The United States Supreme Court
The University of Michigan Admissions Lawsuits

Gratz v. Bollinger

The Reply Brief of Counsel for the University of Michigan (pp 1-10, 21-33, 39-50): Institutions of higher learning have a compelling interest in obtaining the educational benefits of a diverse student body, in which racial and ethnic diversity play an important part. Such diversity at a meaningful level cannot be achieved at a highly selective undergraduate college, like Michigan's, without considering race and ethnicity in admissions. (The complete brief is online here.)

Comments by Curtis Crawford,* indented in italics.


1. The College of Literature, Science, and the Arts ("LS&A") is the flagship undergraduate institution of the University of Michigan, one of the Nation’s leading public institutions of higher education. LS&A seeks to provide an education that will prepare its students to excel in their chosen vocations and to participate as active citizens in our democracy. The University of Michigan enjoys a first-rate reputation for educational excellence and attracts nearly two-thirds of its applicants and a third of its entering class from outside Michigan, including all 50 states and over 80 nations.

*LS&A is a selective institution.* It receives many more applications than it has available spaces.1 See Pet. App. 4a.2 A great many of these applicants are fully qualified to attend LS&A. See id. at 108a. Indeed, it is undisputed that LS&A admits only qualified applicants. See id. at 46a, 111a. Thus, in selecting students from this competitive pool, "admissions officers must decide which set of applicants, considered individually and collectively, will take fullest advantage of what the college has to offer, contribute most to the educational process in college, and be most successful in using what they have learned for the benefit of the larger society." CAJA 1500 (emphasis in original).

 * As evidence of LS&A's selectivity, Note 1 states that in 1997 the school enrolled 3,958 first-year students from more than 13,500 applicants. But comparing the number who applied with the number who enrolled does not show selectivity, since a great many who are admitted choose to attend elsewhere. One needs to compare the number who applied with the number admitted. The Brief does not provide that information. I have it for one year, 1995, the year Petitioner Gratz applied. 16,292 applied, and 12,124 were admitted, giving an acceptance rate of 74%. (Robert Lerner and Althea Nagai, "Racial Preferences in Michigan Higher Education," Center for Equal Opportunity: 1998, p 27, online here.)

Offers of admission are building blocks from which a university constructs its educational environment. There is a common misperception, especially when viewed from a single applicant’s isolated perspective, that offers of admission are entitlements based on grades and test scores. See id. at 1506. That fundamentally misunderstands the [-2-] nature of admissions at a selective institution: with a finite number of available spaces, many qualified applicants must be rejected despite strong credentials. Grades, test scores, and other academic indicators are important in assessing an individual’s potential contribution, but they are by no means the only relevant factors.3

The University considers a broadly diverse student body "an integral component of its mission" because such diversity "increase[s] the intellectual vitality of [its] education, scholarship, service, and communal life." Pet. App. 4a. Thus, the task of the admissions office is to assemble a class that will collectively "create an environment on our campus" that is diverse in many respects and that will foster the most vibrant educational atmosphere. CAJA 1483. To do so, LS&A selects from among qualified applicants with wide-ranging interests, achievements, experiences, talents, and beliefs, and different cultural, racial, ethnic, socioeconomic, and geographic backgrounds.

Racial and ethnic diversity is a single, though crucial, element of the diversity the University seeks. College is a coming of age, and college students, most of whom are between eighteen and twenty-two years old, learn as much from one another as through classroom study. See CAJA 1651-52, 1656-59, 1669-70, 1717, 1740. Living, working, and learning with a racially diverse group of peers provides opportunities for a richer exchange of ideas—whether or not explicitly touching on race—that reflects a wider range of life experiences. Such interactions allow students to develop a broad variety of academic and interpersonal skills, including understanding and tolerance with respect to race and ethnicity and the ability to thrive in a multiracial [-3-] environment. See id. at 1652-53. Neither petitioners nor the United States denies the value of the educational benefits of diversity.

Yet, without the presence of meaningful numbers of minority students on campus, these interactions—and the educational benefits they foster—cannot take place. See CAJA 1668-69, 1732-36, 1836; Pet. App. 40a.4 Meaningful racial and ethnic diversity in a wide variety of educational environments is necessary both to ensure that minority students do not feel isolated or pressured to "represent" their racial or ethnic group, and to break down stereotypes by allowing students to view each other as unique individuals. See CAJA at 1734-40, 1834-36.

2. LS&A endeavors to enroll a broadly diverse student body while simultaneously maintaining its commitment to academic excellence. To achieve these two essential aspects of its mission—in light of the very small pool of potential minority applicants with competitive academic qualifications—*LS&A must consider race or ethnicity as a factor in making admissions decisions. To achieve the educational benefits of diversity, LS&A also pursues means other than considering race in admissions.* LS&A vigorously recruits minority students with competitive academic credentials, both to maximize the number of such students who apply and to increase the percentage of those admitted who choose to enroll (the "yield"). See Pet. App. 42a. LS&A’s minority recruiting and outreach efforts are a year-round campaign that includes identifying and contacting talented minority students from across the country; attending recruiting fairs in areas with substantial minority populations; hosting workshops for high school counselors; maintaining an office in Detroit to recruit local high school students, most of whom are minorities; coordinating campus visits for minority high school students; [-4-] enlisting the aid of current students to encourage admitted minority students to enroll; and hosting a Spring Welcome Day and other events for admitted minority students. See id.; CAJA 1480-81, 1487-88.

 * In noting that Michigan takes race into account, when finding as well as admitting applicants, the Brief implies that if the latter is unlawful as racial discrimination, the former would be also.

The University has found, however, that these targeted recruiting and outreach efforts are simply not enough. See Pet. App. 42a; CAJA 1441. That is because the number of potential minority applicants with competitive academic qualifications is very small—both in absolute terms and relative to the number of qualified non-minority applicants. See Pet App. 42a; CAJA 1441, 1492. For example, *of all high school students in the United States in 1999 with a grade point average of B or above and SAT scores of 1200 or above, only 6% were African-American, Hispanic, or Native American.* See CAJA 4016. Likewise, of all students in the State of Michigan in 1999 with those grades and scores, only 5% were African-American, Hispanic, or Native American. See id. at 4017. In addition, intense competition with other selective institutions for these highly sought-after students compounds this pool size problem by depressing the yield. See id. at 1492, 3992.


