The United States Supreme Court
The University of Michigan Admissions Lawsuits

Grutter v. Bollinger

Reply Brief of Counsel for the Respondent, the University of Michigan (pp 1-12, 14-38, 43-50): The Law School has a compelling interest in the use of race in admissions to secure the educational benefits that flow from student body and racial diversity. The Law School’s consideration of race is individualized, competitive, modest in scope, and does not impose an undue burden on non-minority applicants. (The complete Brief is online here.)

Comments by Curtis Crawford,* indented in italics.


There is no genuine dispute in this case about the relevant historical facts, and the record evidence establishes three central realities that provide firm support for the Sixth Circuit’s conclusion that the admissions program in use at the University of Michigan Law School ("the Law School") does not violate the Equal Protection Clause. First, academic selectivity and student body diversity, including racial diversity, are both integral to the educational mission of the Law School. Second, the Law School successfully realizes both goals through an admissions program that is "virtually indistinguishable" from the Harvard plan that five Justices approved in Bakke. 1 It evaluates the potential contributions and academic promise of every individual and does not employ quotas or set-asides. Third, no honestly colorblind alternative policy could produce educationally meaningful racial diversity at present without enrolling students who are academically unprepared for the rigorous legal education that the Law School offers.

There is accordingly no way for this Court to reverse the Sixth Circuit’s decision without "break[ing] … new ground." U.S. Br. at 10. This Court must instead decide whether, consistent with Bakke, the finest law schools throughout the country may continue to train this Nation’s leaders in integrated classrooms—as they have done so effectively for the past three decades—or *whether they now must choose between maintaining academic distinction and avoiding very substantial resegregation.* 2

 * The word, segregation, implies separation or exclusion based on race. What petitioner demands is the opposite: that the Law School cease to prefer, separate or exclude based on race. If petitioner wins, fewer black and Hispanic applicants will be admitted, not because of their race or ethnicity, but because (as respondent concedes) of their lower college grades and LSAT scores.

[-2-] 1. The Law School is among the Nation’s leading law schools. It has achieved that preeminence by *carefully selecting* and training students of exceptional promise to serve as leaders of the profession and of our Nation. 3 The Law School has determined that effective pursuit of this mission requires a curriculum that "firmly links professional training to the opportunity for reflection about many of our most fundamental public questions, such as … the effects of religious, racial and gender intolerance in our culture" (CAJA 1658), and integrated classes comprising a "mix of students with varying backgrounds and experiences who will respect and learn from each other," each of whom is "among the most capable students applying to American law schools in a given year" and has a "strong likelihood of succeeding in the practice of law and contributing in diverse ways to the well-being of others." JA 110.

 * According to Note 3, out of 3500 applicants the Law School typically admits 1300 (37%), of whom 350 (27%) enroll.

Neither the petitioner nor the United States questions the legitimacy or importance of these goals to the Law School’s mission. Extensive (and virtually unchallenged) evidence confirms that a racially diverse student body helps students to develop the interpersonal skills necessary to "work more effectively and more sensitively" in a world that "is and will be multi-racial" (CAJA 2243); helps to dispel historic stereotypes (CAJA 7697-99); and introduces students to unfamiliar experiences and perspectives to promote the "mutual respect" and "sympathetic engagement with the experiences of other people that are basic to the mature and responsible practice of law" (CAJA 5106). See infra pp. 21-26. The evidence also proves that fully realizing these benefits requires "meaningful numbers" or a "critical mass" of minority students (JA 120)—enough to create significant opportunities for personal interaction, to show that there is no consistent "minority viewpoint" on [-3-] particular issues, Pet. App. 215a, and to ensure that "minority students do not feel isolated or like spokespersons for their race, and feel comfortable discussing issues freely based on their personal experiences." Id. at 28a. Based on the persuasive weight of the educational evidence, the district court emphasized that it "d[id] not doubt that racial diversity in the law school population" promotes "cross-racial understanding," helps to break down racial stereotypes, "enables [students] to better understand persons of different races and better equips them to serve as lawyers in an increasingly diverse society and an increasingly competitive world economy." Id. at 246a. The court also acknowledged that the benefits of diversity are "important and laudable," because "classroom discussion is livelier, more spirited, and simply more enlightening and interesting" when the students have "the greatest possible variety of backgrounds." Id. at 246a, 244a. Indeed, petitioner acknowledged that "no one is contesting that there are educational benefits of diversity. It’s simply not an issue in the case." CAJA 7192.

2. Petitioner and the United States nonetheless challenge the admissions policy that has been carefully crafted to achieve meaningful educational diversity. In 1992, the Law School charged a distinguished committee of legal scholars to design a policy that would further its mission and comply with this Court’s decision in Bakke. CAJA 7486-87, 7546-47. The policy they designed—like the Harvard plan it was modeled on—openly acknowledges that the racial background of a minority applicant can be one of many factors relevant to the admissions decision. JA 121.

Petitioner’s own expert conceded, however, that race is not the predominant factor under that policy. Tr. 2:211-13. Instead, its hallmark is a focus on academic capabilities coupled with a flexible assessment of every individual applicant’s talents, experiences and potential "to contribute to the learning of those around them." JA 111.

First, the policy requires the director of admissions, in consultation with the faculty, to evaluate each applicant based on all of the information available in the file. JA 114- [-4-] 21. The Law School does not use any formulas or set criteria for admission. The policy requires careful consideration of an applicant’s undergraduate grades and LSAT score because they are important (though imperfect) predictors of academic success in law school, and the "minimal criterion is that no applicant should be admitted unless we expect that applicant to do well enough to graduate with no serious academic problems." 4 But "[t]here is no combination of grades and test scores … below which an applicant will automatically be denied admission, or above which admission is guaranteed." Pet. App. 5a. The policy instead requires the admissions office to look beyond grades to other criteria important to the Law School’s educational objectives, such as "experiences . . . likely to be different from those of most students." JA 114. As Dean Jeffrey Lehman explained, an applicant’s potential "contribution to the diversity of the environment" is an important part of his or her qualifications. Tr. 5:195.

Second, the policy does not restrict the types of diversity contributions eligible for "substantial weight" in the admissions process. JA 120. The Law School seriously considers each "applicant’s promise of making a notable contribution to the class by way of a particular strength, attainment or characteristic—e.g., an unusual intellectual achievement, employment experience, nonacademic performance, or personal background." JA 84. *The Law School’s policy explains that effective pursuit of its educational mission has been greatly furthered by the presence of "meaningful numbers" or a "critical mass" of "students from groups which have been historically discriminated against, like African-Americans, Hispanics, and Native Americans" 5 because these students "are particularly likely to have experiences and perspectives of [-5-] special importance to our mission."* JA 120. But many other diversity-related factors are seriously considered, such as a record of "leadership, work experience, [and] unique talents or interests." Pet. App. 27a-28a; see Tr. 1:244-45.


* The Brief rightly distinguishes between viewpoint diversity as an overall goal, and racial or ethnic diversity as an important contributor. However, it is not clear why, of all the races and ethnicities in America, the "meaningful numbers" or "critical mass" necessary to provide racial and ethnic diversity must all be recruited from black, Hispanic and Native Americans (hereinafter BHN). True, these three groups "have been historically discriminated against," but so have many others: Poles, Irish, Italians, Russians, Arabs, Asian Indians, Chinese, Japanese, Filipinos, etc, etc. (Not to mention whites, during the last three or four decades, by affirmative action programs.)

When the Law School seeks racial/ethnic diversity, is it looking for all the differences in experience and perspective that may come from racial or ethnic difference, or primarily the experience and perspective that comes from having been discriminated against? If the former, many more than BHN would be available for recruiting the "critical mass" sought. If the latter, applicants could be preferred based simply on their having been discriminated against, regardless of their race. In either case, the reason alleged for racial or ethnic preference in admissions disappears.

Third, the Law School’s process ensures that candidates have an opportunity to share all relevant information about their background for consideration. The application requests a personal statement, letters of recommendation, and an essay describing the ways in which the applicant will contribute to the life and diversity of the Law School. JA 83-84. The background and experiences revealed by the application "commonly" make a difference in the admissions decision. Tr. 1:244-45. By applying this flexible policy, the Law School has generally succeeded in enrolling classes of superb students from diverse backgrounds that include enough minority students to provide meaningful integration of its classrooms and residence halls.

3. There is literally no chance that these results could be sustained under any race-blind admissions program. In 1997 when petitioner applied, there were only *67 minority applicants, compared to 1236 white and Asian American applicants,* in the LSAT range (164+) from which over 90% of the admitted white students were drawn. JA 172-79. Competition for these minority applicants is extremely fierce, and the Law School cannot hope to enroll more than a few of them. In 2000, there were only 26 African-American applicants nationwide with at least a 3.5 GPA and a 165 on the LSAT compared to 3173 whites and Asian Americans. 6 **Any race-blind methodology applied to the upper and middle grade and test score ranges will therefore invariably select a class with very few minority students.**


* By "minority applicants," the Brief means only blacks, Hispanics and Native Americans. By some administrative or lawyerly magic, Asian Americans are not "minority applicants."

** This is a pseudo problem, created by defining "minority students" to exclude a large number of applicants who, by racial/ethnic background or cross-cultural experience, could make a significant contribution to the School's racial and ethnic diversity. The Law School fails to distinguish between (1) the applicants who could add to the student body's racial/ethnic diversity, and (2) the applicants to whom it grants race-based admission. Although the first group may include all of the second, the second does not include most of the first.

[-6-] Although the district court did not designate the issue of race-neutral alternatives for trial (CAJA 99), it found "no reason to doubt" that the number of minority students would drop "sharply and dramatically" under a "race-blind admission system" that continues to give substantial weight to grades and LSAT scores. Pet. App. 229a. *An unrebutted expert study demonstrated that a class of 400 students selected that way would have included a total of 16 African-American, Hispanic and Native American students—down from 58 under the Law School’s policy. CAJA 6047. The educational benefits that depend upon opportunities for frequent interaction among students of different races cannot be achieved with so few minority enrollments. 7* The district court nevertheless faulted the Law School for its "failure to consider" and "perhaps experiment" with race-neutral programs that would sacrifice academic excellence and selectivity—such as the random selection of applications that satisfy minimum quantitative credentials through a "lottery." 8 Pet. App. 251a. The Sixth Circuit [-7-] rejected that reasoning. It held that "Bakke and the Supreme Court’s subsequent decisions [do not] require the Law School to choose between meaningful racial and ethnic diversity and academic selectivity." Id. at 35a. Petitioner suggests that the Sixth Circuit’s decision should be reversed because it applied de novo review to this and several other issues, but the courts below clearly disagreed only as to matters of law and legal characterization, not historical fact. 9


* Note 7, assuming an entering class under race-blind admissions of only 4% BHN, calculates what this would mean for the percentage of BHN in a typical course section or dormitory. Based on this calculation, the Brief asserts that "The educational benefits that depend upon opportunities for frequent interaction among students of different races cannot be achieved with so few minority enrollments."

First, the calculation ignores the presence of students of color from groups that do not receive racial preference. I don't know which year the referenced study was based on, but I have some relevant numbers for 1995 from the district-court trial of this case.

The Law School's favored groups that year were African, Mexican, Puerto-Rican and Native Americans. Their applicants constituted 14.7% of those admitted. This figure would fall to 4% under race-blind admissions, if we apply the discount found in the study cited by the Brief. But the proportion of students of color rises to 15%, when one adds the Asian-Americans and Other Hispanics admitted, all without racial preference. Also, an undetermined fraction of the 1% of Foreign admits, and of the 14% of Undeclared Ethnicity admits, would presumably be mixed race or nonwhite. This suggests that under color-blind admissions, 15-20% of those admitted in 1995 were students of color. The School wants a "critical mass," "meaningful numbers"? Voilá.

Second, these calculations understate the diversity potential of the 1995 admittees. The School seems to make the foolish (racist?) assumption that the only ethnic diversity of interest is nonwhite. No doubt many Americans of European, or West or South Asian ancestry have lost most of their ethnic distinctiveness, but many have not. 59% of the Law School's admittees were "Caucasian-Americans." Some (many?) of them had much to teach other students about what it has meant to be Irish-, German-, Polish-, Italian-, Scandinavian-, Russian-, Ukrainian-, Czech-, Bulgarian-, Romanian-, Armenian-, Turkish-, Arab-, Iranian-, Indian-, Pakistani-, Serbian-, Croat-, Greek-, Spanish-, or Portuguese-American. (With apologies to the many Caucasian ethnicities not named.)

