnic diversity to be far more important than intellectual (or any other form) of diversity. But Justice Powell condemns Davis’s own vision of diversity as “seriously flawed” and “hindering . . . genuine diversity,” as if he gets to decide how Davis exercises its academic freedom. n33 Justice Powell finally concludes that Davis Medical School only has the academic freedom to imitate Harvard. n34

Of course, the problem is that no one (including Justice Powell, apparently) seriously believes that institutions have the complete freedom to make their own judgments about who may be admitted to study. If they did, presumably a segregationist institution would have the equal freedom to keep out members of disfavored races. n35 The Supreme Court apparently rejected that notion in Runyon v McCrary. n36 Thus, diversity has been left alone, nearly stripped of Justice Powell’s initial academic freedom rationale, as the “compelling interest.”

Runyon highlights yet another peculiarity of Bakke. The defendant in Runyon was a private school, and had no First Amendment right to make race-based decisions on admissions. Davis Medical School is a state institution, and the court holds that it does have that right. Yet the First Amendment is normally [*55] considered to be a shield wielded by private citizens against the state. n37

One need not fear only lesser institutions’ exercise of academic freedom. Harvard itself once focused on reducing the number of Jews from the city, and increasing the number of students from rural (and Protestant) areas because it believed that the latter would be better “raw material” for the production of “Harvard men” that the institution had been known for. n38 That is, Harvard wanted to exercise (and did exercise) its academic freedom to achieve more homogeneity rather than heterogeneity (or, more kindly, a different kind of heterogeneity). n39 Such is the nature of academic freedom when it comes to a school’s right to determine who is admitted.

The only other Justice who has suggested that something akin to “academic freedom” would be a permissible rationale to use race was Justice Stevens in his Wygant v Jackson Board of [*56] Education n40 dissent. n41 Justice Stevens asserted that a school board’s decision to retain an African-American teacher because she was an African American (over an equally-qualified white applicant) should be analyzed by “asking whether the Board’s action advances the public interest in educating children for the future.” n42 Since it was “quite obvious that a school board may reasonably conclude that an integrated faculty will be able to provide benefits to the student body that could not be provided by an all-white, or nearly all-white faculty,” the RCDM policy in Wygant stated a valid public interest. In addressing whether another board might appropriately think that segregation might “lead to better academic achievement,” Justice Stevens just punted. n43 He immediately abandoned the assumption of his hypothetical (that a school board might think that segregation led to better educational outcomes), and assumed that the decision would necessarily rest on the “false premise that differences in race, or in the color of a person’s skin, reflect real differences that are relevant to a person’s right to share in the blessings of a free society.” n44 Because such a decision would be exclusionary, rather than inclusionary–a somewhat contrived set of categories, since every decision for a competitive position both includes someone and excludes someone–Justice Stevens concluded that only the inclusionary RCDM was consistent with the Equal Protection Clause.

Of course, Justice Stevens’s analysis assumes that the Equal Protection Clause has an ultimate goal of “inclusion” (whatever that means) rather than equal treatment; equal (or “more equal”) results rather than equal process. Justice Stevens’s contention that the “inclusionary decision is consistent with the principle that [*57] all men are created equal” n45 is stated in an ipse dixit fashion; one could easily contend that any use of RCDM, inclusionary or exclusionary, is a violation of the general principle of “equality” before the law. n46

Even more peculiar is Justice Stevens’s insistence that consideration of race is absolutely prohibited in determining who may serve on juries because it is “utterly irrational” to believe any benefit could come from it. n47 If it is “far more convincing” to experience the truth that skin color does not matter from an African-American teacher, n48 one might think that experiencing truths about African-American life in America today might be more convincing from an African-American fellow juror rather than a white one. n49 To the extent such experiences are relevant, could not a court administrator, a county, or a prosecutor conclude that racially-integrated juries–which we can assume can be reached in some particular instances in certain areas only through RCDM–“provide benefits to [the administration of justice] that could not be provided by an all-white, or nearly all-white, jury?” n50 Like Justice Powell, Justice Stevens offers no [*58] explanation that really can limit the use of RCDM to education. n51 Indeed, unlike Justice Powell, it would seem that Justice Stevens has no desire to do so.

Justice Stevens’s “public interest” analysis is just Justice Powell’s “academic freedom” interest extended a bit more toward (but not reaching) its logical conclusion. Deference is given to the public policy choices of those with whom the Justice agrees (Harvard), and not to others (Davis or segregationists). The analysis is applied haphazardly to some areas (like education) but not others (juries) without any real explanation as to why there are no benefits from integration in the excluded areas. Race can be used so long as a plausible benefit can be defined (educational improvements from “diversity”). For better or worse, it is a prescription for the use of RCDM in large areas of our society. n52 [*59]

III. More Problems with Bakke

There are various other (mostly well-known) difficulties with the Powell decision in Bakke that I will briefly enumerate. n53 First, although he certainly does not say so, Justice Powell does suggest that there are various viewpoints associated with different ethnicities. n54 Second, the distinction between a “plus” given to a member of a racial minority and the simple set-aside system used by Davis in Bakke has been roundly criticized, including in Bakke itself. n55 This is particularly so given Justice Powell’s assertion that weights applied to race and other diversity factors “may vary from year to year depending upon the ‘mix’ both of the student body and the applicants for the incoming class” n56 and his somewhat vague references to the degree to which a school could look at numbers. n57 Indeed, he ultimately endorsed a Harvard system that seemed to operate with great attention to numbers. n58 [*60]