* This statement may be true, but what is its relevance to the LS&A's potential applicant pool? Michigan draws from high-school seniors or graduates (not "all high school students") and its admittees, apart from those receiving racial, socio-economic, or athletic preferences, are predominantly A (not "B or above") students. On the other hand, combined SAT scores of 1200 are not required and confer only a negligible advantage. Insert the proper specifications, and the percentage of competitive black, Hispanic and Native Americans (BHN) that should be available to Michigan becomes much larger.

In 1997, the proportion of high-school seniors taking the SAT, who were BHN and reported grades of A-, A or A+, was 12.5% (See my piece, "Estimating the Impact on Minority Enrollment of a Nationwide Ban on Racial Preference in College Admissions" here.)

If, instead of A's, Michigan required combined SAT scores of 1200 or above, the number of BHN available would indeed be tiny. In 1997, according to the College Board, the proportion of testees who were BHN and scored above 600 was 1.5% in the verbal, and 1.4% in the math SAT.

This reality has practical consequences for LS&A’s admissions program. It means, first, that in making admissions decisions, LS&A must consider race and ethnicity to attempt to enroll meaningful numbers of minority students and thereby provide all students with the educational benefits of diversity. See Pet. App. 42a-43a; CAJA 1836, 1912-16.5 It also means that, *because the [-5-] number of qualified minority applicants is so limited,* LS&A ends up admitting virtually all minority applicants with competitive academic credentials.6 See CAJA 1472-73, 1491-92.


* In Note 5, to illustrate this point, the Brief asserts that "under a race-blind system, the number of minority students admitted [in 1996] would have dropped from 1,335 to 269—out of 10,363 total admitted students." Which is to say, a BHN proportion of only 2.6%.

**How does one square this estimate with the College Board data cited above, which indicate that the BHN proportion nationwide of A-student college applicants (just a year later) was 12.5%? Why was Michigan's applicant pool in this respect so radically different from the national pool? Unfortunately, there was no opportunity in court to examine the University's estimate of the number of BHN available on a race-blind basis, because Gratz v. Bollinger did not go to trial on the facts.

3. The Office of Undergraduate Admissions ("OUA") employs 20 full-time, professional admissions counselors who spend most of each year reviewing application files and making admissions decisions.7 Each of the more than 17,000 [-6-] applications OUA receives is read from front to back and evaluated by a counselor. The volume of applications requires procedures and routines to promote fairness and consistency, while preserving counselors’ ability to exercise judgment. See JA 223.8 A single, unitary set of guidelines, which is reviewed annually and altered periodically, governs the admissions process.9 The aim of the guidelines is to "blend the consistency of a formula with the flexibility of a review that is ultimately a matter of human judgment." Id. OUA instructs counselors that "[a]dmissions is more art than science, and these guidelines should not be read otherwise." Id.

Each counselor is assigned a geographic territory and is responsible for reviewing all applications from that region, regardless of the race of the applicant. See Pet. App. 38a. There is no separate review of minority applicants, see id., and LS&A does not employ quotas or numerical targets for admission or enrollment of minority students, see id. at 36a. Counselors assess a broad range of academic and other factors that the University believes are important in composing a class. To facilitate consistency, counselors [-7-] calculate a "selection index" by assigning numerical values for a variety of factors, up to a possible total of 150 points. See CAJA 1118, 1147.10 While the selection index is designed to guide counselors’ evaluations, it does not purport to identify every possible factor that might bear on an admissions decision.

Consonant with LS&A’s commitment to academic excellence, the vast majority of available points are for academic factors. Of the 150 total potential points, 110 are available for academic factors, and a maximum of 40 are available for other factors. See JA 224.

The predominant factor is high school grade point average. Eighty points are available for GPA from tenth and eleventh grades, excluding non-academic courses. See id. at 225. Standardized test scores play a relatively smaller role in the admissions process, accounting for up to 12 points. See id. at 224, 230-31. 11


LS&A gives 2 admission points for each 1/10 of a GPA point, up to 80 admission points for a GPA of 4.0, and

0 admission points for an ACT score of 1-19 or for an SAT of 400-920
6 admission points for an ACT score of 20-21 or an SAT of 930-1000
10 admission points for an ACT score of 22-26 or an SAT of 1010-1090
11 admission points for an ACT score of 27-30 or an SAT of 1200-1350
12 admission points for an ACT score of 31-36 or an SAT of 1360-1600

High school grades are very important, while test scores in the upper ranges make almost no difference.

I owe these details to a copy of the official U-M admission guidelines online here, posted by the Michigan Review.

Counselors use the expertise they develop concerning schools in their geographic region to evaluate several non-quantitative academic factors that provide a more detailed [-8-] portrait of an applicant’s academic promise. Applicants may receive up to ten points for the academic strength of their high school. See id. at 226-28, 240.12 In addition, "[g]iven the wide disparity in high school course selection and offerings," counselors are asked to assess the rigor of each applicant’s course of study based on their knowledge of the meaning of course labels (e.g., "honors") used by the schools in their territory. Id. at 228. Counselors subtract up to four points for an applicant who chose a weaker curriculum when a stronger one was available, and add up to eight points for an applicant who selected more challenging courses. See id. at 228-30.

In addition to the 110 points available for academic qualifications, applicants also may receive a maximum of 40 points for other factors that the University believes indicate an applicant’s potential contribution to the life of the campus. See id. at 231. Some of these factors relate to specific attributes; others are more open-ended, allowing counselors to take into account factors not specifically identified on the selection index, such as community volunteering, playing the trombone, writing poetry, or working part-time.