Third, among the white admittees there must be some whose cross-cultural education or experience would make them valuable contributors to an understanding of and sympathy with racial or ethnic diversity. Consider, for example, students who have mastered a second language, or whose previous education has been largely abroad.

One premise is fundamental in the Law School's case: Without "race-conscious" admissions for black, Hispanic and Native Americans, it cannot admit a "critical mass" of racially or ethnically diverse students. The premise is based on a conceptual confusion: The School erroneously equates being a BHN student with being a racially or ethnically diverse student. When the error is detected, the premise collapses. And so does respondent's case.

4. The district court similarly made no factual finding that the Law School was administering a secret quota forbidden by this Court’s decision in Bakke. See Pet. Br. at 41-42; U.S. Br. at 9. To the contrary, it acknowledged that "the law school has not set aside a fixed number of seats for underrepresented minority students, as did the medical school in Bakke." Pet. App. 248a. The district court did reason that the admissions policy should be characterized as the functional equivalent of a quota because the "practical effect" of the policy has been to admit more than 10% minorities each year. Id. The Sixth Circuit rejected that legal conclusion, however, observing that the Harvard plan also pursued "meaningful numbers of minority students" but that did not make it a quota. Id. at 27a-28a.

Indeed, the record confirms that the faculty members who drafted the admissions policy in 1991 took precautions to ensure that the policy would not be read to authorize, require, or encourage admissions officers to admit a predetermined number of minority applicants. An early draft of the policy expressly stated that the Law School was likely to obtain the benefits of a critical mass when minority enrollment ranged between 11 and 17%. Pet. App. 225a. The chair of the Committee responsible for developing the policy explained that this range was derived from the educational experience of the faculty in prior years. CAJA 7564-65. Although one member of the Committee advocated that this numerical range should be retained in the final [-8-] policy because they were "just guidelines" and therefore "permissible under Bakke," the Committee rejected that suggestion in order to avoid "the risk that exists when you put numbers in, even as a guideline," that a future admissions officer might "see these numbers" and feel bound by them. CAJA 7736 (Lehman). 10

Nor is there any evidence that the Law School officials violated the intent of this policy by secretly directing the admission of a predetermined number of minority applicants. Dean Lehman and other Law School officials who administer the policy testified categorically that they did not employ any numerical quota in assembling the class.11 And the district court determined that they "acted reasonably and in good faith in adopting and administering the policy" in an "attempt[] to comply with Bakke" (Pet. App. 254a, 253a)—a finding that cannot be reconciled with any notion that they devised "disguised quotas." U.S. Br. at 9. Between 1993 and 2000, the number of minority students in each class varied from 42 to 73 (13.5-20.1%)—a range inconsistent with the operation of a fixed quota. JA 156-203; CAJA 1536, 4929-96, 5387-93, 5463-69; Pet. App. 30a. Dean Lehman also testified without contradiction that enrolling a critical mass of minority students is merely one "value in [the] composition of the student body that is important to us pedagogically" but "not the only value." CAJA 7767-68. That goal is balanced against competing objectives, such as assembling a class that shows [-9-] exceptional academic promise and is broadly diverse in attributes other than race. Id. at 7251-54, 7521-26.

5. Petitioner also asks this Court to find that the "plus factor" afforded to some minority applicants was just too large. But the district court did not find that the Law School could have admitted *meaningful numbers of minority applicants if it had assigned less weight to these applicants’ contributions to racial diversity; or that the acknowledged educational benefits could have been achieved with fewer minority students*; or that the plus factor was so large that the minority students were not well qualified for the rigors of the Law School’s demanding academic program.


* Note this illustration of the conceptual slippage pointed out in my previous comment. By using the word, minority, to refer only to BHN admittees, the Brief helps the reader forget the large number of other admittees who are racially or ethnically diverse.

Although the court found that the median undergraduate GPA of every underrepresented minority group "has been lower than the median GPA of Caucasians by approximately one-tenth to three-tenths of a point" between 1995 and 2000, and that the median LSAT has been approximately "seven to nine points" lower, 12 it never questioned (and petitioner stipulated) that all of the applicants admitted under the Law School’s policy were qualified. CAJA 8785. The Law School’s minority students have grades and scores which— while not always as exceptional as many white and Asian American admittees—nonetheless are superior to most applicants nationwide. They graduate, pass the bar exam, obtain judicial clerkships, and succeed in the practice of law at rates essentially indistinguishable from their white and Asian American classmates. Id. at 6222-23, 6243-58, 5870-81.

Petitioner nevertheless relies upon certain disparities in numerical credentials reflected in her admissions "grids" as proof that the program is not narrowly tailored to achieve its educational goals. Pet. Br. at 5-10. These grids were [-10-] generated by her statistician. It is undisputed that the Law School used nothing of the kind in its actual admissions process.13 They do, however, illustrate two key points. First, *the grids reveal that an applicant’s college GPA, LSAT score, and ethnic background all influence admissions decisions, but even together those factors fail to explain the outcomes*—either within or across racial categories. More than 40% of the admitted white and Asian American applicants from 1995 to 2000 came from "cells" in which at least 30% of the total white and Asian American applicants were rejected, demonstrating that subjective factors make the difference between acceptance and rejection for a great many of them as well.

 * The grids show not only that "an applicant's college GPA, LSAT score, and ethnic background all influence admissions decisions," but also how much. They accomplish this by showing applicants' rates of admission depending on their grades, scores and status as racially favored or not. Take, for example, the 1995 grids presented at trial by petitioner's statistician and uncontested by respondent. They show that the rate of admission for applicants with GPAs of 3.75+ and LSATs of 154-55 was 50% if they were racially favored (Black, Mexican, Puerto Rican or Native American), but 5% if they were not (White, Asian, Other Hispanic, Foreign, Ethnicity Unknown). With the same GPAs but LSATs of 161-163, the rate of admission was 100% for the racially favored, but 9% for the rest. For applicants at the 30th percentile (i.e., whose grades and scores were lower than 70% of all applicants), the rate of admission was 83% for the racially favored, but 1% for all others. For applicants at the 60th percentile (grades and scores higher than 60% of all applicants), the rate of admission was 93% for the racially favored and 8% for the rest. Complete tables showing the racial disparities in the 1995 admission rates are available on this website here.

And those factors can be given substantial weight. Even crediting the district court’s suggestion that diversity considerations might outweigh differences of up to a third of a letter grade or 7-9 points on the LSAT for minority students, the record shows that white and Asian American applicants frequently receive similar credit for other diversity factors. Holding GPA constant at 3.5-3.74, 53 white or Asian American students were accepted between 1995 and 2000 with an LSAT of 160 or below, whereas 88 with an LSAT of 167 or above were rejected. Holding LSAT constant at 164-166, 189 white or Asian American applicants with a 3.49 GPA or lower were accepted over 283 with a 3.75 or better. 14 Non-minority applicants are also frequently accepted with grades and test scores lower than minority applicants who are rejected. Sixty-nine minority applicants were rejected between 1995 and 2000 with at least a 3.5 GPA and a 159 or higher on the LSAT, while 85 white and Asian American applicants were accepted from the same or lower cells . 15 These observations do not suggest [-11-] that race does not matter in the admissions process. The grids demonstrate, however, that the Law School considers race only in the context of an individualized review seriously weighing many factors, including subjective non-racial diversity factors that make a real and dispositive difference for many white and Asian American applicants as well.

Second, the Law School’s individualized consideration of racial background does not actually affect the outcome of the overwhelming majority of the admissions decisions each year, or unfairly burden other applicants who may have higher test scores but who would not significantly enhance the diversity of the class. Plaintiff’s own expert testified that he was "sure" that grades and test scores had the "strongest association with admissions decisions" relative to any other factors, including race. Tr. 2:211-13. Approximately two-thirds of the Law School’s minority applicants are denied admission each year, and in each of the years between 1995 and 2000 the Law School denied admission to a greater proportion of minority applicants than majority applicants. CAJA 6045, 7585. Nor is there any dispute that the average odds of admission for non-minority applicants would have increased by less than 5% if the Law School had not taken race into account as part of its assessment of diversity contributions. Id. at 6045.

Accordingly, petitioner’s "probabilities" and "odds ratios" comparing white and minority applicants with identical credentials (Pet. Br. at 8-10) would reveal nothing unlawful even if the methodology were sound. See infra p. 44. It would be surprising indeed, in a regime in which race is given any weight, if minority applicants were not admitted at substantially higher rates than otherwise similar non-minority applicants. As the Sixth Circuit [-12-] explained, petitioners have "concede[d] that all admitted students are qualified," and evidence that race "plays an important role in some admissions decisions" is simply the "logical result of reliance on the Harvard Plan." Pet. App. 31a (emphasis added).

* * * * *



A. This Court Held In Bakke That Institutions Of Higher Education May Consider The Race Of Applicants As One Factor Among Many When Attempting To Assemble A Diverse, Racially Integrated Student Body

Although different Justices articulated a range of views about the permissibility of race-conscious admissions practices in Bakke, Justice Powell forged a middle ground that constituted (and has ever since been relied on as) the holding of the case. Joining Justice Powell, a majority of this Court agreed on several important propositions —all of which were essential to the result. *Five Justices reversed the California Supreme Court’s mandate prohibiting the University of California from considering race in admissions decisions,* 438 U.S. at 270 n.**, 271, **agreeing that the "competitive consideration of race and ethnic origin" in the context of a "properly devised admissions program" would be constitutional and consistent with Title VI,** id. at 320. ***Five Justices agreed that the University of California could constitutionally devise such a program even though it was "conceded that [the University] had no history of discrimination" and the University articulated no narrowly remedial justification for considering race.*** Id. at 296 n.36 (Powell, J.). **** And all five agreed that Harvard College’s admissions policy—which also articulated no remedial [-15-] purpose and was solely tailored toward, and justified by, Harvard’s desire to assemble a diverse student body—was "properly devised" and "constitutional."**** Id. at 320, 326 n.1 (Brennan, J., concurring in part).


* True.

** True.

*** True, though not supported by the citation.

**** True, more or less. The five Justices were Powell, Brennan, White, Marshall and Blackmun. The latter four did not concur with Section V-A of Powell's opinion (pp 315-319), in which he expounded a diversity rationale for the use of race in admissions. They did subscribe to a very short Section V-C (p 320), which permitted "the competitive consideration of race and ethnic origin" in "a properly devised admissions program," without defining either phrase. And in Note 1, p 326, they agreed with Powell that "a plan like the 'Harvard Plan'[citation omitted] is constitutional under our approach, at least so long as the use of race to achieve an integrated student body is necessitated by the lingering effects of past discrimination." The goal affirmed in this brief note ("an integrated student body") differs from that of the Harvard plan (overall diversity of the student body), and the note fails to specify in what respects a plan, to be constitutional, must be "like" the Harvard Plan. A consensus concerning the plan is present, but thin, flickering and ambiguous.