Even more perplexing, it is not altogether clear that the difference between a “plus” system and a “setaside” system is simply a difference between a system that is narrowly tailored and one that is not. Rather, Justice Powell suggested that “strict scrutiny” might not be applicable at all to a “plus” system because “a facial intent to discriminate” does not “exist[] in an admissions program where race or ethnic background is simply one element–to be weighed fairly against other elements–in the selection process.” n59 In such a system, “good faith would be presumed,” n60 there would be “a presumption of legality and legitimate educational purpose,” n61 and “there is no warrant for judicial interference in the academic process.” n62 The absence of an intent to discriminate and the presumption of good faith certainly suggests a standard other than strict scrutiny. n63 [*61]

Still there are even more questions raised by Powell’s opinion: must an institution include all possible diversity factors, or is it sufficient to include a few besides race? What is the rationale for excluding religion or sexual orientation? Is it sufficient that the school expects that kind of “diversity” to occur “naturally”? n64 In truth, as everyone knows, it is impossible to give preferences to all different diversifying characteristics, and some groups–indeed, some ethnic or racial groups–will simply not be present in significant numbers. n65 Does a system that provides some advantage for African Americans and Hispanics, but not Arab Americans, indicate a facial intent to discriminate in Justice Powell’s universe?

Finally, Justice Powell’s Bakke opinion is sadly ironic. As Justice Brennan’s opinion in Bakke pointed out, the Davis program was limited to those “within a general class of persons likely to have been the victims of discrimination.” n66 Indeed, the dissenters found this consideration important in concluding that the Davis program met their somewhat lower level of scrutiny:

The Davis admissions program does not simply equate minority status with disadvantage. Rather, Davis considers on an individual basis each applicant’s personal history to determine whether he or she has likely been disadvantaged by racial discrimination. The record makes clear that [*62] only minority applicants likely to have been isolated from the mainstream of American life are considered in the special program; other minority applicants are eligible only through the regular admissions program . . . Specific proof that a person has been victimized by discrimination is not a necessary predicate to offering him relief where the probability of victimization is great. n67

The dissenters also found that the constitutionality of the Davis program “is buttressed by its restriction to only 16% of the positions in the Medical School, a percentage less than that of the minority population in California.” n68

Justice Powell’s opinion is ironic because it ignored the limitation on preferences that the Davis “quota” actually provided, confining it to likely victims of societal discrimination and within reasonable proportions of the total student population. n69 Even more importantly, it rejected a program that at least made some effort to conform itself to the original purpose of RCDM–which, at least to some degree, hearkened back to President Johnson’s “chains” analogy–and gave its imprimatur to a program that provided preferences to those for whom the “chains” analogy would be least applicable. This result is perhaps the saddest part of the Bakke legacy. n70 [*63]

IV. RCDM Litigation in Higher Education

Justice Powell’s Bakke opinion has skewed the litigation surrounding RCDM in admissions in higher education and created all sorts of litigation problems for those defending RCDM. In this section, I explore a few of these phenomena.

The first problem is that the Court has said that the “compelling interest” identified when defending RCDM must be the “real” motivation. n71 This presents a problem since many candid admissions officials would admit that Justice Powell’s vision of diversity is not the motivation for their admissions programs and/or is not particularly important. n72 Indeed, the origins of such programs [*64] seem to have little connection with “diversity” as a concept. n73 It also presents an opportunity for opponents of RCDM; it usually would not be too difficult to find some statement somewhere suggesting that Powell-type diversity was not the goal of the program. n74

The second major problem is raised by the “academic freedom” rationale, which, as already suggested, would permit (were it taken seriously) the negative consideration of race as well as its positive consideration. n75 The litigation answer to this problem is primarily to ignore “academic freedom” and focus on diversity. [*65] More specifically, defenders of RCDM now place great weight on evidence that purports to support the value of diversity in education. n76 This evidence does not really address the fact that, however valuable diversity may or may not be, many institutions have done just fine without a tremendous amount of racial diversity. n77 And, of course, its mere presentation is inconsistent with the notion that pursuing diversity is simply a matter of academic freedom. Academic freedom permits colleges, universities, and professors the freedom to pursue ideas regardless of what evidence there may be to support them; it protects unpopular and highly disputed ideas, and those with no studies to support them, as [*66] well as those in the mainstream. n78 Moreover, by presenting evidence supporting the value of diversity, defenders of RCDM suggest that it is a question of fact rather than a question of law, and that its value must be proven in each instance. Defendants in the University of Michigan cases have certainly accumulated evidence in an effort to factually support the educational benefits of diversity, but what happens in a future case if the defendants do not accumulate such evidence?

A third litigation problem from Justice Powell’s opinion is its assumption that racial diversity will lead to intellectual diversity, and the unstated stereotyping corollary that people of a particular race have a particular perspective. Defenders of RCDM have rejected the proposition that such a stereotype could be true, while maintaining the more defensible position that race has a significant influence on perspective. n79 They also have asserted that diverse viewpoints within a given race provides significant benefits because students learn of this intraracial diversity. n80 But this argument seems to treat college and graduate students as if they were ignoramuses, unaware of the fact that not all African Americans (or Hispanics or Asian Americans, etc.) think alike. n81 Even if such ignoramuses were being admitted in droves to the elite schools of our nation, one would think that the lesson of intrarace heterogeneity could be taught in many other ways (like, for example, having students read the decisions of Justice [*67] Marshall and Justice Thomas). And if it cannot be taught that simply, then it probably cannot be taught at all.

Supporters of RCDM must also deal with the holding of Bakke: that a set-aside of seats, where members of a particular race cannot compete for the seats, is illegal under Title VI. Having a literal set-aside is usually fairly easy to avoid. n82 But most institutions will have some objective for the matriculation of minority students, usually stated in terms of “critical mass.” n83 At this point, the line between a “critical mass” (which sounds a lot like a minimum) and a set-aside begins to blur.