Reflecting the University’s commitment both to Michigan residents and to broader geographic diversity, counselors assign ten points for Michigan residency, six additional points for residency in underrepresented Michigan counties (e.g., those in the State’s Upper Peninsula), and two points for residency in underrepresented states (e.g., many in the West and South). See id. at 232-33. Applicants receive four points if a parent is an alumnus of the University or one point if another close relative, such as a sibling or grandparent, is an alumnus. See [-9-] id. at 233-34. An applicant may also receive 20 points for one—but only one—of the following: socioeconomic disadvantage, membership in an underrepresented minority group13 [defined in Note 13 as blacks, Hispanics or Native Americans], attendance at a predominantly minority or predominantly socioeconomically disadvantaged high school, recruitment for athletics, or at the Provost’s discretion. See id. at 238-39. Based on information in the application, essay, or high school counselor’s recommendation as to activities, work experience, and awards, counselors may assign up to five points for leadership and service. See id. at 236-38. Counselors may give up to five additional points for personal achievement as evidenced by persistence, character, commitment to high ideals, and level of awards. See id. at 235-36. Finally, counselors also may award up to three points for the personal essay, taking into account originality, organization, subject matter, and writing quality. See id. at 234. No matter how many non-academic factors might apply to a given applicant, no applicant may receive more than 40 (of the 53 available) points for any combination of such factors. See id. at 231.


Thus, an applicant may receive a maximum of 40 points from non-academic factors, plus 110 points for academic factors, a total of 150. The weight of the latter is more than two/thirds, but by no means all. The 10 points for Michigan residency, plus 6 more for "underrepresented" county residency, plus 4 for alumni parentage helps to explain why two/thirds of the successful applicants are from Michigan. The 20 points for membership in an "underrepresented" minority group, socio-economic disadvantage, attendance at a mostly minority or disadvantaged high-school, or athlete status, give a B-student who belongs to such a group the points of an A-student who does not.

How many points are needed for admission? This Brief does not say, but the Brief for Petitioner Gratz (p 8) asserts that LS&A's guidelines specify definite admission for applicants with 100 to 150 points, and possible admission for those scoring 90-99. Thus, A-students who lack the 20-point preferred-group advantage may still be admitted by a combination of non-academic points for leadership and personal achievement, and academic points for harder high-school work and high SAT scores.

Recognizing that the selection index score may not always reflect all of the ways an applicant might contribute to LS&A, in 1999, the University formed the Admissions Review Committee ("ARC") to evaluate more complex admissions cases. For these cases, every member of the ARC closely reviews each applicant’s entire file; the whole committee discusses the applicant’s strengths and weaknesses; and the committee decides whether to admit based on its assessment of the candidate (without further reference to the selection index). See id. at 273-75. [-10-]

Counselors have broad discretion to decide whether an application warrants this additional level of in-depth review. See id. at 257. A counselor may, in his or her "discretion and judgment," "flag" an application for ARC review if the counselor determines that the applicant meets three criteria: (1) is academically prepared to do the work at LS&A; (2) has a selection index score greater than 75 for non-Michigan residents and greater than 80 for Michigan residents; and (3) *possesses at least one of a variety of qualities or characteristics important to the University’s composition of its freshman class, such as underrepresented race, ethnicity,* or geography; high class rank; socioeconomic disadvantage; unique life experiences, challenges, circumstances, interests or talents; connections to the University community; or athletics. Id. at 257-60. **These wide-ranging criteria allow counselors the discretion to flag for ARC review a broad array of "any number of applicants, including applicants other than under-represented minorit[ies]."14** Pet. App. 40a.


* This gives BHN applicants a second bite at the apple. In addition to the 20 bonus points based on their race or ethnicity, they may now receive an unlimited number of extra points to carry them over the top. Applicants in various other categories may receive the same boost, but preference based on those categories, unlike race, is not presumptively against constitutional and statutory law.

** Note 14 denies that all applicants who receive ARC review are BHN, but avoids stating what proportion are.

During the admissions season, decisions are made periodically, based on selection index scores, for all applications then pending. To avoid overenrollment, LS&A sets and adjusts, when necessary, the selection index levels that trigger the three possible admissions outcomes— admittance, deferral, and denial. See JA 275. Flagged applications that are not admitted based on the selection index are sent to the ARC for review and final decision. See id.

The result of this admissions process is a student body of remarkable talent and diversity.

* * * * *



* * * * *

B. Institutions Of Higher Learning Have A Compelling Interest
In Obtaining The Educational Benefits Of Diversity
That Justifies The Consideration Of Race And Ethnicity.

Setting aside stare decisis considerations, Justice Powell’s controlling opinion in Bakke was correct to hold that obtaining the educational benefits that flow from a [-22-] diverse student body is a compelling interest. See 438 U.S. at 306. 32

In concluding that "the interest of diversity is compelling in the context of a university’s admissions program," id. at 314, Justice Powell noted that "our tradition and experience lend support to the view that the contribution of diversity [to education] is substantial," id. at 313. He emphasized that a far-reaching consensus recognized the direct link between positive educational outcomes and a broadly diverse student body: "[t]he atmosphere of ‘speculation, experiment and creation’—so essential to the quality of higher education—is widely believed to be promoted by a diverse student body." Id. at 312 (quoting Sweezy v. New Hampshire, 354 U.S. 234, 263 (1957) (Frankfurter, J., concurring)).33 This consensus is deeply rooted in our Nation’s history and case law, confirmed by abundant empirical evidence, and consistent with the requirements that define a compelling interest. [-23-]

1. Our Nation has struggled throughout its history with the challenge of harnessing our diversity as a source of strength rather than a cause of friction. When our first President, George Washington, became concerned that the cultural, ethnic, religious, and regional differences among the colonists were creating tensions that endangered the new Nation, he drew upon his experience as leader of the Revolutionary Army. That Army was comprised of young men from across the colonies who, in Washington’s words, had "imbibed" "prejudices" against one another. Exposing this diverse group of soldiers to each other, Washington found, served to "eradicate" many of these "prejudices" and create a shared sense of purpose. Seeking to harness this educational process, Washington advocated the creation of a National University to meet the urgent need for education through exposure to America’s diversity:

 [T]hat which would render [a National University] of the highest importance, in my opinion, is, that the Juvenal period of life, when friendships are formed, & habits established that will stick by one; the Youth, or young men from different parts of the United States would be assembled together, & would by degrees discover that there was not that cause for those jealousies & prejudices which one part of the union had imbibed against another part . . . . prejudices are beginning to revive again, and never will be eradicated so effectually by any other means as the intimate intercourse of characters in early life, who, in all probability, will be at the head of the councils of this country in a more advanced stage of it.