The minimum core or essential holding of Bakke, therefore, is that a University may consider race in admissions, even if it has no historical discrimination of its own to remedy, at least in the manner exemplified by the Harvard plan appended to Justice Powell’s opinion. These observations require no sophisticated analysis, and they alone are sufficient to support the Sixth Circuit’s judgment in this case—since as that court held the Law School’s policy is "virtually indistinguishable" from the Harvard plan. Petitioner nevertheless asserts (Pet. Br. at 27-28) that the holding in Part V-C of Bakke has no precedential force here because it says nothing about the permissible purposes of a "competitive" race-conscious plan. Although the broad language of Part V-C certainly leaves a great deal unspecified, that paragraph was not the only proposition to garner a majority. Five Justices also specifically agreed that the Harvard admissions policy was constitutional. Because the sole justification advanced in that plan was student body diversity, it *necessarily* follows that five Justices agreed that diversity was a sufficient justification. Petitioners resist that obvious conclusion based on the fact that Justices Brennan, White, Marshall and Blackmun "agree[d] with Mr. Justice Powell that a plan like the ‘Harvard’ plan … is constitutional under [their] approach, at least so long as the use of race to achieve an integrated student body is necessitated by the lingering effects of past discrimination." 438 U.S. at 326 n.1 (Brennan, J. concurring in part) (emphasis added). But the italicized language in Justice Brennan’s opinion means only that a policy like Harvard’s must cease considering race once the disparities in applicants’ numerical qualifications produced by our Nation’s discriminatory past have been eliminated, because a racially diverse class could then be assembled by other [-16-] means. The Law School agrees with that caveat, and Justice Powell plainly did too; sadly, that day has not yet arrived. 16

 * Not "necessarily," since the footnote by the four Justices that endorses something "like the Harvard Plan" does not say in what respects, and states a different purpose ("an integrated student body"), subject to the need for remedying "the lingering effects of past discrimination." This is the need which they held to justify the use of race. As petitioner's Brief argues in the section cited, "Justice Brennan’s opinion contains no affirmative indication of support for the proposition that diversity is a compelling interest." (p 27)

Justice Powell’s reasoning was also the "narrowest ground" articulated by any of the Justices supporting reversal, and is therefore a holding of this Court under Marks v. United States, 430 U.S. 188, 193 (1977). It is possible to tie oneself in metaphysical knots when applying Marks, by postulating creative and endlessly different theoretical axes along which one opinion or another might be considered the most "narrow." But Justice Powell’s opinion was "narrowest" in every sense that mattered: it completely defined, as a practical matter, the universe of race-conscious admissions programs that a majority of this Court regarded as constitutional.17 *It was immediately obvious to courts,18 commentators 19 and countless public [-17-] officials and institutions after Bakke that Justice Powell’s analysis and the Harvard plan were the coherent, if narrow, common ground for this Court’s judgment—and therefore the law of the land.20* As the U.S. Department of Education announced to the higher education community, "[t]he Court affirmed the legality of voluntary affirmative action" in order to "attain a diverse student body." 21

 * That Justice Powell's analysis did not receive majority support should be clear to any careful reader of the other Justices' opinions. It is true that "countless public officials and institutions" acted as if Powell's diversity rationale were part of the "Court's judgment," but this would not make it so.

B. Settled Principles Of Stare Decisis Strongly Counsel Against Overruling Bakke

Because Bakke has proven to be a landmark decision, *the principles outlined by Justices O’Connor, Kennedy and Souter in Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833 (1992), provide a useful framework for analyzing whether there is any "‘special justification’" 22 for reconsidering that decision.* There is not.

 * The criteria for reconsideration set forth in Casey applied to a decision whose holding and rationale were clear and supported by a majority. At issue was whether the case had been correctly decided. In Bakke the issues include not only whether the holding was correct, but what it was. The criteria for reconsidering such a case would presumably be different or differently weighed.

[-18-] First, intervening decisions of this Court have not left Bakke "a doctrinal anachronism discounted by society." Id. at 855. This Court has never questioned the core holding of Bakke, and indeed has uniformly assumed its continuing validity. 23 Bakke has become a "long-established precedent … integrated into the fabric of the law," Adarand, 515 U.S. at 233, and of our "national culture," Dickerson v. United States, 530 U.S. 428, 443 (2000).

Second, the standards established in Bakke have proven to be "[]workable," Casey, 505 U.S. at 855, as demonstrated by a 23-year history of enforcement under Department of Education regulations. Shortly after Bakke, the Department undertook a comprehensive reexamination of its regulations governing higher education admissions under Title VI. It concluded that universities could, "consistent with Bakke and the Department’s regulation, … [c]onsider race, color, or national origin as a positive factor, with other factors … in selecting from among qualified candidates," and that "[t]he relative weight granted to each factor is properly determined by institution officials; race, color or national origin may be accorded greater weight than other factors." 44 Fed. Reg. at 58,510. The Department has reaffirmed its regulations and guidance many times over the past two decades and five presidential administrations, and has used the standards established in Bakke as an effective, workable framework for the enforcement of Title VI in both [-19-] admissions and financial aid. 24 Although the United States remarkably fails to mention its regulations or policy interpretations in its brief, all of them remain in force and none would be constitutional if Bakke is overruled. 25

Third, Bakke could not be overruled "without serious inequity to those who have relied upon it." Casey, 505 U.S. at 855. Over the past twenty-five years, universities and professional schools, public and private, have made countless decisions about their faculty hiring, physical plant, capital spending and curriculum in reliance on this Court’s assurance that they would not be forced to a stark choice between racial diversity and radically lower academic standards and ambitions. The State legislators and private donors who fund these institutions have chosen to support them instead of countless other worthy causes because, in part, they are both racially integrated and committed to academic excellence. 26 If Bakke is overruled, the Law School will have to become a very different kind of institution than it, its philanthropic donors, and the State of Michigan have worked so hard to build.

Finally, *overruling Bakke would cause "significant damage to the stability of the society governed by it." Casey, 505 U.S. at 855. Overruling Bakke would force this Nation’s elite and selective institutions of higher education, public and private, to an immediate choice between dramatic [-20-] resegregation and abandoning academic selectivity. If they chose to maintain academic standards, the representation of African-American students at the 89 most selective law schools would fall from approximately 7% now to less than 1%. Three-quarters of the African-American students who are currently admitted to accredited law schools would not be accepted anywhere, and 40% of those still admitted would be admitted only to schools with predominantly minority student populations.* 27 Those predictions are confirmed by experience. In the year after the Fifth Circuit prohibited the University of Texas Law School from considering race in its admissions process, for example, Hispanic admissions fell by 33% and African-American admissions fell by 86%— to four students, out of a class of about 500. 28


* We may glimpse in this paragraph the real motivation for the Law School's policy of racial preference. We discovered above that the stated goal (obtaining a "critical mass" of students who are racially or ethnically diverse) was easily reached without racial preference, or any sacrifice of academic standards. But now an alarm is sounded that may explain much. The Law School fears that the abandonment of racial preference would cause a substantial decrease in black law school enrollment, which would result in a "significant damage to the stability of the society."

The argument based on diversity as a compelling interest is probably camouflage. The honest argument is more likely that social peace with blacks (and Hispanics?) requires discrimination in their favor until such time as the distribution of their abilities and achievements resembles that of whites.

As the United States recognizes (U.S. Br. at 16), if higher education is not "broadly inclusive to our diverse national community, then the top jobs, graduate schools and the professions will be closed to some." Yet a decision to overrule Bakke would cut the minority lawyers currently being trained by half or three-quarters, resulting in the near-complete absence of *minority students* from the schools that train most of our federal judges, prosecutors and law clerks (to say nothing of the new lawyers at our country’s leading law firms). 29 That is a chilling prospect. As our country becomes increasingly racially diverse, the public confidence in law enforcement and legal institutions so essential to the coherence and stability of our society will be [-21-] difficult to maintain if **the segments of the bench and bar currently filled by graduates of those institutions again become a preserve for white graduates**, trained in isolation from the communities they will serve.


* That is, black, Hispanic and Native American students.

** This language implies that under race-neutral, academically rigorous admission standards only white applicants would be smart enough to get into Law School, which is demonstrably untrue, and racially insulting.

C. Educational Experience, Social Science Research, And Common Sense Confirm That Diversity Has Compelling Educational Benefits

This Court recognized long before Bakke that preparing students for work and citizenship in our diverse society is exceedingly difficult in racially homogenous classrooms and on racially segregated campuses. In Sweatt v. Painter, 339 U.S. 629, 634 (1950), this Court held that Heman Sweatt could not receive an equal legal education at a law school which "excludes from its student body members of the racial groups which number 85% of the population of the State and include most of the lawyers, witnesses, jurors, judges and other officials with whom petitioner will inevitably be dealing." Id. "The law school, the proving ground for legal learning and practice, cannot be effective in isolation from the individuals and institutions with which the law interacts." Id. This Court has acknowledged the educational benefits of a diverse student body repeatedly since then. See Brown v. Bd. of Educ., 347 U.S. 483, 493-95 & n.11 (1954); Wash. v. Seattle Sch. Dist. No. 1, 458 U.S. 457, 472 (1982) ("[I]t should be equally clear that white as well as Negro children benefit from exposure to ‘ethnic and racial diversity in the classroom.’") (citation omitted). And those conclusions have been confirmed by congressional findings, educational experience, social science, and common sense.

1. Congress has repeatedly made specific findings that "elimination of racial isolation has significant educational benefits," even in the absence of any prior discrimination or remedial purpose, when authorizing federal financial assistance for local school districts seeking to eliminate both de jure and merely de facto segregation in their schools. 30 [-22-] The legislative history reveals Congress’s firm conclusion that "racially integrated education improves the quality of education for all children," H.R. Rep. No. 92-576, at 10 (1971), and that "[e]ducation in an integrated environment, in which children are exposed to diverse backgrounds, is beneficial to both" white and minority students, S. Rep. No. 92-61, at 7 (1971). The recently enacted No Child Left Behind Act of 2001 reaffirmed Congress’s findings that "[i]t is in the best interests of the United States … to continue the Federal Government’s support of local educational agencies that are … voluntarily seeking to foster meaningful interaction among students of different racial and ethnic backgrounds …." 31

2. These findings, which reflect the longstanding conviction of the United States government on a set of critically important issues of fact and national policy, were also supported by a powerful and essentially uncontested evidentiary record in this case. 32 The United States filed an amicus curiae brief in the district court summarizing the social science research, and concluding that diversity "in the higher education context improves students’ education, racial understanding, cultural awareness, cognitive development and leadership skills." CAJA 803. In its filing in this Court, the United States repeatedly emphasized its belief that educational diversity is an "important goal[]," U.S. Br. at 8, and that keeping undergraduate and graduate institutions open to "people of all races and ethnicities" is "a paramount government objective." Id. at 13.

That belief does not depend on "crude stereotypes." Pet. Br. at 38. It simply acknowledges the elephant in the room—that despite the recent advent of formal equality under the law and indisputable progress in race relations (in part because of the growing racial diversity in institutions [-23-] like the Law School), America remains both highly segregated by race and profoundly and constantly aware of its significance in our society. Many white Americans underestimate those realities because, of course, "[t]o be born white is to be free from confronting one’s race on a daily, personal, interaction-by-interaction basis." By contrast, "[t]o be born black is to know an unchangeable fact about oneself that matters every day." 33 The evidence for that fact, anecdotal and scientific, is beyond serious dispute. The House Judiciary Committee recently found that:

 millions of African-Americans and Hispanics alter their driving habits in ways that would never occur to most white Americans. Some completely avoid places like all-white suburbs, where they fear police harassment for looking "out of place." Some intentionally drive only bland cars or change the way they dress. Others who drive long distances even factor in extra time for the traffic stops that seem inevitable. 34

African-American men are asked to pay almost twice the markup that white men are asked to pay for automobiles. 35 Recent studies have shown dramatic disparities in the treatment of whites and African-Americans trying to rent an apartment over the telephone (most people can identify a caller’s race by dialect and the sound of their voice). 36 The issue is much more complex and subtle than just the unfortunate persistence of widespread racial discrimination. [-24-] Stereotypes and unthinking assumptions about social roles pervasively influence daily life. The Dean of the School of Education at Berkeley, who is Hispanic, testified in this case that when he was cutting the grass in front of his own house, a neighbor approached him to ask what he charges for yard work. CAJA 8472. And the renowned historian John Hope Franklin testified that "in recent years he has been approached more than once by a white person in a hotel lobby or private club who asked him to fetch her coat or car." Pet. App. 267a.

Minority students draw as wide a range of conclusions from experiences like these, and from the ideas they have been exposed to, as white students do from their own lives and influences. The Law School’s minority students are, like its white students, liberals and conservatives, communitarians and libertarians, devotees of both Mill and Kant. But the presence of persons who have had such experiences enriches the educational environment, if only because *it is human nature to undervalue or fail to see burdens that we haven’t truly experienced ourselves.* 37

 * It is surely an educational benefit to get to know people who have been racially discriminated against (unless of course it persuades one to discriminate in their behalf). But why are nonwhites assumed to be the only persons who can provide this benefit? After 35 years of race-based affirmative action at the expense of whites, it should be clear that people of color are not the only victims of racial discrimination in America. Perhaps "it is human nature" for Michigan's administrators "to undervalue or fail to see burdens that we haven't truly experienced ourselves."