Finally, Justice Powell’s opinion forces universities to explain which groups they have included and/or excluded. Some schools will not include all minority racial groups because, for example, Asian Americans do not require a preference to be admitted in reasonable numbers. It is unclear to date whether “admitted in reasonable numbers without a preference” will sustain a racially selective admissions program, and, even if it does, a clever plaintiff will be able to find some racial group that is not present in significant numbers. n84 Other programs do include preferences for Asian Americans, but do not include a preference for other diversifying characteristics like sexual orientation on the ground that homosexuals are admitted in reasonable numbers without a “diversity” preference. n85 It seems safe to say that the question of which groups to include has not yet been completely resolved.

Both sides of the RCDM debate have a problem explaining their choices for compelling interests in a coherent way. Supporters [*68] of RCDM must explain why education is so different from criminal law that we are willing to tolerate RCDM in the former, but not in juries (assuming they do not wish to advocate RCDM in jury selection, which most do not). n86 Conversely, those who attack RCDM must explain why “remedying identified past discrimination” is the only compelling interest (or, for that matter, a compelling interest at all). RCDM, after all, means providing preferences to those who are not necessarily victims of that past discrimination. n87 Why precisely should non-victims receive a preference, and if they should, then why is better education (or juries or child rearing) not just as important a reason as making sure that a given state body has the same racial make-up it hypothetically would have had in the absence of discrimination? n88

Opponents of RCDM have two other problems they must face, albeit ones not directly related to Bakke. The first is responding to the defenders’ argument that maintaining high academic standards along with the elimination of RCDM will reduce dramatically the number of African Americans and other minorities attending elite institutions. n89 Ironically, those attacking RCDM will frequently point out that universities are not obligated to maintain their emphasis on grades and standardized test scores, and would be able to attain more diversity if they emphasized those [*69] criteria less. n90 While this is true, it usually is the case that those attacking RCDM have no great predeliction for lowering standards.

The lowering of standards raises yet another problem: the “race-neutral” alternative. The Supreme Court has occasionally asserted in other contexts that narrow tailoring requires that the state actor consider “race-neutral alternatives.” n91 But a “race-neutral alternative”–for example, creating a minimum academic standard and choosing students based upon their nonracial diversity characteristics–may be just as bad as racial preferences if the purpose behind the standard lowering is simply to increase the percentage of minorities at the institution. n92 Indeed, it may be worse since it simply tries to achieve the same goal in an inefficient way and indirect way. n93 In a sense, the “raceneutral alternative” argument echoes the worst parts of Justice Powell’s opinion–that using race in a surreptitious way is better than using it in an overt and candid way. n94 [*70]

V. The Consequences of Adopting the “Diversity” Rationale

The Powell diversity paradigm has had significant negative effects. Its lack of clarity about means has led colleges and universities to use race in extraordinary ways. Its rejection of increasing minority presence in the medical field as a goal and its focus on diversity has led university officials to deny or hide what they are really trying to do–which, in the end, is simply putting more minorities into the professions and the middle class. Most importantly, it has eliminated the fundamental moral basis for RCDM, and replaced it with a utilitarian model in which minority students are simply used as pawns to better the educational experience for others. n95 The increasing unpopularity of RCDM has followed. People can understand the moral argument, whether they agree or disagree with it. I tend to think that many fewer understand the utilitarian argument. n96

The rule against race discrimination was never really a utilitarian-based rule. n97 As Professor Epstein has demonstrated, discrimination can frequently be rational, particularly where people’s utility functions have a significant element of prejudice or a fear of those different. n98 Despite that, we have reached a fundamental moral consensus in this country–perhaps because our [*71] history has taught us that we cannot be trusted with making ad hoc utilitarian decisions about the propriety of discrimination in specific instances–that RCDM is wrong. In contrast, Justice Powell’s Bakke opinion states that ad hoc decisions are proper, so long as they are made by our educated elite. This will not do. To convince the public that there should be exceptions to that general moral rule of non-discrimination, a moral claim of competing weight must be offered. There may be such claims out there, but they are not in Justice Powell’s Bakke opinion.


n1 In this article, and as a general rule, I try to avoid using the phrase “affirmative action” because its definition is so elastic. See Lungren v Superior Court, 55 Cal Rptr 2d 690, 694 (App 1996) (“The term ‘affirmative action’ . . . is rarely defined . . . so as to form a common base for intelligent discourse.”), quoting Dawn v State Personnel Board, 154 Cal Rptr 186, 190 (App 1979) (Paras concurring). See also Minnick v California Department of Corrections, 452 US 105, 128 (1981) (Stewart dissenting from denial of certiorari) (“[Under the Equal Protection Clause,] a sovereign State may never [engage in racial discrimination], and it is wholly irrelevant whether . . . the discrimination is called ‘affirmative action’ or by some less euphemistic term.”) Although I have occasionally (in my role as a litigator challenging such programs) used the phrase “reverse discrimination,” the word “discrimination” frequently (perhaps unfairly) has pejorative connotations. Accordingly, I will use “race-conscious decision-making” in this piece.

n2 Cal Const Art I, § 31 (codifying Proposition 209). See also Washington Civil Rights Act, Rev Code Wash § 49.60.400 (2002) (codification of Initiative Measure No 200).

n3 438 US 265, 312-15 (1978).

n4 President Lyndon B. Johnson, Commencement Address at Howard University: To Fulfill These Rights (June 4, 1965), available online at johnson/archives.hom/speeches.hom/650604.asp> (visited Oct 7, 2002) [on file with U Chi Legal F]. See also Gabriel J. Chin, ed, 1 Affirmative Action and the Constitution: Affirmative Action Before Constitutional Law 1964-1977 21-26 (1998), quoted in Kenneth Jost, Affirmative Action, 11 Cong Q 739, 745 (Sept 21, 2001).

n5 In Lutheran Church-Missouri Synod v FCC, 141 F3d 344 (DC Cir 1998), the FCC justified licensing requirements for RCDM in radio station employment as promoting “diversity.” The D.C. Circuit felt that this showed how much burden the term ‘diversity’ has been asked to bear in the latter part of the 20th century in the United States. It appears to have been coined both as a permanent justification for policies seeking racial proportionality in all walks of life (‘affirmative action’ has only a temporary remedial connotation) and as a synonym for proportional representation itself.