Letter to Alexander Hamilton, September 1, 1796, Washington Writings (Library of America 1997); see also Joseph Ellis, Founding Brothers 154 (Knopf 2001). 34 [-24-]


A lovely foreword, but does it support Respondent's policies? It beautifully expresses the desirability of combating faction and prejudice by means of a regionally diverse college student body. However, the means that Michigan has chosen include preferential admission for members of some racial and ethnic groups at the expense of others. Not so, Washington. When he wrote, Americans already included many ethnic groups: English, Scots, Germans, Dutch, French, etc. But his proposal does not encompass discrimination in favor, say, of the Germans and French against the more numerous and powerful English.

As part of the promotion of racial and ethnic diversity in today's colleges and universities, faculty appointments, course offerings and reading assignments are commonly based on race or ethnicity. Nothing in Washington's letter sides with such means for eradicating "jealousies & prejudices."

Echoing Washington’s point nearly two centuries later, Justice Powell observed in Bakke that "it is not too much to say that the 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." 438 U.S. at 313 (internal quotation marks and citation omitted). Well before Bakke, this Court recognized the importance of diversity in educational settings, noting in Sweatt v. Painter, 339 U.S. 629, 634 (1950), that legal education "cannot be effective in isolation from the individuals and institutions with which the law interacts." See also Brown v. Board of Educ., 347 U.S. 483, 493-95 & n.11 (1954). Bakke thus was a capstone to—not a departure from—an established line of precedent acknowledging the importance of diversity in education. 35

2. This Court’s recognition that diversity generates important educational benefits is reinforced by a remarkably uniform and non-ideological consensus among the country’s leaders, educators, and social scientists. The President of the United States and the President of the University of Michigan share a conviction that college students "live and learn" better "with people from many backgrounds," and both "strongly support . . . racial diversity in higher education." 36 The record below contains unrebutted "solid [-25-] evidence" explaining how and why a student body that includes meaningful racial and ethnic diversity generates educational benefits for all students. Pet. App. 22a. Even petitioners do not disagree. See CAJA 4157.

*Racial and ethnic diversity is educationally important because, notwithstanding decades of progress, there remain significant differences in our lives and perceptions that are undeniably linked to the realities of race. Continuing patterns of residential segregation, for example, mean that "the daily events and experiences that make up most Americans’ lives take place in strikingly homogenous settings." Id. at 1968. 37 As a result, most students entering college have had few opportunities for meaningful interactions across lines of race and ethnicity. This separation contributes to misconceptions and mistrust, and provides little opportunity to disrupt racial stereotypes or to experience the richness of different racial and ethnic communities.* See id. at 1953, 1955-56. **Petitioners’ position—that race has no place as a consideration in college admissions outside of remedying discrete instances of past discrimination—would require universities to ignore the world in which they are educating their students to live and lead.**


* Largely true.

** False. "Petitioners' position" does not "require universities to ignore the world in which they are educating their students to live and lead." On the contrary, it demands that they stop ignoring the historical consequences of governmental discrimination based on race or ethnicity: racial/ethnic injustice, oppression, resentment, conflict, hatred, warfare.

Rather than relying on racial stereotypes or race as a "proxy," the benefits of diversity depend on the hardly debatable proposition that being part of a racial majority or minority, in a society in which race still so profoundly matters, will inform one’s perspective and base of knowledge. Cf. J.E.B. v. Alabama, 511 U.S. 127, 148-51 (1994) (O’Connor, J., concurring) ("We know that like race, gender matters . . . one need not be a sexist to share the intuition that in certain cases a person’s gender and [-26-] resulting life experience will be relevant to his or her view of the case.").

In jury selection cases, for example, members of this Court have recognized that jurors of different racial backgrounds may well bring varying experiences and prejudices to bear, such that the "distorting influence of race is minimized on a racially mixed jury." Georgia v. McCollum, 505 U.S. 42, 68 (1992) (O’Connor, J., dissenting); see also id. at 60 (Thomas, J., dissenting) ("the racial composition of a jury may affect the outcome of a criminal case . . . I do not think that this basic premise of Strauder has become obsolete."). *Exclusion* of an identifiable group of persons removes from consideration

 qualities of human nature and varieties of human experience, the range of which is unknown and perhaps unknowable. It is not necessary to assume that the excluded group will consistently vote as a class in order to conclude
. . .that its exclusion deprives the jury of a perspective on human events that may have unsuspected importance[.]

Taylor v. Louisiana, 419 U.S. 522, 531-32 & n.12 (1975) (internal quotation marks omitted). While the unique interpretations of experience that different individuals bring may not "make an iota of difference" on any given issue, their *exclusion* deprives the proceedings of something valuable. Id. at 532. 38

 * At this point, the Brief characterizes the relative absence of "an identifiable group" as an "exclusion," a loaded word, that biases the argument. From the standpoint of diversity, the relative absence of particular groups from a student body may be a disadvantage. But when that absence is not intended by the school, it is misleading to call it an exclusion. The only racial exclusion in this case is the exclusion of white and Asian applicants in favor of black, Hispanic and Native Americans.

[-27-]Petitioners seem to believe that any acknowledgment that race has some bearing on an individual’s experience is to deny that person’s individuality. That is not so. *There is a sound distinction between acknowledging that race may affect an individual’s experience, and assuming that it determines an individual’s behavior or beliefs.* The particular conclusions that individuals draw, or the interpretations they give to their own experiences will, of course, be as varied as the individuals themselves.39

 * True.