3. The importance of these differences in lived experience is particularly trenchant in the context of legal education. The legal system is the epicenter of our Nation’s ongoing struggle to overcome racial divisions that persist in our society. Indeed, monitoring and mediating the progress of that struggle has become one of the most important jobs of the federal courts. Discrimination suits under Titles VI, [-25-] VII and IX, ongoing school desegregation cases, Voting Rights Act enforcement and racial-profiling lawsuits have all become staples of the case load. What legal consequences if any should follow from the disparate impact of the criminal justice system in general, and certain criminal statutes in particular, on racial minorities is one of the most oft-debated and important challenges that our society faces. Against this backdrop, law schools need the autonomy and discretion to decide that *teaching about the role of race in our society and legal system,* and preparing their students to function effectively as leaders after graduation, are critically important aspects of their institutional missions. And it hardly requires extensive proof that pursuit of those goals is greatly enhanced by the presence of meaningful racial diversity among the law school’s student body—enhanced in ways that white students alone, no matter what their viewpoints are or even what their experiences have been, cannot possibly supply.

 * What the law schools teach "about the role of race in our society and legal system" is indeed immensely important. The greatest evil in the admissions policies defended by respondent is their racial lesson: that racial discrimination, while wrong in behalf of some racial and ethnic groups in our society, is right in behalf of others. This undermines the authority of a moral and legal principle fundamental to the well-being of a multi-racial, multi-ethnic society: no unequal treatment by government, no second-class citizenship, based on race or ethnicity.

The presence of minority students is also essential to the Law School’s educational mission in other ways. At its most successful, the educational process is a productive collision not only of facts and ideas, but also of people. The Law School is training lawyers and leaders for a society in which, within the careers of its current students, white citizens will become a minority of the population. Those students need to learn how to bridge racial divides, work sensitively and effectively with people of different races, and simply overcome the initial discomfort of interacting with people visibly different from themselves that is a hallmark of human nature. 38 As then-Provost Condoleezza Rice recently explained, "differences in talent, in background, in racial and ethnic identity, in creed" in an educational environment can open "a small window on perhaps the greatest challenge [-26-] before us as human beings—finding a way that people who are different can live together in peace and move forward together." Stanford Class Day Speech (June 12, 1999). 39

4. Effective pursuit of these goals requires more than an isolated handful of minority students, for several reasons. First, the educational benefits of diversity depend on opportunities for interaction—in classrooms, cafeterias, or residential settings. The Law School is a large institution, and a few minority students obviously could not be everywhere at once, or establish meaningful personal ties with more than a small fraction of their classmates.

Second, the presence of more than one or two minority students in a classroom encourages students to think critically and reexamine stereotypes. Kent Syverud, Dean of the Vanderbilt Law School, testified that the classroom "dynamic is different within the class among the students and between me and the students, when the class is homogenous" or has a "token minority student" than "when there are enough minority students … that there is a diversity of views and experiences among the minority students." CAJA 7698; id. at 5618-20. When there are more than a token number of minority students, "everybody in the class starts looking at people as individuals in their views and experiences, instead of as races." Id. at 7699.

Third, as the Harvard plan recognized, there is a powerful body of evidence that very low numbers of minority students tend to create a "sense of isolation among the [minority] students themselves" that would "make it more difficult for them to develop and achieve their potential." 438 U.S. at 323. That sense of isolation particularly inhibits the willingness of many minority students to participate freely in class discussions. 40 [-27-]

 The above paragraphs continue the fiction that the only racially and ethnically diverse admittees under race-blind admissions would be a tiny remnant of BHN.

D. The Law School’s Educational Objectives Are Sufficiently Compelling To Satisfy Strict Scrutiny

Petitioners’ arguments boil down to the assertion that only one interest can be characterized as "compelling": remedying an institution’s own past discrimination. But this Court has steadfastly refused to embrace a rigid interpretation of the Equal Protection Clause that would preclude "case-by-case" scrutiny of the justifications advanced for the consideration of race in this or any future case. Croson, 488 U.S. at 519 (Kennedy, J., concurring in part); see also Wygant, 476 U.S. at 286 (O’Connor, J., concurring in part) (recognizing "the possibility that the Court will find other governmental interests … to be sufficiently ‘important’ or ‘compelling’ to sustain the use of affirmative action policies"). As Adarand recently confirmed, not all decisions influenced by race are "‘equally objectionable’" and strict scrutiny is designed to provide a framework for "carefully examining" the importance and the sincerity of the reasons advanced. 515 U.S. at 228 (citation omitted). Those reasons—while certainly few in number— are potentially as varied and difficult to predict as the challenges facing our Nation.

By way of example, few would question the State’s need to take race into account when choosing an undercover law enforcement officer to infiltrate a racially homogenous terrorist cell, or when acting to quell a race riot in a prison. See Croson, 488 U.S. at 521 (Scalia, J., concurring). Indeed, this Court has recognized a variety of governmental interests—from combating corruption to promoting health and safety—as sufficiently "compelling" to justify incursions upon other rights to which strict scrutiny applies and which [-28]- are, in their own way, no less weighty than those granted by the Equal Protection Clause. 41


I agree that the rule against governmental racial discrimination, though fundamental, is not absolute. But its moral and legal authority can be maintained, only when the exceptions to it are based on genuine necessity, and narrowly defined.

The Law School’s interest in achieving the educational benefits of diversity plainly satisfies the standards set by this Court. First, race is relevant to a core mission of the Law School that is vitally important and plainly "legitimate." Adarand, 515 U.S. at 227-28. Second, the program does not use racial classifications as a proxy for other more germane considerations, or in a way that suggests reliance on impermissible stereotypes that demean any racial or ethnic group. Id. at 226. Third, the asserted interest in considering race to achieve the benefits of student body diversity has a "logical stopping point," Wygant, 476 U.S. at 275, sufficient to ensure that it will not justify indefinite or unconstrained consideration of race.


The arguments in this section fail at three levels. First, campus diversity of viewpoint as a compelling interest: After Powell offered viewpoint diversity as a justification for racial preferences in admissions, it gained widespread popularity as an academic goal. But the appeal seems to lie much less in diversity as an end than in preference as a means. Schools that extol viewpoint diversity when justifying the use of race seem indifferent to the comparative lack of politically and religiously conservative views on their campuses. Such hypocrisy is indefensible. However, the remedy is not to add preference for political and religious minorities, but to drop it for racial and ethnic minorities.

If viewpoint diversity in academia were seriously established as a compelling public interest, schools would ask applicants for their religious, political, moral, philosophical, cultural and other opinions. Admission decisions would be based partly on the answers to these questions. School administrators and other government officials would establish goals for attaining "meaningful numbers" or a "critical mass" of students with particular views. Complaints to government offices or courts would allege that schools were failing to promote diversity, or were discriminating for or against applicants based on their opinions. There would be appeals to the First Amendment protections of freedom of thought, but constitutional rights are not absolute, and may be superseded by compelling interests. Establishing viewpoint diversity on campus as a compelling public interest is a road we should not take. If the Law School believes otherwise, it should make the case. Falling back on one man's opinion in a nine-Justice Court does not make the case.

Second, campus racial and ethnic diversity as a compelling interest: The Brief argues for this diversity as a necessary component of viewpoint diversity. If the latter fails as a compelling public interest, does the former remain? The Brief cites some uncontested educational benefits. However, the existence of benefits does not a public necessity make. We need to be shown that these benefits are essential, despite the evils of racial discrimination candidly faced. The Brief is so far from such candor, that it carefully avoids even the words, racial discrimination and racial preference, in designating the admissions policy it defends.

These obstacles belong to the nature of the issue. Before reaching them the Brief had demolished its own position by its definition of the students that provide racial and ethnic diversity. One might have expected that at least these would include all students of color, and perhaps those whites who retained some of their ethnic distinctiveness. The Brief claimed that a "critical mass" was necessary. The trial data on 1995 admittees showed that without racial preference or any sacrifice of academic standards, at least 15% would be students of color. That sounds like "meaningful numbers," a "critical mass." But the Brief, ignoring the Asian and Other Hispanic admittees, insisted that under a race-blind process, the students providing racial or ethnic diversity would be only the tiny proportion (4%) of black, Mexican and Native Americans. Having fashioned an absurd definition of "racial and ethnic diversity," the Brief rendered its rational defense as a compelling interest impossible.

Third, the admission, as a compelling interest, of substantially more black, Hispanic and Native Americans into selective universities and professions than would be possible under the standards that other applicants have to meet: I suspect that this is much closer to the Law School's actual purpose than viewpoint or racial diversity. And the School may well believe that the successful pursuit of this goal is a national necessity. But in the Brief thus far, such views have only been hinted; no argument has been launched to persuade the Court to adopt them.

1. As Justice Powell recognized in Bakke, the Law School’s interest in the educational benefits of a diverse, racially integrated student body is both unquestionably legitimate and "of paramount importance in the fulfillment of its mission." 438 U.S. at 313. The cultivation of a diverse and vibrant academic environment is the most important "‘business of a university,’" and the selection of students who will best enrich that environment is one of its "‘four essential freedoms.’" Id. at 312 (quoting Sweezy v. N.H., 354 U.S. 234, 263 (1957)). Indeed, "[t]he ‘nation’s future depends upon leaders trained through wide exposure’ to the ideas and mores of students as diverse as this Nation of many peoples." Id. at 312-13 (Powell, J.) (quoting Keyishian v. Bd. of Regents, 385 U.S. 589, 603 (1967)). The Law School’s desire for a diverse student body is at the very core of its proper institutional mission. 42

[-29-] Indeed, racial diversity is simply far more relevant to the core mission of a university or professional school than to virtually any other government endeavor. See Adarand, 515 U.S. at 227 (race is "‘in most circumstances irrelevant’" to governmental action and "‘therefore prohibited’") (quoting Hirabayashi v. United States, 320 U.S. 81, 100 (1943)). 43 This Court has recognized that universities have an unparalleled need for pluralism that is essential to the vitality of our society. 44 Although the City of Richmond could install the finest possible plumbing fixtures in its jail using an all-white work force, Croson, 488 U.S. at 481-82, this Court recognized in Sweatt and Bakke that the Law School cannot provide the finest possible legal education with a nearly all-white student body.

[-30-] 2. The educational interest in a diverse student body does not employ historic stereotypes, "directly equate race with belief and behavior," or use race as a poor proxy for characteristics that could be pursued directly. Metro Broad., 497 U.S. at 618 (O’Connor, J., dissenting). The Law School does not premise its need for a racially integrated student body on any belief that minority students always (or even consistently) express some characteristic minority viewpoint on any issue. To the contrary, breaking down such stereotypes is a crucial part of its mission, and one that cannot be advanced with only token numbers of minority students. Supra p. 26. *The Law School values the presence of minority students because they will have direct, personal experiences that white students cannot—experiences which are relevant to the Law School’s mission.* To the extent there are any proxies at work in the Law School’s policy, the "nexus [is] nearly complete," if not perfect. 497 U.S. at 626 (O’Connor, J., dissenting). 45

 * Notice how the campus reality is reduced and distorted by the Brief's conceptual framework. From this sentence, knowing the Brief's definition of "minority students," one would think that the student body consisted only of blacks, Hispanics, Native Americans, and whites. Without the first three groups, the School would be all white. Absent in this picture are all the Asian-Americans, and whatever students of color or of mixed race are in the Law School's categories of Unknown Ethnicity and Foreign. Also absent is the considerable ethnic variety among white students.