Id at 356. See also Alan M. Dershowitz and Laura Hanft, Affirmative Action And The Harvard College DiversityDiscretion Model: Paradigm Or Pretext, 1 Cardozo L Rev 379, 404 (1979) (“The concept of ‘diversity’ is so vague that it lends itself to a myriad of widely divergent and ever-changing definitions capable of masking the criteria actually at work”); Samuel Issacharoff, Bakke In the Admissions Office and the Courts: Can Affirmative Action Be Defended?, 59 Ohio St L J 669, 677-78 (1998) (noting that “one of the clear legacies of Bakke has been to enshrine the term ‘diversity’ within the legal lexicon to cover everything from curricular enrichments to thinlyveiled set-asides”); id at 678 (“The problem with diversity as a justification for a challenged affirmative action program is that it is an almost incoherent concept to operationalize, unless diversity means a predetermined number of admittees from a desired group.”).

n6 See Washington Post/Kaiser/Harvard Racial Attitudes Survey, Question 50, available online at htm> (“Post Survey Web Site”) (visited Oct 17, 2002) [on file with U Chi Legal F]. See also The Center for Individual Rights, No Retreat: The Alamo of Affirmative Action, available online at (visited Oct 17, 2002) [on file with U Chi Legal F]. Additional breakdowns from the poll data can be obtained on the first web site through interactive processes. Some of the data on the second web site were obtained in this way. Conversation with Curt Levey of the Center for Individual Rights (Aug 9, 2002).

n7 Post Survey Web Site, Question 51 (cited in note 6). The racial breakdown in responding to this question was somewhat starker than the question concerning whether race should be a factor. 49 percent of whites, 77 percent of African Americans, 62 percent of Hispanics, and 63 percent of Asians supported making an “extra effort” to recruit. There may have been some confusion over the wording of this question. In a related article, the Post quoted an individual who it said purportedly opposed outreach programs, but who said: “If they want to recruit minorities, fine, as long as an equally qualified white isn’t replaced. If that’s a problem, make the school bigger.” See Richard Morin, Misperceptions Cloud Whites’ View of Blacks, Wash Post A1 (Jul 11, 2001).

n8 Post Survey Web Site, Question 52b (cited in note 6). Significant majorities of every racial/ethnic category opposed using race for this purpose: 90 percent of whites, 70 percent of African Americans, 83 percent of Hispanics, and 78 percent of Asians.

n9 Morin, Misperceptions Cloud Whites’ View of Blacks, Wash Post at A1 (cited in note 7). The primary focus of the article reporting the poll results was that whites who were informed about the condition of racial and ethnic minorities were more likely to support an obligation upon government to insure that African Americans and other minorities were treated equally and fairly in public schools, by the police, etc. Id (“Misinformed whites were far less likely to view black problems as being serious, or to favor government action to correct persistent social and economic disparities.”).

n10 Id. The author did not provide exact numbers for African Americans, said nothing at all about the responses of the other minority groups surveyed, and reported only that there was little difference between informed and uninformed whites. Id.

n11 Morin, Misperceptions, Wash Post at A1 (cited in note 7).

n12 See Jost, Affirmative Action at 748 (cited in note 4) (quoting Professor John Jeffries of the University of Virginia Law School to the effect that “every public institution in America . . . takes racial diversity in admissions”). Peter H. Schuck, Affirmative Action: Past, Present, and Future, 20 Yale L & Pol Rev 1, 9 (2002) (“no other domain practices and supports [affirmative action] so enthusiastically”). Certainly, the litigants who defend such programs take that position. In their brief to the Sixth Circuit, defendants in Gratz v Bollinger, 277 F3d 803 (6th Cir 2001), cert granted at 2002 US LEXIS 8681, a case challenging the consideration of race in a system of undergraduate admissions at the University of Michigan, wrote (perhaps a bit hyperbolically) that “this case presents an issue of great national importance whose resolution will affect the admissions program of every public and private institution of higher education in this country.” Final Brief of Appellees (01-1333 & 01-1418) and Appellants (01-1416) at viii, Gratz v Bollinger, No 011333, 01-1416 & 01-1418, (6th Cir Jul 30, 2001), available online at admissions/legal/gratz/gratzappeal.html> (visited October 18, 2001) [on file with U Chi Legal F]. See also Issacharoff, 59 Ohio State L J at 670 (cited in note 5) (noting that the admissions program of University of Texas Law School “mirrored that of most other comparable institutions”); Lee Bollinger and Nancy Cantor, The Educational Importance of Race, Wash Post A17 (Apr 28, 1998) (“Virtually all institutions with selective admissions consider race when they choose among applicants.”).