*Thus, the argument for diversity in no way turns on the assumption that individuals of any given race will necessarily hold any particular set of views.* The educational benefits of diversity stem in large part from a very different phenomenon. Exposing students at a critical period of personal development to situations in which they cannot predict viewpoint or behavior based on race actually undermines and deters stereotypical thinking. See CAJA 1656-59, 1736-40. This type of learning can occur only in an environment with meaningful racial and ethnic diversity and corresponding opportunities for students of different races and ethnicities to interact in and out of the classroom. See [-28-] id. at 1732-34. 40 Put bluntly, "[t]eaching that ‘not all blacks think alike’ will be much easier when there are enough blacks around to show their diversity of thought." Glenn C. Loury, The Anatomy of Racial Inequality 147 (2002). 41 And such diversity of thought is much more likely to emerge in settings where there are more than token numbers of minority students, so that individual minority students do not feel isolated or pressured to act as "representatives" of a racial group. See CAJA at 1734-36, 1835-36; see also Bakke, 438 U.S. at 323 (recognizing risk that lack of meaningful numbers of minority students leads to "sense of isolation"). The give-and-take from perspectives informed by the widest variety of human experience makes the most of the diversity of our Nation. This is the heart of the "robust exchange of ideas," Bakke, 438 U.S. at 312, that the University seeks to foster.

 * True. Unfortunately, this process is often impeded by university encouragement of separate minority living quarters, student associations and academic departments, all of which are fostered by the university's acceptance of the principle of racial preference.

Diversity leads not only to greater understanding of the ways in which race may—or may not—be relevant to a whole range of issues, but it also generates broader educational benefits. Abundant empirical evidence confirms [-29-] educators’ long-held beliefs and experience that "[s]tudents who experienced the most racial and ethnic diversity in classroom settings and in informal interactions with peers showed the greatest . . . growth in intellectual engagement and motivation[.]" CAJA 1652. 42 Likewise, experience with diverse peers in an educational setting better equips students "to understand and consider multiple perspectives, deal with the conflicts that different perspectives sometimes create, and appreciate the common values and integrative forces that harness differences in pursuit of the common good." Id. at 1652-53. A successful education, particularly at the undergraduate level, encourages students to move beyond familiar habits of thinking by confronting the "unknown." This is a crucial aspect of learning how to approach and understand the unfamiliar—whether it be a classmate, a foreign culture, or a newly-encountered text.43


Let us agree that a university needs diversity of viewpoint, and that racial and ethnic difference can both add to this diversity, and bring other educational benefits. Foremost among these benefits, especially when this opportunity has previously been lacking, is the chance to experience and mix with students of other racial and ethnic groups. With luck, this can reduce interracial prejudices and antagonisms.

The question, however, is whether a university established by the states and/or supported by the federal government should practice racial discrimination in order to obtain these benefits. Throughout human history, on every continent, governmental discrimination based on race has led to racial injustice, oppression, resentment, antagonism, hatred, conflict and warfare. Anyone familiar with our history should know that the United States is no exception to this rule. By the mid-20th Century, it seemed that the Western world had recognized the disastrous folly of racial discrimination. In the United States, the Civil Rights Act of 1964, and kindred statutes at the local, state and federal level, repudiated discrimination for or against anyone based on race or ethnicity, in education, employment, housing, public accommodations, government services, voting, etc.

But for many, the lesson had only been half learned. They agreed that government must not discriminate against people of color, but it was all right, under the name of affirmative action, to discriminate in favor of nonwhites at the expense of whites. This was touted as a temporary exception to the rule, but soon ceased to be temporary or exceptional. Government and the universities are powerful teachers. When their policies teach that racial discrimination in behalf of certain minorities is praiseworthy, millions learn the lesson. And the practice spreads in colleges from the admission of students to the allocation of rooms to the organization of student activities to the hiring of faculty to the assignment of readings to the creation of courses to the establishment of academic departments to the definition of offenses to the provisions of due process.

Petitioners and some of their amici claim that any consideration of race in admissions causes stigmatic harm to minority students which outweighs any benefits that increased levels of diversity might bring. See, e.g., Brief for National Association of Scholars as Amicus Curiae 23. William Bowen and Derek Bok definitively refute that [-30-] assertion in their comprehensive study of the effects of race-conscious admissions policies at selective colleges and universities. See Shape of the River at xxxi.44 Their findings demonstrate that this "assertion withers in the light of the evidence," id., because *"if minority students were truly demoralized, one would expect that they would be less likely than whites to succeed in graduate and professional schools, less likely to appreciate their college experience, and less inclined to report that they benefited intellectually by having attended a selective school," but "[n]one of these results appears in [the] data," id. at 261. Bowen and Bok found that the overwhelming majority of African-American graduates of selective colleges and universities with race-sensitive admissions programs performed well and were very satisfied with their undergraduate educational experience.* See id. at 265. 45

 * On the question whether preferences stigmatize blacks in their own and/or in the eyes of others, the Bowen/Bok conclusions are relevant, but hardly sufficient. In their survey of black and white graduates, the authors could have asked respondents to state their belief as to whether black admissions had been substantially due to racial preference, and whether this belief had affected their view of black ability and/or performance. Since no such questions were asked, the existence or degree of stigmatization was indirectly inferred from other answers. The omission of the questions was not impressive social science, but the authors did not pretend to be scientists. They were ex-presidents of Princeton and Harvard, respectively, strong proponents during their tenure of race-based admissions. An extensive critique of the Bowen/Bok study is on this website here.