The United States reads this Court’s cases to hold that any recognition that members of racial minorities have relevant "life experiences" rests on an impermissible "stereotype." U.S. Br. at 20, 25 n.8. That is plainly incorrect. This Court has condemned the fiction that race determines a person’s "belief and behavior"46 —not the inescapable reality that race affects life experiences in our society. See J.E.B. v. Ala. ex rel. T.B., 511 U.S. 127, 148-49 [-31-] (1994) (O’Connor, J., concurring) ("[L]ike race, gender matters" in one’ s "resulting life experience[s]"). 47

3. As the dissenters applying strict scrutiny in Metro Broadcasting explained, "[a]n interest capable of justifying race-conscious measures must be sufficiently specific and verifiable, such that it supports only limited and carefully defined uses of racial classifications." 497 U.S. at 613. The interest in remedying societal discrimination that Justice Powell rejected as "amorphous" in Bakke itself, 438 U.S. at 307, and in Wygant, 476 U.S. at 276, plainly failed that test. 48 Because no individual employer or educational institution could hope to actually remedy societal discrimination, the enormity of that challenge would justify consideration of race without any "logical stopping point."49

As Justice Powell recognized, the Law School’s interest in achieving the educational benefits of racial diversity in its classrooms is entirely different. Precisely because those benefits are educational, any program that genuinely seeks to obtain them is constrained by its own logic and by other [-32-] pressing educational goals. It would be inconsistent with a sincere pursuit of those benefits, for example, to admit minority students who are unprepared to "be the intellectual peers of their fellows in the classroom," and whose presence would detract from, rather than enhance, the learning environment. CAJA 7756 (Lehman). Those benefits can justify the Harvard plan’s modest and flexible "attention to numbers," but not racial balancing of any kind—which Justice Powell famously condemned as "discrimination for its own sake." 438 U.S. at 307. 50 And, as Justice Powell explained in depth in Bakke, it would be inconsistent with a genuine interest in the educational benefits of racial diversity not to constantly weigh that interest against other academic goals—including the educational benefits of other kinds of diversity. 438 U.S. at 315-16. Taken seriously, the educational benefits of racial diversity justify only an individualized admissions system along the lines of the Harvard plan.

Finally, petitioner argues that an interest in educational diversity cannot be recognized as compelling because it would "give the Nation its first permanent justification for racial preferences." Pet. Br. at 33. The argument rests on an unspoken premise that should not be countenanced. The Law School of course recognizes that race-conscious programs must have reasonable durational limits, and the Sixth Circuit properly found such a limit in the Law School’s resolve to cease considering race when genuine race-neutral alternatives become available. Pet. App. 38a; JA 121; CAJA [-33-] 7750-51. The disparities in academic preparation that make such alternatives impossible today are rooted in centuries of racial discrimination. The district court found that these disparities will eventually be eliminated as our society "invest[s] greater educational resources in currently underperforming primary and secondary school systems." Pet. App. 291a. Any assumption that they are inevitably "permanent" merely because three decades of modest effort have not yet erased them should not be dignified with a place in our constitutional jurisprudence.


A. There Are No Race-Neutral Alternatives Capable Of Producing
A Diverse Student Body Without Abandoning Academic Selectivity

Petitioner and the United States assert that there are race-neutral alternatives available to the Law School. 51 Many of the ideas they present are not genuinely race-neutral, and all are demonstrably unworkable or would substitute a different institutional mission for the one that the Law School has chosen. 52 The Law School has studied [-34-] this issue for many years, and would like nothing better than to find a race-neutral admissions formula that would produce meaningful diversity without doing unacceptable damage to its other educational goals. Steady improvement in the quantitative credentials of the minority applicant pool will make such alternatives possible. At this point, however, every race-blind alternative requires a dramatic sacrifice of diversity, the academic quality of all admitted students, or both. 53

Recruiting and outreach. The Law School already engages in significant recruiting and outreach activities targeted at minority applicants, but such efforts have never proven sufficient to enroll a critical mass of minority students without the consideration of race in admissions. CAJA 401, 7668-70. *Given the small size of the pool of highly qualified potential applicants nationwide, and the recruiting efforts already directed at them by the Law School and its peers, such efforts have largely become a zero-sum competition. They are also not "race-neutral."

"Percentage Plans." The United States touts admissions policies adopted recently by the public undergraduate institutions in Texas, Florida and California, which guarantee admission to all students above a certain class-rank threshold in every high school in the State. There are serious and well-documented problems with that approach even for undergraduate schools.54 But the United States does not even attempt to articulate how such a program could work for graduate and professional schools. [-35-]

No elite law school could responsibly assemble a class by guaranteeing admission to every applicant who had secured a high grade point average in college, without regard to the institution or course of study. Moreover, such an approach could not produce meaningful diversity. The Law School draws from a national pool and is too small to guarantee admission to even a tiny percentage of graduates from every university in the country. At the universities from which it currently draws the vast majority of its students, minorities make up no more than around 3% of the students graduating in the top five or ten percent by GPA.55 The only way to produce a diverse, racially integrated class at the law school level through a "percentage plan" would be to limit and gerrymander the undergraduate institutions allowed to participate, such that an artificial proportion of them were highly segregated majority-minority schools. That is not race-neutral. If affirmative action for minority students is unconstitutional, then affirmative action for minority colleges would be a thin and cynical proxy that would be vulnerable under cases like Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252 (1977), and Shaw v. Reno, 509 U.S. 630 (1993).

Indeed, that vulnerability points up a deeper problem with the percentage plans at any level of higher education. The Law School’s current admissions policy considers race only as one factor among many, in an effort to assemble a student body that is diverse in ways much broader and richer than race. Because a percentage plan makes that kind of nuanced judgment impossible, it effectively sacrifices all other educational values—including every other kind of diversity. By subordinating traditional admissions criteria to a single-minded focus on race, these plans make race the "predominant factor" in the design of the entire admissions system. E.g., Hunt v. Cromartie, 526 U.S. 541, 547 (1999).

Abandon academic selectivity. The United States repeatedly suggests that the Law School "eas[e] admissions [-36-] requirements for all students," and "discard facially neutral criteria that, in practice, tend to skew admissions in a manner that detracts from educational diversity." U.S. Br. at 13-14. Those are in fact the same recommendation, since the only facially neutral criteria that the Law School considers that have a significant disparate impact on minority candidates are academic in nature.

As the grids and the chart at JA 219 demonstrate, the difficulty with such proposals is the composition of the applicant pool. There are so many more white and Asian American applicants throughout the upper and middle score ranges that no incremental lowering of standards will create a pool with meaningful racial diversity. Setting the bar so low that academic criteria are nearly irrelevant might allow a lottery (or academic-blind subjective review) to produce a racially diverse class, but any such plan would require the Law School to become a very different institution, and to sacrifice a core part of its educational mission.

Socio-economic criteria. The Law School already considers the light that a history of overcoming poverty or disadvantage may shed on every applicant’s likely contributions. But if petitioner is suggesting that the Law School could enroll a critical mass of minority students by giving even greater weight to socio-economic criteria in an honestly race-blind manner, the problem is, again, the facts. There is a strong correlation between race and poverty in our country. Nonetheless, there are still many more poor white students than poor minority students in the pool from which the Law School draws. "[T]here are almost six times as many white students as black students who both come from [low socio-economic status] families and have test scores that are above the threshold for gaining admission to an academically selective college or university."56 Again, this is not a way the Law School could enroll an academically talented class that is diverse in many ways, including race. Boalt Hall recently experimented with admitting more low- [-37-] income students but abandoned that experiment after one year, concluding that it could not produce racial diversity.57

"Experiential diversity." Finally, petitioner and the United States suggest that the Law School focus its admissions process on identifying those students, without regard to race, who have had the particular experiences and perspectives that the Law School regards as uniquely salient to its academic mission. That suggestion simply elides the central question, which is whether the Law School would still be permitted to consider "the experience of being an African-American, Hispanic or Native American in a society where race matters." Pet. App. 35a. If not, this proposal could not produce meaningful racial diversity, supra pp. 5-6, yet it would deny minority students the opportunity to have their own backgrounds and experiences "weighed fairly" in the admissions process. 58 If so, it is not clear how the proposal would differ from what the Law School currently does. As Dean Lehman testified, "the extent to which we take race and ethnicity into account is actually going to vary by individual. And it’s going to depend on the admissions file, and what they say in their essays about who they are, and the extent to which race is part [of] their experiences." CAJA 7755.

In its efforts to assemble a broadly diverse class, the Law School already looks for minority applicants who say interesting things about the ways that race has, or has not, influenced their lives. It would not, however, endorse an [-38-] admissions system that could consider the unique contributions that minority applicants can make to the educational environment only if they describe their experiences as "‘victims’ of discrimination," Pet. Br. at 37. As Gerhard Casper recently put it when explaining his support for race-conscious admissions programs at Stanford and other selective universities: "[i]n order to survive as a sane society, we should not create incentives for ever more people to think in terms of victimhood or to play the role of victims, or to suggest that one must be disadvantaged to be given serious consideration in the college admissions process." Casper, Statement on Affirmative Action at Stanford University (Oct. 4, 1995).59

* * * * *

C. The Law School’s Consideration Of Race Is Individualized, Competitive, Modest In Scope, And Does Not Impose An Undue Burden On Non-Minority Applicants

Petitioner’s brief points to various statistical measures of academic qualifications and odds of admission, and concludes that the Law School employs a "plus factor" that is too large. That is not truly a narrow tailoring argument at all. Narrow tailoring scrutiny of the size of a "plus factor" [-44-] must be focused on the questions to which that issue is genuinely relevant: the closeness of the "fit" between means and ends, and the burden imposed on innocent parties. The Law School’s policy satisfies both standards.

1. It is important to recognize at the outset that the statistical measures relied upon by petitioner cannot bear the weight that she places on them. Differences in average or median scores are unrevealing for reasons already explained. Supra p. 9 n.12. *Petitioner’s probabilities and "odds ratios"71 within individual cells on her admissions grids (or at a given index score) certainly establish some attention to race, but are inherently incapable of measuring its weight.* Because applicants within each cell have (by definition) identical quantitative qualifications, even a very modest "tie-breaking" plus factor would often produce enormous differences in probabilities or relative odds ratios.72 In addition, the composite relative-odds ratios (Pet. Br. at 9) are highly misleading because this methodology required petitioner’s statistician to exclude all of the cells (a majority of the total) in which white and minority applicants were treated the same. 73 A methodology that would quantify even a tie-breaking plus factor as an "enormous" one and exclude all data that reflects equal treatment is simply not useful to the constitutional inquiry. 74


* By "probabilities" on petitioners "admissions grids," the Brief refers to the trial testimony of Grutter's statistician, showing racially disparate rates of admission in 1995. These figures (previously noted in my comments at p __ supra) were not contested by the Law School. Several examples:

1. In the cell of applicants with UGPAs of 2.50-74, and LSATs of 148-50, the rate of admission for racially favored applicants was 8%. Other applicants did not reach this rate until the cell with UGPAs of 3.50-74 and LSATs of 161-63.
2. In the cell with 2.50-74 and 154-55, the rate for racially favored applicants was 25%. Other applicants reached this rate with 350-74 and 164-66.
3. In the cell with 3.25-49 and 154-55, the rate for racially favored applicants was 58%, a rate reached by other applicants with 3.50-74 and 167-69.
4. In the cell with 2.75-99 and 164-66, the rate for racially favored applicants was 100%. The rate for other applicants never reached 100%. Their highest was 95% (143/151) in the top cell, with 3.75+ and 175+.

In the first example the racial advantage equalled a full grade point and 13 LSAT points. In the second, a full grade point and 10-11 LSAT points. In the third, ¼ of a grade point and 13-14 LSAT points. In the fourth, a full grade point and 11+ LSAT points. The Brief to the contrary notwithstanding, such figures are a strikingly good measure of the size of the racial thumb on the Law School's admissions scale. (Tables showing the Law School's racially disparate rates of admission in 1995 are on this website here.)

Grutter's use of odds ratios as a measure, however, is problematic. Her statistician calculated enormous odds of black relative to white admissions, ranging from 53-443 to 1, for the years 1995-2000. It is hard to believe that the degree of preference varied that much, and hard to understand what the odds ratios actually measure. If the probability of admission is 20% for one applicant and 80% for another, it is easy to understand that the second probability is four times as large as the first. But what does it mean to say that the odds for white admission in general are, say, 3:7, and the odds for black admission are 53 or 443 times that large? The answers (on this website here) of petitioner's statistician to these questions at trial were unsatisfactory.

2. The bulk of petitioner’s narrow tailoring argument proceeds as if a "plus factor" is automatically [-45-] unconstitutional if it appears to have any significant impact upon which students are admitted. To the contrary, a race-conscious policy that did not meaningfully alter the outcomes of the admissions process could not, for that very reason, possibly be narrowly tailored. Such a policy would incur most of the costs associated with governmental consideration of race, while achieving nothing at all.