n13 With respect to the amounts of money spent by RCDM’s defenders see Liz Cobbs, New U-M Attorney A Winner in High Court, Ann Arbor News (Sept 5, 2002), available online at /base/news-1/1031236849188085.xml> (visited Oct 17, 2002) [on file with U Chi Legal F] (reporting that the University of Michigan has spent $ 8.5 million in legal fees defending its admissions systems). With respect to aspersions against their opponents, see, for example, Defendant-Intervenors’ Brief In Support Of Defendants’ Motion For Summary Judgment and In Opposition to the Plaintiff ‘s Motion For Summary Judgment at 2, Grutter v Bollinger, No 97-CV-75928-DT (E D Mich filed Nov 20, 2000), available online at (visited Nov 30, 2002) [on file with U Chi Legal F] (“DefendantIntervenors’ Brief “) (“Plaintiff asks the Court openly to spurn the past gains and present claims of the Civil Rights Movement”); id at 11 (citing deposition of UM President Lee Bollinger for the proposition that eliminating RCDM “would result in the ‘resegregation’ of the UM Law School and of higher education in general”); id at 24 (describing plaintiff ‘s position as a “racist view of equality”); id (“Plaintiff appeals to other privileged whites to join her in falsely equating index scores with merit”); Defendant-Intervenors’ Brief at 30 n 17 (noting that an organization established “to support eugenics and the ‘Caucasian race'” had given money to the Center for Individual Rights); id at 35 (“Plaintiff . . . would have the Court impose a racist double standard on the UM Law School and the students who apply to go there”); id (asserting that plaintiff ‘s arguments are characterized by “crassness, superficiality, and the presumption of black inferiority”).

n14 Indeed, my fellow panelist at the University of Chicago Legal Forum’s 2001 Symposium, “The Scope of Equal Protection,” Professor Samuel Issacharoff, has argued passionately that it is the very elitism of the elite public institutions that should be considered a “compelling interest” (along with the interest in integrating racial minorities into society as a whole). Issacharoff, 59 Ohio St L J at 684 (cited in note 5). RCDM, argues Issacharoff, is the only way to achieve reasonable amounts of integration while maintaining the high academic standards that elitism requires. Id. My guess is that Professor Issacharoff ‘s factual assertion is probably true, although I am not as convinced of his corollary prediction that an elite school that maintains high academic standards in the face of a directive to use race-neutral admissions standards quickly will lose public support. Further, his insistence that maintaining elite public institutions of higher education is a compelling governmental interest is questionable. In any event, though, this is plainly a defense of RCDM in modern admissions policies that is far more attuned to reality than Justice Powell’s. See also Dershowitz and Hanft, 1 Cardozo L Rev at 419 (cited in note 5) (increasing the number of minority group members in the universities and professions “surely is the concern that actuates the movement for university affirmative action today”).

n15 Professor Schuck surveys other opinion polls, and arrives at a similar conclusion concerning its unpopularity, concluding that “no researcher in this field doubts, however, that the public’s opinion remains decidedly and intensely negative.” Schuck, 20 Yale L & Pol Rev at 56 (cited in note 12). He also notes that opposition to RCDM seems to span the political spectrum. Id at 57. See also Richard A. Posner, The Bakke Case and the Future of Affirmative Action, 67 Cal L Rev 171, 172 (1979) (vast majority of American people and almost 2/3 of all non-whites would have been pleased with a decision in Bakke outlawing reverse discrimination broadly).

n16 Judge Laurence H. Silberman, who served in the Nixon Labor Department and had significant responsibility for formulating the “Philadelphia Plan” that crystallized RCDM in federal contracting, reports that “although other minorities were nominally included in our affirmative action efforts, it was the plight of American blacks that drove the whole policy.” Laurence H. Silberman, The Origin of Affirmative Action as We Know It–The Philadelphia Plan Pivot 10 (Federalist 2001) (addressing the Labor & Employment Practice Group of the Federalist Society for Law & Public Policy, October 10, 2001). See also Schuck, 20 Yale L and Pol Rev at 59 (cited in note 12) (attention in early years was entirely on African Americans); Terrance Sandalow, Book Review, Identity and Equality: Minority Preferences Reconsidered, 97 Mich L Rev 1874, 1875 n 2 (1999) (reviewing William G. Bowen and Derek Bok, The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions (Princeton 1999)) (“The policies were initially adopted because of, and I believe the primary impetus for their attention remains, the perceived importance of increasing black enrollment in colleges and universities”).

n17 See Mark S. Nadel, A Four-Element Affirmative Action Alternative to Racial Preferences 22 n 136 (manuscript on file with author).

n18 See generally Stephan and Abigail Thernstrom, America in Black and White: One Nation, Indivisible; Race In Modern America 183-202 (Touchstone 1997). Schuck, 20 Yale L & Pol Rev at 62 (cited in note 12) (progress has been “‘nothing short of astonishing'”), quoting Orlando Patterson, The Ordeal Of Integration: Progress and Resentment in America’s Racial Crisis 15 (Counterpoint 1997).The Post article discussed earlier in this section reported that “blacks have made dramatic progress in many, if not most areas of American life. There have never been more blacks in the middle class or a larger share who have graduated from high school, gone to college, or entered professional schools.” Morin, Misperceptions, Wash Post at A1 (cited in note 9). The article quoted Keith Reeves, a political scientist at Swarthmore College, to the effect that increasing and visible African American success may be the reason for the white misperceptions that the poll reported. Id. Indeed, even in 1979, Dershowitz and Hanft asserted that “not all minority group members have been educationally disadvantaged,” a proposition which is surely even more true today than it was then. 1 Cardozo L Rev at 416 (cited in note 5).