Bakke’s holding has an even stronger empirical foundation today than it did in 1978. As the court below correctly concluded, see Pet. App. 22a, 27a-28a, extensive social science research confirms the soundness of Justice Powell’s conclusion that "the attainment of a diverse student body. . . . clearly is a constitutionally permissible goal for an institution of higher education." Bakke, 438 U.S. at 311-12. [-31-]

3. The interest in obtaining the educational benefits of diversity is not only firmly grounded as an empirical matter, but it also fits squarely within the framework defining whether an interest is compelling as a matter of law. While strict scrutiny clearly requires a "skeptical view of all governmental racial classifications," Adarand, 515 U.S. at 228, it rejects petitioners’ argument that no interest other than remedying institution-specific past discrimination is capable of justifying race-conscious measures. See Pet. Br. 40. Strict scrutiny analysis requires "carefully examining the interest asserted by the government" precisely because it "does take relevant differences into account" in assessing whether an interest is sufficiently compelling. Adarand, 515 U.S. at 228 (emphasis in original) (internal quotation marks omitted); cf. Calif. Democratic Party v. Jones, 530 U.S. 567, 584 (2000) (cautioning that the compelling interest "determination is not to be made in the abstract"). Contrary to petitioners’ characterizations, the interest in obtaining the educational benefits of diversity is neither inappropriately amorphous nor timeless. Cf. Metro Broad., 497 U.S. at 613 (O’Connor, J., dissenting).

The University’s interest in providing its students with the concrete educational benefits generated by a diverse student body bears no resemblance to the generalized interest in remedying societal discrimination that has been rejected as "amorphous." Bakke, 438 U.S. at 307; Wygant, 476 U.S. at 276. Petitioners suggest that this Court’s specific rejection of the role model theory in Wygant is actually a broad rejection of any interest that might produce educational benefits. See Pet. Br. 35. This argument misconstrues the nature of the interest repudiated in Wygant. The plurality opinion, authored by Justice Powell, characterized the role model theory—as the school district did—as a subset of the broad remedial rationale. The Court did not treat the role model theory as an interest in the educational benefits of diversity. Rather, it described the interest in "providing minority role models for its minority students, as an attempt to alleviate the effects of societal discrimination[.]" Wygant, 476 U.S. at 274; see also id. at [-32-] 289 n.* (O’Connor, J., concurring) (distinguishing the asserted interest from "the very different goal of promoting racial diversity among the faculty" for educational reasons).

Nor does the interest in obtaining the educational benefits of diversity justify the consideration of race in admissions without any "logical stopping point." Id. at 275 (Powell, J.). The need to consider race and ethnicity in admissions is inherently time-limited because it stems from the disparities in academic qualifications, such as grades and test scores, between minorities and non-minorities, and *the correspondingly small number of minority students with competitive academic credentials.* **These phenomena are by no means permanent fixtures of our educational landscape. For example, recent studies show that test score gaps have narrowed, see, e.g., Shape of the River at 22 (noting that SAT gap between whites and African-Americans shrank from 282 to 181 points from 1976 to 1989)**; Back to Basics at B9 (observing additional improvement through 1995), and the University looks forward to the day when it is no longer necessary to consider race in admissions to compose a diverse student body.


* The "number of [BHN] students with competitive academic credentials" is not as small as Michigan indicates. (See p 4 supra and my comment.) According to 1997 data from the College Board, BHN students were 12.5% of the testees who placed in the top third with respect to high-school grades, and 8% of the top third in SAT scores. Also, they were 11.4% of the top fifth in grades, but only 1.5% of the top fifth in scores. (It is easy to see why a selective college that wished to maximize BHN enrollment would, like Michigan, give more weight to grades than to scores.)

** The "test score gaps" may or may not be "permanent"; but if not, they may well last for generations or centuries rather than years or decades. The 101-point reduction cited by the University is not a national figure: it is based on the applicants to only two colleges. The following figures for all tested students are from the College Board, which began publishing SAT mean scores by race in 1976. Between 1976 and 1997, the black-white gap shrank 63 points from 258 to 195. During the same period, the gap between whites and Mexican-Americans increased from 136 to 143. The Brief does not address the gap between Latinos and whites, which may grow rather than decrease, due to the influx of poorly educated, non-English speaking immigrants.

A comparison of the results of six nationwide tests given to black and white high-school seniors between 1965 and 1992 indicates a decline in the gap of .36 standard deviations from 1.18 to .82. (Larry V. Hedges and Amy Nowell, "Black-White Test Score Convergence since 1965," p. 154, in Christopher Jencks and Meredith Philipps, eds, The Black-White Test Score Gap. 1998: Brookings) However, no significant change occurred in the proportion of whites and blacks who scored in the top tenth, a category from which highly selective colleges and professional schools draw most of their applicants. (Ibid, p. 158.)


LS&A’s current admissions program is "properly devised" to achieve the meaningful racial and ethnic diversity that allows the University to realize the educational benefits of diversity. It hews closely to the "plus" system that Bakke approved, in which race and ethnicity are accorded competitive consideration in combination with other factors relevant to an applicant’s potential contribution to the student body. Because LS&A is unable to achieve, through active recruiting and outreach programs, sufficient racial and ethnic diversity to make possible the opportunities for interactions that facilitate the educational benefits of diversity, see CAJA 1836, 1912-16, and because no race-neutral admissions alternatives [-33-] currently are viable, LS&A properly considers race and ethnicity as factors in admissions.

A. LS&A’s Admissions Program Is Narrowly Tailored.

In devising and refining its admissions program, LS&A has relied on the principles set forth in Bakke. As Justice Powell observed, a system that considers race as one of many factors and treats each applicant as an individual reflects an appropriate fit between the objective of obtaining the educational benefits of diversity and the means by which that objective is achieved.

* * * * *

To the extent that petitioners use the term "separate standards" to capture the statistical fact that the average grades and test scores of admitted minority students are somewhat lower than the average grades and test scores of admitted non-minorities, that fact, standing alone, has no constitutional significance. Further, this phenomenon cannot be said to measure the "extent" to which race is taken into account, as it is in part an unavoidable consequence of the reality that, nationally and in the State of Michigan, average grades and test scores are lower for minority students. See CAJA 4012-23. This disparity means that, for example, *even if the University selected a particular SAT score and admitted all applicants with scores [-40-] above that level, regardless of race, the average scores of the admitted minority students would still be lower.* In any event, the gap in academic credentials is smaller than petitioners suggest. **Between 1995 and 1999, the average academic GPA was approximately 3.5 for non-minority applicants and approximately 3.2 for minority applicants.** See id. at 1879, 1906. This marginal difference in average GPAs remains roughly the same when comparing the data for admitted and enrolled students: it is approximately the difference between an A- (3.67) and a B+ (3.33).