The most important "fit" question in this case, therefore, is whether the scale of the Law School’s plus factor is appropriately tailored to the achievement of its educational goals. See Croson, 488 U.S. at 507. On that question, the record supplies a clear and undisputed answer. A ruling that the Law School must place measurably less weight on race will preclude it from enrolling a meaningful number of minority students. However one measures the scale of the Law School’s plus factor, it is clearly the minimum required to make the policy, in Justice Powell’s words, an "effective means" "to the attainment of considerable ethnic diversity in the student body," 438 U.S. at 315, in light of the current applicant pool. 75

Justice Powell also recognized in Bakke that an admissions program does not genuinely "fit" the interest in educational diversity unless it considers race only in the context of a genuine commitment to diversity in a "broad[] array of qualifications and characteristics of which racial or ethnic origin is but a single though important element." 438 U.S. at 315. The program must therefore proceed on an "individualized, case-by-case basis," id. at 319 n.53, cannot isolate any applicants from competition with all others, and must be "flexible enough to consider all pertinent elements of diversity in light of the particular qualifications of each applicant, and to place them on the same footing for consideration, although not necessarily according them the same weight." Id. at 317 (emphasis added).

[-46-] The Law School similarly engages in a highly individualized, holistic review of each file, and gives serious consideration to all of the ways that applicants might contribute to a diverse educational environment. The Law School does not, of course, accord all such potential contributions the same weight, but it does weigh them "fairly" and "place them on the same footing for consideration." Id. at 317-18. *The Law School’s "plus," however measured, is far smaller than the disparities at UC-Davis in Bakke (0.61 to 0.94 points of GPA and 35 to 54 percentiles on the MCAT), id. at 277 n.7,* and the subjective diversity contributions of white and Asian American students are frequently given similar weight. Supra p. 10.


* The "disparities at UC-Davis," reported in the Bakke footnote, were between the average grades and scores of regular admittees and racially favored admittees. The Brief does not provide comparable figures for the Law School. Petitioner's statistician tabulated separately, but did not aggregate, the median admittee grades and scores for the racial groups. In 1997, the year Grutter applied, the group median UGPAs of racially favored admittees ranged from 3.35 to 3.54; the group medians for other admittees, from 3.48 to 3.71. The group median LSAT scores of the racially favored admittees were 160-164; the group medians for other admittees were 168-169.

Group distributions of UGPAs and LSATs are useful indicators of whether members of one group will view members of another as intellectual peers, but they do not prove the degree or even the fact of preference. When admissions are raceblind, racial group differences in the distribution of these factors is common and often substantial. A better way to estimate the degree of preference, whether based on race, gender, residence or something else, is to determine how much larger an unpreferred applicant's UGPA and/or LSAT would have to be to give him the same probability of admission, other things being equal, as the preferred applicant.

It would also indicate a poor fit between the scale of the Law School’s plus factor and its educational goals if the minority students being admitted in fact detracted from rather than enhanced its educational environment—or did not achieve the kinds of success, and provide the kinds of leadership, that the Law School expects from its students after graduation. The record in this case conclusively dispels such notions. Supra p. 9.

The idea that minority students themselves are somehow injured by being admitted to highly selective institutions also wilts under scrutiny. Such students graduate at significantly higher rates and earn much more in later life than their peers with identical grades and test scores who attend less selective schools. Shape of the River 54-68, 128, 264. *Graduates of all races from selective institutions support continued use of race in admissions to achieve diversity by wide margins*(much wider than the population as a whole), indicating that the consequences of such programs were enlightening—not stigmatizing. 76 In one recent study, 91% of the Law School’s graduates reported that racial diversity was a positive aspect of their [-47-] experience. 77 The Law School’s consideration of race is, in intent and effect, no more stigmatizing than the "plus" it gives to some white students to ensure geographic diversity, or to build a community across generations by admitting children of alumni.

 * To support this claim, Note 76 cites Shape of the River. However, its data contain no information on how "graduates of all races from selective institutions" feel about the "use of race in admissions." The data include strong support for racial diversity as an end, but nothing concerning the means of attaining it. A survey was conducted of white and black graduates of selective schools in 1976 and 1989. They were asked to "indicate how much emphasis your undergraduate school currently places and should place on . . . g. a racially/ethnically diverse student body." The survey could have inquired what means these graduates favored for attaining racial diversity, and specifically whether they favored racial preference in admissions and if so, how much. But no such question was asked. Indeed, since its academic practitioners have commonly refused to acknowledge the existence of racial preference in admissions, we cannot tell what proportion of these graduates were even aware of it.

3. The Law School’s program also does not "unduly burden individuals who are not members of the favored racial and ethnic groups." Metro Broad., 497 U.S. at 630 (O’Connor, J., dissenting). As Justice Powell recognized in Wygant, the burden imposed by race-conscious "school admission[s]" decisions, like the burden imposed by hiring goals, "is diffused to a considerable extent among society generally." 476 U.S. at 282, 283 n.11. Unlike a job layoff, in which "the entire burden of achieving racial equality" is imposed on identifiable individuals, "resulting in serious disruption of their lives," an admissions decision "often foreclos[es] only one of several opportunities." Id. at 283.

The Law School of course understands that these decisions are enormously important to all of its applicants, and that failure to gain admission can be very disappointing. But the Law School’s consideration of race imposes a burden on non-minority applicants so small and "diffuse" that it barely affects their chances at all. By way of example, an entirely race-blind process would have reallocated an average of 41 seats in each incoming class between 1995 and 2000, CAJA 6047, among the approximately 2200 applicants rejected each year. The Law School’s policy thus offers white and Asian American students a slightly smaller chance of attending a school that is thereby able to offer them (and others) a substantially better educational experience if admitted—hardly an unreasonable burden. 78 [-48] Moreover, because the Law School’s admissions policy is typical of those used at law schools nationwide, the handful of rejected students who would have been admitted under a rigidly race-blind policy may be expected to have gained admission to a comparable law school that is itself able to offer the benefits of a racially integrated environment.

The burden imposed on non-minority applicants by the Law School’s policy is wholly different in nature from that created by the FCC programs in Metro Broadcasting. The FCC’s distress sale program "created a specialized market reserved exclusively for minority controlled applicants"; literally, a "100% set-aside." 497 U.S. at 630 (O’Connor, J., dissenting). And because "[t]he basic nonrace criteria [were] not difficult to meet" in the comparative program, race was "clearly the dispositive factor in a substantial percentage of comparative proceedings"—perhaps "overwhelmingly the dispositive factor." Id. at 630-31.

By contrast, the record demonstrates beyond question that academics, not race, is the dispositive factor in the vast majority of the Law School’s admissions decisions. CAJA 7476, 7585, 7637; Tr. 2:210-13. The Law School’s academic criteria are overwhelmingly difficult to meet—so difficult that only a small fraction of our Nation’s college graduates can meet its standards. From among that group, the Law School considers each applicant as an individual and strives to admit a student body that will best further its educational goals. The Law School (while appropriately conscious of the racial and ethnic background of most applicants) has not, in other words, subordinated traditional criteria in a way that would make race the "predominant factor" in the admissions process. Hunt, 526 U.S. at 547.

4. Close scrutiny of the fit between means and ends and of the burden on non-minority applicants imposes [-49-] meaningful constraints on the consideration of race within the framework established by Bakke. Petitioner’s abstract contention that the Law School’s "plus" is simply too large offers no workable alternative. A holding along those lines would, as a practical matter, likely preclude any selective institution from employing any plus program to enroll meaningful numbers of minority students. The difficulty of measuring the precise weight given to race versus other diversity factors, coupled with the difficulty of articulating a reasoned but clear definition of how much weight this factor among others may be given, means that such a ruling would create far too much exposure to disruptive and costly litigation. As Justice Powell properly recognized in Bakke, if the standards described above are met—and they are here—there "is no warrant for judicial interference in the academic process." 438 U.S. at 319 n.53.

D. The Law School’s Special Attention To African-American, Hispanic And
Native American Applicants Is Based On Reasoned Principle

Petitioner contends (Pet. Br. at 43) that the Law School’s policy is illogical and "haphazard" in the choice of racial or ethnic groups for which it shows a particular concern. It is not. That policy’s objective is to assemble a class that is both academically superior and richly diverse in a variety of ways that include, but certainly are not limited to, race and ethnicity. *The Law School therefore pays attention to the racial or ethnic background of every applicant, to the extent that it sheds any light on their experiences* and "likely contributions to the intellectual and social life of the institution." CAJA 314; id. at 7783, 7248. 79 [-50-]

 * This sounds quite different from the Brief's earlier position that black, Hispanic and Native American students were the only contributors to racial and ethnic diversity.

But the Law School’s desire for meaningful numbers of African-American, Hispanic and Native American students is, in several important respects, unique. *By virtue of our Nation’s unfortunate past and ongoing struggle with racial inequality,* such students are both uniquely likely to have had experiences of particular importance to the Law School’s mission, and uniquely unlikely to be admitted in meaningful numbers on criteria which ignore those experiences. The Law School’s goal of fostering interaction and understanding across traditional racial lines also particularly requires African-American, Hispanic and Native American students, since those are *the groups most isolated by racial barriers in our country.* For similar reasons, the educational pitfalls associated with isolation are particularly salient for these students. See supra p. 26.

 This last-page paragraph begins to address a question long neglected. If a school is determined to confer admissions preference based on race or ethnicity, to which groups should the preference be granted, and why? The criteria suggested in this paragraph are that the group be one of those that has suffered most from "our Nation's unfortunate past" of "racial inequality," and is "most isolated by racial barriers in our country." Such formulations raise many questions: Should the children of recent Hispanic and black immigrants, many of whom broke the law in their eagerness to live in the United States, receive preference based on "our Nation's unfortunate past"? (What brought them here is more likely their native country's "unfortunate past.") Are the "barriers in our country" that isolate many "African-American, Hispanic and Native American students" primarily "racial," or differences in language, family structure, parenting, cultural values and habits, educational preparation, verbal and mathematical reasoning ability, etc.? Dealing with such questions at this time, the Brief is not only at the disadvantage of having reached its last page. Criteria for giving preference imply justifications. Thus, in suggesting criteria, the Brief is constricted by the justifications for racial discrimination that the Court under its precedents is likely to permit.

If educational experience revealed a similar confluence of issues with respect to other discrete ethnic groups, the Law School would modify its policy to acknowledge that fact. 80 Petitioner’s complaint that the Law School ignores the "dozens of separate racial or ethnic groups" from which its white and Asian American students hail is wrong (because, as noted, the Law School does consider such information) and misses the point. Narrow tailoring does not require the Law School to blindly give the same "plus" to every ethnic group it can identify, regardless of its salience to the educational mission—and regardless of whether members of that group would be well represented in the student body anyway. Such a regime would be impossibly unwieldy, self-negating, and would serve no coherent interest whatsoever.


1. Pet. App. 29a. "Pet. App." refers to the Petition Appendix; "JA" refers to the Joint Appendix filed in this Court; "CAJA" refers to the Joint Appendix filed in the Sixth Circuit; "Tr." refers to the transcript of the trial, Record 331 (Vol. 1) through Record 345 (Vol. 15). [return to text]

2. Because this Court has held that Title VI imposes substantive obligations coextensive with the Equal Protection Clause, the decision in this case will bind private as well as public institutions. Pet. Br. at 20. Petitioner offers no basis for applying any different standards under 42 U.S.C. § 1981, and failed to preserve that argument in any event. See also Gen. Bldg. Contractors Ass’n v. Pa., 458 U.S. 375, 389-90 (1982) (declining to impose broader obligations under § 1981). [return to text]


[That liability may not be imposed under 1981 without proof of intentional discrimination is the thesis in this section of the opinion.]