n19 Dershowitz and Hanft, 1 Cardozo L Rev at 416 (cited in note 5) (“The fact that certain advantaged minority persons who benefit under race-specific programs would no longer receive windfall benefits under a race-neutral program [assisting the socioeconomically disadvantaged] should not be cause for distress; these are precisely the persons who do not–under any principle of morality–deserve to be given any special advantage.”). RCDM supporters repeatedly assert that (not surprisingly) using socioeconomic status as a substitute diversity factor would not achieve the same racial diversity as using race directly. See, for example, Grutter, 288 F3d at 764 (Clay concurring). If so, that can only be because some minority beneficiaries of RCDM would not be the beneficiaries of a program emphasizing socioeconomic diversity–because they are not socioeconomically diverse (i.e., poor).

n20 Mark C. Niles, Clarence Thomas: The First Ten Years Looking for Consistency, 10 Am U J Gender Soc Pol & L 327, 338 n 53 (2002). Thomas has denied at least part of this contention. Id at 339 & n 54.

n21 See Dershowitz and Hanft, 1 Cardozo L Rev at 381 (cited in note 5) (noting prophetically that the Bakke decision “will be all the more important if other guidance is not forthcoming from the Court”); Anthony T. Kronman, Is Diversity a Value in American Higher Education?, 52 Fla L Rev 861, 861 (2000) (noting that the word “diversity” has become increasingly important in a generation); Samuel Issacharoff, Law And Misdirection in the Debate Over Affirmative Action, 2002 U Chi Legal F 11:

I suggest that diversity came to its current life with the narrow window left open by Justice Powell’s opinion in Bakke. Assuming that swing opinion to have the force of the Court, the defenses of affirmative action were limited to either the internal claim of educational product enhancement or the retrospective claim of remedying institutional discrimination.

Id at 16 & n 29 (footnotes omitted).

n22 Bakke, 438 US at 312-15.

n23 Id at 269-72.

n24 Adarand Constructors, Inc v Pena, 515 US 200, 218 (1995) (“Bakke did not produce an opinion for the Court.”). In Alexander v Sandoval, 532 US 275 (2001), the four dissenters in Adarand made a similar (if not more trenchant) observation, viz., that the Bakke majority for overturning the injunction against using race was “divided over the application of the Equal Protection Clause–and by extension Title VI–to affirmative action cases. Therefore, it is somewhat strange to treat the opinions of those five Justices in Bakke as constituting a majority for any particular substantive interpretation of Title VI.” Id at 308 n 15 (Stevens dissenting).

n25 Bakke, 438 US at 312 (Powell).

n26 Id at 313-315.

n27 Id at 315.

n28 See note 24 and accompanying text. Lower court judges have disagreed about its precedential value. Compare Grutter v Bollinger, 288 F3d 732 (6th Cir 2002) (5-4 majority of en banc Sixth Circuit concluding that Powell’s opinion had precedential value under doctrine of United States v Marks, 430 US 188 (1977)), cert granted at 2002 US LEXIS 8677; Smith v University of Washington Law School, 233 F3d 1188, 1200 (9th Cir 2000) (upholding precedential value of Powell’s rationale); and Gratz v Bollinger, 122 F Supp 2d 811, 822 (E D Mich 2000) (somewhat ambiguous with respect to precedential value of Powell’s rationale, but upholding diversity as a compelling governmental interest), with Grutter 288 F3d at 776-77 (Boggs dissenting) (concluding that Powell’s opinion is not precedential); Johnson v Board of Regents of the University of Georgia, 263 F3d 1234, 1249 (11th Cir 2001) (same); Hopwood v Texas, 78 F3d 932, 944 (5th Cir 1996) (same); Hopwood v Texas, 236 F3d 256, 274-75 (5th Cir 2000) (adhering to earlier Hopwood decision and explicitly disagreeing with Ninth Circuit’s Marks analysis in Smith); Grutter v Bollinger, 137 F Supp 2d 821, 847 (E D Mich 2001) (holding that Powell’s opinion is not precedential), revd, 288 F3d 732 (6th Cir 2002); and Peters v Moses, 613 F Supp 1328, 1335 (W D Va 1985) (same).

n29 Bakke, 438 US at 312 (Powell) (“The freedom of a university to make its own judgments as to education includes the selection of its student body.”).

n30 Id at 314.

n31 Id at 315.

n32 Id.

n33 Bakke, 438 US at 315 (Powell).

n34 Id at 316-17 (discussing Harvard College’s admissions program). Professor Alan Dershowitz and Laura Hanft already have described the irony of adopting an admissions system whose original focus on “wellrounded” students had its origins in a system to reduce the number of Jews attending Harvard. Dershowitz and Hanft, 1 Cardozo L Rev at 386-99 (cited in note 5).

n35 See Issacharoff, 2002 U Chi Legal F at 37-38 (cited in note 21) (no serious scholar would argue that an empirical claim that homogeneity benefits educational outcomes could be used to consider race as a means of reinforcing segregation).

n36 427 US 160 (1976).

n37 Hopwood, 78 F3d at 943 n 25.

n38 See Dershowitz and Hanft, 1 Cardozo L Rev at 390-91 (cited in note 5). The Admissions Chair described the policy as follows:

Race is part of the record. It is by no means the whole record and no man will be kept out on grounds of race; but those racial characteristics which make for race isolation will, if they are borne by the individual, be taken into consideration as a part of that individual’s characteristics under the test of character, personality, and promise.

That if there should result in fact any substantial change in the proportion of groups in the College following application of the test, this will be due, not to race discrimination or any quota system, but to the failure of particular individuals to possess as individuals those evidences of character, personality and promise which weighed with other evidences render them more fit than other individuals to receive all that Harvard has to offer. Of course there will be criticisms. It will be said that Harvard is discriminating on grounds of race. That will not be true.