* True. This is a function of the normal bell curve. When the mean of one distribution of scores is lower than another, its means will also be lower in the overlapping areas in the right tail.

** This is surprisingly close, but not very different from other figures I have seen. Lerner (op. cit. supra, p 1) gives the median high-school GPAs of LS&A's 1995 admittees. They were: blacks - 3.3, Hispanics - 3.4, Asians - 3.7, whites - 3.7. Median SAT scores tell a different story: blacks - 1020, Hispanics - 1120, Asians - 1300, whites - 1250.

The 20 points that the University attributes to race and ethnicity in its admissions process reflects LS&A’s judgment, based on its experience, regarding the proper balance to be struck among competing considerations: the goal of obtaining the educational benefits of a racially diverse student body; the need for more than token numbers of minority students on campus to make it possible to generate those benefits; the small size of the pool of minority applicants with competitive academic credentials; the interest in assuring that all students admitted are prepared to succeed academically; and the objective of enrolling a class that is broadly diverse in ways other than race and ethnicity. See Bakke, 438 U.S. at 316-18 (noting that universities may accord different weights to different admissions criteria in light of "the ‘mix’ both of the student body and the applicants" and the universities’ goal of obtaining the educational benefits of diversity). Another institution might strike the balance in a different place, depending on the particular values that institution seeks to further in the admissions process.

For LS&A, allotting fewer than 20 points would undermine its ability to achieve the educational benefits of diversity, because even with a 20-point "plus," LS&A is not able to achieve sufficient racial diversity to ensure that meaningful numbers of minority students consistently are present in formal and informal educational settings on campus. At the same time, it is LS&A’s judgment that allotting more than 20 points may tip the scale too far in terms of admitting students who might not be academically [-41-] prepared to succeed at the University. That would not only unacceptably undercut the University’s commitment to academic excellence, but it also would undermine its ability to secure the educational benefits of diversity because it would hinder the University’s efforts to provide *an educational environment in which students view each other as peers.*

 * In 1995 at U-M, could black admittees view white admittees "as peers," and vice versa? On the ACT college aptitude test, 75% of the white admittees scored 26 or above, while 75% of the black admittees scored 25 or below. This means that less than 25% of each group had scores in the same range. That same year, at Michigan State, 75% of the white admittees scored 26 or below, a range of scores similar to the black range at U-M. If the black contingent admitted to U-M had gone to MSU instead, there would have been indeed a racially diverse campus at which whites and blacks could realistically "view each other as peers." As things turned out, the gulf at U-M was nearly duplicated at MSU. On the ACT test, 75% of the white admittees scored 21 or above, while 75% of the black admittees scored 22 or below. MSU had lost, to more selective universities like U-M, most of the blacks who could have matched its white students in academic ability. (Data from Lerner, op. cit. supra, p 1)

Striking the proper balance among these factors is an exercise of educational expertise and judgment. *Competing admissions objectives, such as academic selectivity and broad diversity across categories other than race, will always limit the extent to which an institution considers race as a factor in admissions.* Requiring the University to accord measurably less weight to race would, as a practical matter, preclude any selective institution of higher education from employing any admissions program that gives a "plus" to race. Justice Powell was right to conclude that, within the framework set forth in Bakke, this fine-tuning ought to be left to the exercise of good-faith judgment by **education professionals who have the expertise and experience to balance the consideration of race against other concerns to advance the institution’s educational mission.** That is the workable balance that Bakke struck in prohibiting quotas, while permitting universities to consider race as one of many factors in an individualized, comprehensive review that evaluates each applicant as a whole person.


* Notice that the social evils and presumptive illegality of racial discrimination are not mentioned among the factors that "will always limit the extent to which an institution considers race as a factor in admissions."

** If the shoe of discrimination were on the other foot, if it were a question of racial preference for white and Asian over black and Hispanic applicants, would Michigan still argue that the decision should be left to "education professionals"?

It is surely a good thing for college students of different racial groups to experience and mix with each other, and for colleges to enroll enough minority students to make that possible. But is it good enough to justify the use of racial discrimination in order to secure it? Michigan extols the benefits of the end, but is silent on the evils of the means. Indeed, it avoids speaking even the names (discrimination or preference based on race or ethnicity) of the means. Thus, the many arguments against state-sponsored racial discrimination are left wholly unanswered in this Brief.

If racial discrimination in universities ceased, what would happen to the opportunity for campus interracial mixing? Many blacks and Hispanics would attend less competitive Michigan State, instead of Michigan U. Other blacks and Hispanics would attend less competitive Ferris State, instead of Michigan State. After all, the push for racial diversity has not increased the total number of blacks and Hispanics who go to college, and thus are available for interracial mixing. It simply has shifted part of this total from less to more selective schools. If the push stopped, many highly selective schools would be unable to give their students an extensive campus experience of blacks and Hispanics. On the other hand, the vast majority of schools—less or non-selective—would have more black and Hispanic students, with academic abilities more in line with their classmates. White and Asian students who want the experience of college life with a generous number of blacks and Hispanics would be free to apply to schools where this would exist.