3. JA 110. The Law School receives more than 3500 applications each year and makes approximately 1300 offers of admission to fill a class of around 350 students. See generally JA 156-203. [return to text]

4. JA 111-13; Tr. 14:110-11 (Lempert). As Dean Lehman explained, "there is one absolute baseline criterion upon which we will not compromise," and that is that "[w]e don’t want to admit students who we think won’t be able to make it. It’s not right and it’s not fair." Tr. 5:147. [return to text]

5. Members of these groups are referred to as "minority" students. [return to text]

6. And there were only 170 African-Americans with at least a 3.0 GPA and 160 LSAT, compared to 11,348 whites and Asian Americans. Law School Admissions Council, National Statistical Report, 1995-96 through 1999-00 (2001) (lodged with the Court by LSAC). Because the United States has relied on data outside the record to support its assertion that race-neutral alternatives are available, we reference some responsive information appropriate for judicial notice. See Fed. R. Evid. 201. [return to text]

7. The study demonstrated the impact of a race-blind policy on typical learning environments. With a hypothetical 58 minority students, there is a 76% chance that a first-year small section of about 43 students will have more than one or two African-American students and more than one or two Hispanic students. With only 16 minority students, the probability is 4%. The chance of having such modest, concrete diversity in a residential dormitory section would fall from 34% to 1%. The probability of a student being the only African-American in a small section would increase from 4% to 51%; and in a dormitory section from 18% to 69%. CAJA 6045-49.[return to text]

8. Petitioner’s year 2000 grids, JA 196-203, show that even a race-blind lottery for every applicant scoring above 150 (50th percentile) on the LSAT would have offered admission to about 15 African-American, 16 Hispanic, and 3 Native American applicants—a 79% decline. By preventing the Law School from accepting students with truly exceptional academic qualifications at a higher rate than those with less impressive credentials, a lottery would also seriously undermine its other educational goals. (The number of offers extended to applicants scoring 170 or over on the LSAT would, for example, fall by 88%.) Moreover, even these bleak results could not be sustained once it became known that the Law School was conducting such a lottery, because the pool would immediately be flooded with applications from lower-scoring white students who do not currently apply, CAJA 7902-03 (Orfield), and abandoned by high-scoring students—who place great weight on academic selectivity, and the national rankings driven by it. [return to text]

9. See infra pp. 33 n.51, 41 n.69. [return to text]

10. Petitioner asserts that the 1992 policy merely "ratif[ied]" a previous policy that included a goal of 10 to 12% minorities. Pet. Br. at 3-4. Although the 1992 policy was intended to "ratify [the Law School’s] attention in the past to race for purposes of establishing a diverse law school class," CAJA 7533 (Lempert), the policy "represent[ed] a major change" in the way applications were processed, id. at 7504. The mission of the Committee was to "rewrite, rethink, [and] redo the admissions policy" in order to ensure that "the policy was … constitutional" under Bakke. Id. at 7492; see also Tr. 3:70 (Bollinger). [return to text]

11. Pet. App. 26a; CAJA 7749-50 (Lehman), 7313 (Munzel), 7667, 7693 (Shields); Tr. 3:64 (Bollinger). [return to text]

12. Pet. App. 275a-76a. These disparities significantly overstate the size of the Law School’s "plus factor." As a vertical line drawn anywhere on the graph at JA 219 will illustrate, there would be large differences in average test scores between admitted white and minority students even if the process were entirely race-blind—because most of the minority students would still be in the bottom half of the pool. See Bowen & Bok, Shape of the River 29, 42 (2000) (demonstrating that race-blind admissions would eliminate only 14% of the test score gap at selective universities). [return to text]

13. CAJA 7289-90, 7687-88; Tr. 5:139-42. [return to text]

14. These numbers would be significantly higher but for the fact that the Law School naturally finds quite substantial diversity in a variety of attributes within the large pool of white and Asian American applicants with the highest academic credentials. JA 121-22. [return to text]

15. See generally JA 156-203. Note that petitioner’s "Selected Minority" grids exclude "Other Hispanic" applicants—all of whom are properly included among the category described as Hispanic in the Law School policy. CAJA 321, 477; see also, e.g., CAJA 7311 (confirming that some minorities are rejected even though whites with lower quantitative credentials are accepted) (Munzel); JA 182-83 (white applicant in 1998 in LSAT 151-53/GPA 2.75-2.99 "cell" admitted while all five African-American applicants in the same cell were rejected). [return to text]

16. Justice Brennan and his colleagues did not mean that Harvard’s admissions practices were constitutional only if justified by a remedial purpose. The language they chose ("so long as … necessitated by" rather than "if … justified by") makes that clear. See Pet. App. 18a & n.7, 19a & n.8. It also would have made no sense. Harvard’s policy was forthrightly non-remedial in motivation, 438 U.S. at 321-22, and therefore (for reasons ably explained by petitioners themselves (Pet. Br. at 35)) it could not have been rendered constitutional by an unarticulated remedial rationale. [return to text]

17. The other Justices forming that majority believed that the Constitution permits much more extensive and varied consideration of race in admissions; indeed, they voted to affirm the rigid 16-seat quota employed by UC-Davis. In other words, those Justices had much broader reasons for reversing the California Supreme Court because they believed it improperly foreclosed a wider spectrum of legal conduct than Justice Powell did. See City of Richmond v. J.A. Croson Co., 488 U.S. 469, 496-97 (1989) (noting that Justice Powell’s opinion in Bakke would permit consideration of race only to pursue narrowly "focused" objectives, not the "amorphous" goal of remedying societal discrimination). [return to text]

18. The State and federal courts have widely regarded Justice Powell’s opinion in Bakke as the controlling law. See, e.g., Univ. & Cmty. Coll. Sys. of Nev. v. Farmer, 930 P.2d 730 (Nev. 1997); McDonald v. Hogness, 598 P.2d 707 (Wash. 1979); United States v. Brown Univ., 5 F.3d 658 (3d Cir. 1993); Talbert v. City of Richmond, 648 F.2d 925 (4th Cir. 1981); UWM Post, Inc. v. Bd. of Regents of Univ. of Wis. Sys., 774 F. Supp. 1163 (E.D. Wis. 1991); Davis v. Halpern, 768 F. Supp. 968 (E.D.N.Y. 1991); Uzzell v. Friday, 592 F. Supp. 1502 (M.D.N.C. 1984); Martin v. Charlotte-Mecklenburg Bd. of Educ., 475 F. Supp. 1318 (W.D.N.C. 1979), aff’d, 626 F.2d 1165 (4th Cir. 1980). The Fifth Circuit's contrary decision nearly twenty years later in Hopwood v. Texas, 78 F.3d 932 (5th Cir. 1996), has gained few adherents. See, e.g., Smith v. Univ. of Wash. Law Sch., 233 F.3d 1188 (9th Cir. 2000), cert. denied, 532 U.S. 1051 (2001). [return to text]

19. See, e.g., Scalia, Commentary, The Disease as Cure: "In Order to Get Beyond Racism, We Must First Take Account of Race.", 1979 Wash U. L. Q. 147, 148 (describing Powell’s opinion as "the law of the land"); Blasi, Bakke as Precedent: Does Mr. Justice Powell Have a Theory?, 67 Cal. L. Rev. 21, 30 (1979) ("The Powell opinion is, after all, the key to assessing the precedential significance of the Bakke decision."); Dixon, Bakke: A Constitutional Analysis, 67 Cal. L. Rev. 69, 69 (1979) ("The actual ‘ruling’ in Bakke, stemming only from Justice Powell’s tiebreaking opinion, that race may be a factor but not the factor in the admissions criteria … has acquired wide pragmatic appeal."). [return to text]

20. Petitioner cites Nichols v. United States, 511 U.S. 738 (1994), in which this Court elected to forego Marks analysis altogether and overrule Baldasar v. Illinois, 446 U.S. 222 (1980), on the merits. Pet. Br. at 28-29. Bakke has not produced anything like the confusion that followed Baldasar. In any event, this Court could take that tack in Nichols only because it was resolved to overrule Baldasar either way. A similar approach here would require this Court to assume for purposes of decision that Bakke produced a binding holding, and then consider whether to overrule it under traditional principles of stare decisis. [return to text]

21. 44 Fed. Reg. 58,509, 58,510 (Oct. 10, 1979). [return to text]

22. Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 231 (1995) (quoting Ariz. v. Rumsey, 467 U.S. 203, 212 (1984)). [return to text]

23. See, e.g., Adarand, 515 U.S. at 218-19, 224-25 (describing adoption of intermediate scrutiny in Metro Broadcasting, Inc. v. FCC, 497 U.S. 547 (1990), as a "surprising turn" from the use of strict scrutiny in decisions such as Justice Powell’s opinion in Bakke); Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 286, 288 n* (1986) (O’Connor, J., concurring in part) (recognizing that this Court’s affirmative action cases reveal a "fair measure of consensus," including that "the promotion of racial diversity has been found sufficiently ‘compelling,’ at least in the context of higher education, to support the use of racial considerations in furthering that interest"); Metro Broad., 497 U.S. at 568; id. at 619, 621, 625 (O’Connor, J., dissenting); Johnson v. Transp. Agency, 480 U.S. 616, 638 (1987) (upholding gender-based affirmative action policy, and drawing a favorable analogy to the Harvard plan). [return to text]

24. See, e.g., 56 Fed. Reg. 64,548 (Dec. 10, 1991) (applying Bakke to financial aid); 59 Fed. Reg. 8756 (Feb. 23, 1994) (same); see also CAJA 787 (Brief for the United States as Amicus Curiae). [return to text]

25. The race-conscious admissions policies employed by the United States at its own universities, the military academies, would also be unconstitutional under petitioner’s reasoning. See Brief for Lt. Gen. Julius W. Becton, Jr., et al. as Amicus Curiae. [return to text]

26. State legislatures must, for example, constantly choose between distributing limited resources evenly across an array of relatively non-selective institutions designed to bring the benefits of higher education to the greatest number of citizens, or disproportionately funding a selective flagship research institution. States like California, Michigan, North Carolina and Virginia that have historically chosen the latter course might reasonably conclude that they can no longer justify it if the flagship institution can no longer admit a significant number of minority students. [return to text]

27. See Wightman, The Threat to Diversity in Legal Education: An Empirical Analysis of the Consequences of Abandoning Race as a Factor in Law School Admission Decisions, 72 N.Y.U. L. Rev. 1, 27-28 (1997); CAJA 2254-56 (Bok) (discussing Wightman’s findings). [return to text]

28. JA 209. Texas has increased these numbers marginally in recent years, but has not achieved meaningful diversity. And the strategies it employs are neither race-neutral nor a realistic option for the Law School. See, e.g., Brief for American Law Deans Association as Amicus Curiae. [return to text]

29. Nearly 600 of this Court’s 824 judicial clerks since 1980 were graduates of just six of these law schools (including the Law School). There would be serious negative consequences at the state government level as well. See Brief for Arizona State University College of Law. [return to text]

30. See Emergency School Aid Act, Pub. L. No. 92-318, §§ 701-720, 86 Stat. 354 (1972); Magnet Schools Assistance Program, Pub. L. No. 98-377, §§ 701-712, 98 Stat. 1299 (1984). [return to text]

31. Pub. L. No. 107-110, § 5301(a)(4)(A), 115 Stat. 1425, 1806 (2002) (codified at 20 U.S.C. § 7231). [return to text]

32. See generally, e.g., CAJA 2240-709 (expert reports), 5617-23 and 5641-42 (Syverud reports), 7515-18 (Lempert testimony), 7699-706 (Syverud testimony), 7749 (Lehman testimony). [return to text]

33. Aleinikoff, A Case For Race-Consciousness, 91 Colum. L. Rev. 1060, 1066 (1991). [return to text]

34. 34 House Judiciary Committee, Traffic Stops Statistics Study Act of 2000, H.R. Rep. No. 106-517, at 3-4 (2000); see also id. at 4-5 (noting that in some jurisdictions African-American drivers are five to twenty-one times more likely to be subject to traffic stops than are white drivers). [return to text]

35. Ayres & Siegelman, Race and Gender Discrimination in Bargaining for a New Car, 85 Am. Econ. Rev. 304, 313 (June 1995). [return to text]

36. See, e.g., Baugh, Racial Identification by Speech, 75 Am. Speech 362-64 (2000); Massey & Lundy, Use of Black English and Racial Discrimination in Urban Housing Markets: New Methods and Findings, 36 Urban Affairs Rev. 452, 461 (Mar. 2001). [return to text]

37. These lessons cannot be learned from books, or by lecture in a racially homogenous classroom. Dean Syverud testified that "the best active, Socratic teaching" provokes "direct and often painful dialogue between students who are forced by the method to confront and make explicit their deepest unexamined convictions about legal issues." CAJA 5619. As Dean Lehman explained, that classroom dynamic does not "work[] really, really well" unless its participants are "drawing on a broad range of backgrounds and experiences which are personal." Id. at 7747. Dean Syverud testified that racial heterogeneity improves the classroom dynamic even in classes "far removed from issues traditionally associated with race" such as insurance, id. at 5622, and that he has been unable to "recreate the dynamic of a diverse Socratic classroom" in racially homogenous classes. Id. at 7710-11. [return to text]