The New Admissions Plan, The Gadfly 4 (May 1926), quoted in id at 391. See also Kronman, 52 Fla L Rev at 874 (cited in note 21) (“At the beginning of the twentieth century, the goal of liberal education was widely understood to be the production of cultured gentlemen.”).

n39 The ultimate Harvard plan in the 1920s rejected discrimination, but involved an increased effort to “nationalize” Harvard by making the school more accessible to students from the South and West. Dershowitz and Hanft describe this as “perpetuating the white Protestant homogeneity of its student body.” 1 Cardozo L Rev at 398 (cited in note 5). They assert that “the Midwestern farm boy and the Southern lawyer’s son were admitted not so much because of the diversity they might provide but primarily because they were seen as far more similar to typical Harvard students than was the first generation immigrant from New York.” Id at 400. They do concede that those geographically unusual students did “add a modicum of diversity by way of some different values,” id, and, of course, Harvard could have exercised its academic freedom to avoid even that modicum. That is, Harvard was seeking a kind of diversity, even if some might question its motive.

n40 476 US 267 (1986).

n41 Id at 313 (Stevens dissenting).

n42 Id.

n43 Id at 316 (Stevens dissenting).

n44 Wygant, 476 US at 316 (Stevens dissenting). Thus, Justice Stevens abandoned the suggestion that the “exclusionary” decision could be a school board’s legitimate effort, for example, to reduce racial tensions that were interfering with the educational process. Long before Justice Stevens, Professor Van Alstyne took the hypothetical much more seriously. Writing in opposition to any sort of “compelling interest” at all, Van Alstyne demonstrates how simple it would have been for legislatures to devise a compelling interest for segregated schools, and how difficult it would be for an honest court to either disprove the “compelling” nature of the interest or demonstrate that it was a pretext, had the compelling interest test been widely known prior to Brown v Board of Education, 337 US 443 (1954). William Van Alstyne, Rites of Passage: Race, the Supreme Court, and the Constitution, 46 U Chi L Rev 775, 796-798 (1979).

n45 Wygant, 476 US at 313 (Stevens dissenting). Of course, the reference is to the second paragraph of the Declaration of Independence, and is one of the “selfevident truths” to which that document refers. It has been generally understood that the “equality” to which the Declaration refers is an equality before the law, an equality of rights. Darlene C. Goring, Private Problems, Public Solution: Affirmative Action in the 21st Century, 33 Akron L Rev 209, 238-39 (2000) (tracing Justice Powell’s concept of “equality” to the Declaration and describing debates around the time of the Civil War Amendments to the Constitution that refer to the Declaration as providing equal rights).

n46 See Goring, 33 Akron L Rev at 238-39 (cited in note 45).

n47 Wygant, 476 US at 313 (Stevens dissenting).

n48 Id at 315 (Stevens dissenting).

n49 Alternatively, race-based jury selections may serve the same interest as race-based selection of judges: assuring minorities of the fairness of the system. See Schuck, 20 Yale L & Pol Rev at 31 (cited in note 12) (“For minorities to accept their outcomes as minimally just or at least acceptable, they must view these institutions as inclusive and procedurally fair. Being tried by minority judges, for example may advance that goal.”); Sandalow, 97 Mich L Rev at 1912 (cited in note 16) (black lawyers, judges, and police officers may be needed to provide assurances of fairness and inclusion to blacks). The American Bar Association makes such an argument to support RCDM in higher education, and indeed, analogizes to the importance of juries not excluding racial minorities. See Brief of the American Bar Association, as Amicus Curiae at 23-25, Grutter v Bollinger, 288 F3d 732 (6th Cir 2002). It does not assert that we should consciously include racial minorities in juries (through RCDM), which would be the more relevant analogy in my view. See also United States v Allen-Brown, 243 F3d 1293, 1298-99 (11th Cir 2001) (defense attorney’s effort to obtain more diversity on predominantly-white jury by using peremptory challenges was not in pursuit of a sufficiently compelling reason and hence unconstitutional).

n50 Wygant, 476 US at 315 (Stevens dissenting). Justice Stevens also asserted that race could never rationally be used to determine who would be a fit parent. Id at 313. This, of course, is hardly a noncontroversial assertion. Adoption agencies do sometimes use the race of an adopting couple as a consideration in determining the most appropriate home. Indeed, they have been known to prevent a cross-racial adoption even if no other couple is ready to adopt the child. See, for example, Cynthia R. Mabry, “Love Alone Is Not Enough!” in Transracial Adoptions: Scrutinizing Recent Statutes, Agency Policies, and Prospective Adoptive Parents, 42 Wayne L Rev 1347, 1359 (1996) (noting such instances as a motivation for the Multiethnic Placement Act). Indeed, in Palmore v Sidoti, 466 US 429 (1984), the case cited by Justice Stevens for the proposition that race cannot be used to determine who is a fit parent, see Wygant, 476 US at 313 n 5, the lower courts had explicitly determined that the “best interests of the child” would not be served by placing a child with his mother, who had remarried interracially, because of the social stigma that interracial marriages would entail. The Court in Palmore did not dispute the factual findings of the lower court. Indeed, the Court agreed that “there is a risk that a child living with a stepparent of a different race may be subject to a variety of pressures and stresses not present if the child were living with parents of the same racial or ethnic origin.” Palmore, 466 US at 433. Thus, contrary to Justice Stevens’s reinterpretation of Palmore, the Court never suggested that it was “utterly irrational” to consider race in such situations because “it is completely unrelated to any valid public purpose.” Wygant, 476 US at 313 (Stevens dissenting). Rather, the Court held that the valid public purpose had to give way to the more important constitutional rule of equal treatment, primarily because society could not allow private biases to influence the outcome under that valid public purpose.