38. See CAJA 7909 (Orfield), 310, 2243, 5044, 5106. Dean Lehman testified that "there are significant numbers of Michigan students who come to the law school with very little prior contact with people of other races." Tr. 5:158; see also Tr. 6:116 (Orfield) (half of Michigan students have no or very little interracial contact prior to Law School). [return to text]

39. Available on the internet at http://www.stanford.edu/dept/news/ report/news/june16/classday-616.html [return to text]

40. See CAJA 8145-46 (James); Pet. App. 28a; CAJA 432-33, 473; see also, e.g., United States v. Va., 518 U.S. 515, 523 (1996) (noting district court finding that 10% female enrollment would be "‘a sufficient "critical mass" to provide the female cadets with a positive educational experience’") (citation omitted). As a result UCLA School of Law, for example, had to decide whether to place one or two of the 13 African-American students in this year’s entering class into each of eight first-year sections (which raises educational concerns related to isolation) or to place all those students into a subset of the sections, creating meaningful diversity there but leaving the other sections with no African-American students. See Brief of Amici Curiae UCLA School of Law Students of Color at 15-16. [return to text]

41. See, e.g., Bush v. Vera, 517 U.S. 952, 990 (1996) (O’Connor, J., concurring) (compliance with § 2 of the Voting Rights Act is a compelling interest); Roe v. Wade, 410 U.S. 113 (1973) (recognizing compelling interests in both maternal health and fetal life, at different stages of pregnancy); Austin v. Mich. State Chamber of Commerce, 494 U.S. 652 (1990) (compelling interest in reducing political corruption). [return to text]

42. In contrast, the interest in "broadcast diversity" asserted in Metro Broadcasting was (at best) on the periphery of the FCC’s legitimate functions and actually threatened to interfere with important First Amendment values. 497 U.S. at 616-17 (O’Connor, J., dissenting). [return to text]

43. This Court has frequently held that constitutional doctrines must be flexible enough to accommodate the unique needs of the educational environment. See, e.g., Bd. of Regents of Univ. of Wis. Sys. v. Southworth, 529 U.S. 217, 231-32 (2000) (First Amendment compelled-speech/funding doctrines modified for academic environments); Regents of Univ. of Mich. v. Ewing, 474 U.S. 214, 225 (1985) (due process review of "genuinely academic decision[s] … should show great respect for the faculty’s professional judgment"); Healy v. James, 408 U.S. 169, 171 (1972) (student speech rights limited by "the mutual interest of students, faculty members, and administrators in an environment free from disruptive interference with the educational process"); Widmar v. Vincent, 454 U.S. 263, 268 (1981) ("A university differs in significant respects from public forums such as streets or parks or even municipal theaters.").[return to text]

44. See Keyishian, 385 U.S. at 603; Sweezy, 354 U.S. at 250 (Warren, J.); id. at 262-63 (Frankfurter, J., concurring); Bakke, 438 U.S. at 312-14 (Powell, J.). The Law School’s desire for the educational benefits of such pluralism is not, contrary to Judge Boggs’s suggestion, either the moral or practical equivalent of the rigid Jewish quotas of an earlier era. The Law School does not have a quota of any kind, infra pp. 38-48, and there is a world of difference between a policy which strives for some diversity for educational reasons—and in which white and Asian American students compete for all the seats and consistently receive the overwhelming majority of them—and one which capped Jewish enrollment at a low, arbitrary number, dramatically limiting educational opportunities for no purpose other than expressing animus or disdain for Jews. [return to text]

45. That nexus is certainly much tighter than in Metro Broadcasting, where the FCC not only "presume[d] that persons think in a manner associated with their race," 497 U.S. at 618, but also that they would insist on disseminating that characteristic "minority" viewpoint regardless of market incentives, id. at 626-27. [return to text]

46. Metro Broad., 497 U.S. at 618 (O’Connor, J., dissenting); see also Shaw v. Reno, 509 U.S. 630, 647 (1993) ("think alike"); Miller v. Johnson, 515 U.S. 900, 911-12 (1995) (same); United States v. Va., 518 U.S. at 517 (stereotypes are "‘fixed notions concerning the roles and abilities of [minorities]’" which are "‘likely to … perpetuate historical patterns of discrimination") (citations omitted) (omission in original).[return to text]

47. As the United States itself explained in an amicus brief to this Court in Hopwood at 16 (No. 95-1773), the fact that a minority "student reared in this country is likely to have had different life experiences, precisely because of his or her race" does not "rest on impermissible stereotypes; … equate race with particular viewpoints; … [or] presume that all individuals of a particular race act or think alike." (Emphasis added.) [return to text]

48. Petitioner’s suggestion (Pet. Br. at 34) that the "role model" justification forwarded in Wygant was directed at "educational benefits" is simply incorrect. The school board in Wygant expressed a desire to produce minority role models "as an attempt to alleviate the effects of societal discrimination." 476 U.S. at 274. As Justice O’Connor recognized, that objective "should not be confused with the very different goal of promoting racial diversity among the faculty" for educational reasons-- which was not asserted in Wygant. 476 U.S. at 288 n.*. [return to text]

49. Wygant, 476 U.S. at 275 (Powell, J.). The FCC’s asserted interest in "broadcast diversity" in Metro Broadcasting was similarly untethered, because the FCC could not "suggest how one would define or measure a particular viewpoint that might be associated with race, or even how one would assess the diversity of broadcast viewpoints." 497 U.S. at 614 (O’Connor, J., dissenting). The interest therefore threatened to justify unlimited regulation of broadcasting to produce whatever mix of purportedly "racial" viewpoints the FCC chose to identify and favor. [return to text]

50. As explained above, "critical mass" is an educational concept and the range of overall minority enrollments likely to produce it is not "a matter for mystical and metaphysical inquiry," Pet. Br. at 31, but a straightforward inference from the Law School’s desire to have, for example, more than one or two African-American and Hispanic students in a typical small section. Supra p. 6 & n.7. That number is not based on the percentage of minorities in the population or the applicant pool. The Law School’s minority enrollment percentages do not correlate with Michigan’s population, see Respondents’ Br. in Gratz v. Bollinger at 48 n.68, and diverged from the percentages in the applicant pool by as much as 17.7% from 1995-2000. See JA 156-203; CAJA 1536, 5584, 5586. [return to text]

51. Petitioner also argues that, apart from whether race-neutral alternatives actually exist, the district court "found" that the Law School did not in fact consider them. Any such finding would be clearly erroneous (although review should in fact be de novo because the issue was decided on summary judgment, not at trial, see CAJA 99). As the Sixth Circuit properly recognized, Pet. App. 33a-34a, the record establishes beyond question that the Law School did consider, and implement, a wide variety of race-neutral recruiting and outreach strategies before and after its adoption of the 1992 policy. See also CAJA 401, 358, 7754-55, 7667-78. The district court actually faulted the Law School officials only for failing to write memos to the file about or "experiment with" options (such as lotteries, percentage plans, and lowering academic standards, see Pet. App. 251a) that obviously could not work without serious injury to the Law School’s other legitimate and central educational goals. That was an error of law. Infra p. 34 n.53. return to text]

52. The United States touts the minority enrollments in Florida’s "graduate, medical, and business schools." U.S. Br. at 16. As the dean of Levin College of Law at the University of Florida recently explained, race-conscious scholarships have been "crucial" to its (limited) success. Mills, Diversity in Law Schools: Where Are We Headed in the Twenty-First Century?, 33 U. Tol. L. Rev. 119, 129 (2001).return to text]

53. See Wygant, 476 U.S. at 280 n.6 (alternatives must serve the interest "‘about as well’" and "‘at tolerable administrative expense’") (citations omitted); Croson, 488 US at 509-10 (city had a "whole array of race-neutral" alternatives because changing requirements with a disparate impact "would have [had] little detrimental effect on the city’s [other] interests"). return to text]

54. The issue is more relevant to Gratz v. Bollinger, and is dealt with in greater detail in the University’s brief in that case. See also Brief for American Law Deans Association as Amicus Curiae. return to text]

55. Bowen & Rudenstine, Race-Sensitive Admissions: Back to Basics, Chron. Higher Educ., Feb. 7, 2003, at B7, B9. return to text]

56. Shape of the River 51; see generally id. at 46-52. return to text]

57. Moran, Diversity and its Discontents: The End of Affirmative Action at Boalt Hall, 88 Cal. L. Rev. 2241, 2247-48 (2000). return to text]

58. Bakke, 438 U.S. at 318 (Powell, J.). As the Sixth Circuit recognized, a focus on "experiential diversity" that willfully ignores experiences associated with race would produce "a narrowed and inferior version of the academic diversity currently sought by the Law School." Pet. App. 34a-35a. Judge Boggs’s dissent offered no real response, other than skepticism that "an experience with [racial] discrimination" was really "so much more important than any other experience germane to other legal issues." Id. at 120a. As this Court has recognized, "[i]t is not for the Court to say what is or is not germane to the ideas to be pursued in an institution of higher learning." Southworth, 529 U.S. at 232. return to text]

59. Available on the internet at http://www.stanford.edu/dept/pres-provost/ president/speeches/951004affaction.html return to text]

71. "Odds ratios" do not mean the same thing as the "probability" of admission. For example, petitioner’s statistician explained that an odds ratio of 81 means that an applicant was nine times as likely to be admitted. Tr. 2:121-23.[ return to text]

72. CAJA 7625-28 (Raudenbush). Petitioner’s statistician conceded the accuracy of this observation. See id. at 7469-70, 8597-99, 7466-67. [return to text]

73. 73 CAJA 7456-58, 7613-14. In 1995, for example, this methodology resulted in the exclusion of almost 40% of the minority admissions decisions from the analysis. CAJA 8603-05, 8982, 8595. Indeed, petitioner’s statistician found statistically significant differences in rates of admission only for 21 of the 240 cells in 1995. Tr. 2:143. [return to text]

74. The district court approved of the Larntz methodology, Pet. App. at 227a-28a, but did not actually rely upon it in resolving the narrow tailoring issues. Id. at 246a-52a. [return to text]

75. The pool of high-scoring minority students was much smaller in 1978 than today; Harvard and similar institutions were necessarily giving a substantial plus to minority students in order to achieve that goal. [return to text]

76. Shape of the River 118-255, 269; CAJA 2251 (almost 80% of white graduates support retaining or expanding race-conscious admissions). [return to text]

77. Orfield & Whitla, Diversity and Legal Education: Student Experiences in Leading Law Schools, in Diversity Challenged: Evidence on the Impact of Affirmative Action 160 (Orfield & Kurleander eds. 2001); see also CAJA 2251, 5870-81, 6210, 6213-18; Brief for Michigan Black Law Alumni Society as Amicus Curiae. [return to text]

78. Barbara Grutter’s application illustrates the point well. Although the Law School’s consideration of race may have decreased her chances of admission slightly in the abstract, if the issue were tried the evidence would show that she would not have been admitted even under a rigidly race-neutral policy. One hundred and thirty-five other white applicants in the same or higher "cells" than Ms. Grutter were rejected along with her in 1997; 35 white applicants from lower cells were admitted; and the wait list she was on included more than 500 applicants. See JA 175; CAJA 458. [return to text]

79. Petitioner argues that the Law School has drawn a special distinction between Puerto Rican applicants born on the mainland and those born in Puerto Rico, and between Mexican American and other Hispanic applicants. The Law School’s pre-1992 system did draw distinctions like these, but its current policy was revised to provide a special commitment to enrolling a "critical mass" of "Hispanics" generally. Supra p. 10 n.15; CAJA 321, 7263 (Munzel, director of admissions), 477 (Dean Lehman). Bulletins were printed for several years that failed to reflect the change, but that mistake was corrected by 1997. Compare CAJA 1729 with 1885. [return to text]

80. See, e.g., Pet. App. 213a n.15 (recognizing that Asian and Jewish Americans are also likely to have had unique experiences because of their ethnicity, but that they are "already being admitted to the law school in significant numbers" on race-neutral criteria); CAJA 7520-21 (same). [return to text]



Return to