n51 Perhaps recognizing this dilemma, Dean Kronman focuses his defense of diversity on the improvements in classroom learning, and not learning that takes place outside of the classroom, that it provides. Kronman, 52 Fla L Rev at 896 (cited in note 21) (“If one want to know why diversity of experience and values is an educational good not just in the general sense that interaction with others from different backgrounds is an occasion for learning in any organizational setting, but in the more specific sense that it contributes to the distinctive goals of our colleges and universities, it is necessary to ask how such diversity promotes the specialized activity of disciplined instruction for whose sake these institutions exist, and to shift our attention from the dorm to the classroom.”). But to say that classroom learning is unique to the educational goals of universities hardly demonstrates that it is more compelling than “dorm learning” or “jury learning” or any other kind of learning.

n52 See generally Eugene Volokh, Diversity, Race As Proxy, and Religion As Proxy, 43 UCLA L Rev 2059 (1996) (demonstrating that assumption that race can be used as a proxy for intellectual diversity can justify RCDM in a wide variety of situations).

n53 I identify here only the problems with the substantive rationale for RCDM. I have discussed elsewhere other jurisdictional and procedural problems, most of which (again) are fairly well-known. See Michael E. Rosman, The Error of “Hopwood’s Error”, 29 J of L & Educ 355, 357-58 (2000).

n54 This was certainly a concern for the Hopwood court. Hopwood, 78 F3d 932, 946 (5th Cir 1996) (“To believe that a person’s race controls his point of view is to stereotype him.”).

n55 Bakke, 438 US at 379 (Brennan concurring in the judgment in part and dissenting in part) (there was no basis “for preferring a particular preference program simply because in achieving the same goals that Davis is pursuing, it proceeds in a manner that is not immediately apparent to the public.”). A number of lower courts seem to agree with Justice Brennan on this point, and have concluded that there really is no difference between a “plus” system and a set-aside. See Middleton v City of Flint, 92 F3d 396, 412-13 (6th Cir 1996) (“We note that quotas and preferences are easily transformed from one into the other.”), citing Bakke, 438 US at 378 (Brennan, concurring and dissenting); Hopwood, 78 F3d at 948 n 36 (noting that “even if a ‘plus’ system were permissible, it likely would be impossible to maintain such a system without degeneration into nothing more than a ‘quota’ program”), citing Bakke, 438 US at 378 (Brennan concurring in the judgment in part and dissenting in part); Valentine v Smith, 654 F2d 503, 510 n 15 (8th Cir 1981) (“Any distinction between goals, quotas, and targets is primarily semantic.”), citing Bakke, 438 US at 378 (Brennan concurring in the judgment in part and dissenting in part).

n56 Bakke, 438 US at 318 (Powell).

n57 Justice Powell quoted the amicus brief of various institutions, which included an appendix describing the policy at Harvard College:

In Harvard College admissions the Committee has not set target-quotas for the number of blacks, or of musicians, football players, physicists or Californians to be admitted in a given year . . . But that awareness [of the necessity of including more than a token number of black students] does not mean that the Committee sets a minimum number of blacks or of people from west of the Mississippi who are to be admitted. It means only that in choosing among thousands of applicants who are not only ‘admissible’ academically but have other strong qualities, the Committee with a number of criteria in mind, pays some attention to distribution among many types and categories of students.

Id at 316, quoting Appendix to Brief for Columbia University, Harvard University, Stanford University, and the University of Pennsylvania, as Amicus Curiae, Regents of the University of California v Bakke, No 76811, *2-3 (filed Jun 7, 1977), reprinted in Philip B. Kurland and Gerhard Casper, eds, 99 Landmark Briefs of the Supreme Court of the United States 689 (Univ Pubs of Am 1978). Id at 323 (Appendix to opinion of Justice Powell) (Harvard College Admissions Program) (“A truly heterogeneous environment that reflects the rich diversity of the United States, . . . cannot be provided without some attention to numbers.”).

n58 Dershowitz and Hanft note that, in practice, the Harvard College admissions program yielded an uncanny stability in the proportion of African Americans in each class. 1 Cardozo L Rev at 383 n 13 (cited in note 5) (noting that the percentage of African Americans in Harvard classes from 1973 through 1981 was 7 percent in every year but one, when it was 8 percent).

n59 Bakke, 438 US at 318 (Powell).

n60 Id at 318-19.

n61 Id at 319 n 53.

n62 Id. Thus, if these words were taken seriously, review of “plus” programs might be similar to the review of the use of race in redistricting programs, where strict scrutiny is not even reached unless race is a “predominant” factor. Hunt v Cromartie, 121 S Ct 1452, 1464 (2001) (“‘Strict scrutiny does not apply merely because redistricting is performed with consciousness of race'”), quoting Bush v Vera, 517 US 952, 958 (1996) (O’Connor’s principal opinion). To my knowledge, all the courts that have analyzed RCDM in higher education apply strict scrutiny, and thus implicitly reject the analogy (or, conversely, assume that race is a predominant factor). Nonetheless, Representative John Conyers (D-Mich), among others, submitted an amicus brief in the Grutter case that argued by analogy to the redistricting cases. Amicus Brief of John Conyers, et al., at 9-11, Grutter v Bollinger, 288 F3d 732 (6th Cir 2002).

n63 Of course, this too is entirely incomprehensible. How can the consideration of race as a factor, even if considered along with other factors, not reflect an intention to discriminate (in the sense of differentiate)? Se