B. The University Cannot Achieve Meaningful Diversity
Without Considering Race And Ethnicity As Factors in Admissions.

Without suggesting any viable race-neutral alternatives for achieving diversity, petitioners flatly assert that the University has not "meaningfully considered" any such alternative. Pet. Br. 12, 30. But as the district court specifically found, the University came forward with substantial, credible, and unrebutted evidence that "a race-neutral admissions program would substantially reduce the number of under-represented minority students in the [-42-] LSA’s incoming student body." Pet. App. 40a. The court rejected petitioners’ suggestion that a random selection of applicants who passed some fixed "qualification threshold" might suffice, because of the small size of the relevant applicant pool; noted that another so-called alternative— bare reliance on test scores at the expense of all other relevant criteria—would result in significant additional rejections of qualified minority applicants; and credited expert testimony as to the inefficacy of relying solely on family income as a predictor of racial and ethnic diversity, given the even smaller pool of highly qualified minorities in the lowest income strata. See id. 53 The United States echoes the unsupported assertion that race-neutral plans were ignored. See U.S. Br. (Gratz) 10, 18. It further asserts that LS&A has "ample race-neutral alternatives" at its disposal, id. at 18—approaches that it claims have "proven effective in meeting the . . . laudable goals of educational openness, accessibility and diversity in other States," id. at 11. The United States actually offers only one such alternative: percentage-based admissions programs such as those currently in place in the public university systems of Texas, Florida, and California. See id. at 13-14. 54 These programs, which guarantee some [-43-] form of admission to a fixed percentage of the graduates from each high school in the State, 55 are by no means "race-neutral." Nor have they been proven effective.56 Further, even were such plans minimally "effective" in maintaining some level of racial and ethnic diversity in those states’ public systems of undergraduate education, they would have [-44-] devastating effects on the University of Michigan’s ability to define and pursue its educational mission. 57 As an initial matter, the suggestion that the percentage plans in Texas, Florida, and California are "race-neutral" is erroneous. First, while facially race-blind, they unquestionably were adopted with the specific purpose of increasing representation of African-Americans and Hispanics in the public higher education system.58 Indeed, petitioners, the United States, and the universities themselves measure the success of these percentage plans by examining a single, race-driven data point, that is, the level of minority representation in the student body—the very criterion petitioners would have LS&A ignore. Such programs are deemed effective when minorities represent a specific percentage of the student body—a percentage roughly equivalent to that in place in the last year in which the university explicitly considered race in individual admissions.59

 I strongly agree. The X% admissions plans of Texas, Florida, and California were designed to serve a racial purpose: to increase the representation of blacks and Hispanics in state university enrollment. Plans designed to serve a racial purpose are not race-neutral.

Second, these percentage plans are not race-neutral because they knowingly are premised on racial segregation in a state’s elementary and secondary public school system. In Texas, for example, granting automatic admission to all [-45-] students in the top 10% of their high school’s graduating class would, it was thought, guarantee a pool of African-American and Hispanic freshmen, because the racial segregation of Texas high schools had resulted in sufficient majority-minority schools as to guarantee that almost all of the top 10% of students in many schools would be minorities.60 *It is simply hiding the ball to hold out percentage plans as a race-blind alternative, where they in fact manifest a less forthcoming form of race-consciousness.*

 * Exactly!

Moreover, the assertion that the few extant percentage programs have "proven effective" in attaining educational diversity is premature and inaccurate. U.S. Br. (Gratz) 11. Research on these recently-instituted programs is just beginning to generate data measuring their short-term impact. And that data shows that percentage plans have not been nearly so successful in achieving racial and ethnic diversity as their proponents might claim, particularly with respect to selective flagship institutions.61 *Further, such [-46-] plans may create perverse incentives, encouraging African-American and Hispanic parents to keep their children in low-performing, segregated schools,62 discouraging students from taking challenging classes in the interest of boosting class rank, and deterring students from pursuing extracurricular activities.63*

 * True, on all three points.

More fundamentally, even were the percentage plans proffered by the United States shown eventually to have some moderate success in achieving racial and ethnic diversity, they would require institutions of higher education that seek to enroll a highly qualified, broadly diverse student body to radically alter their missions. In Michigan, such a plan would be destructive.

First, imposition of a percentage plan in Michigan would fundamentally change the composition of LS&A’s student body, on which the school’s character, reputation, and educational excellence depends. With their exclusive focus on class rank, percentage plans are incompatible with the individualized, "whole-person" file review that LS&A employs and that Justice Powell praised. See Bakke, 438 U.S. at 318. Rather than focusing separately on each of the ways in which an individual might contribute to the overall strength and diversity of the class, percentage plans are a blunt instrument that deprives admissions professionals of [-47-] the flexibility to achieve the broad diversity that is crucial to the University’s mission.

For this reason, percentage plans cannot be considered more narrowly tailored to achieve the genuine compelling interest—the educational benefits of diversity—than the current LS&A admissions system. Indeed, in subordinating the interest in the whole-file review that Justice Powell emphasized to the interest in enrolling a certain percentage of minority applicants, these plans more closely resemble the Davis quota than either the Harvard plan or LS&A’s admissions program.

Percentage plans also would hinder the University’s ability to maintain its character as a national institution with a geographically diverse student body. Except for the University of California’s campuses at Berkeley and Los Angeles, none of the three percentage-plan States’ systems includes a flagship campus comparable to the University of Michigan at Ann Arbor,64 and the student body at LS&A is by far the most national in scope.65 In that sense, the University is similarly situated to many selective private [-48-] institutions. The effect of such a sea change on its ability to define itself cannot be overstated: the University would be an entirely different institution.

Second, a percentage plan in Michigan could not possibly generate a racially and ethnically diverse student pool. California, Florida, and Texas all are extremely populous States with very substantial populations of both African-Americans and Hispanics.66 Each encompasses numerous geographical areas that are majority-minority, and correspondingly embraces relatively large numbers of high schools in which African-Americans and Hispanics form the overwhelming majority of students at all levels of class rank.67 While Michigan’s secondary school system is highly segregated by race, the overwhelming concentration of African-Americans in the Detroit area means that the statewide number of majority-minority schools is dwarfed by the far greater number of Michigan schools that are virtually all-white. See CAJA 1985, 1951. Indeed, Michigan is overwhelmingly white.68 Hispanics and Native Americans [-49-] are not a majority in any county or school district in Michigan. See id. at 1985-88, 1965-67 (Table 3), 1974 (Table 7).

Thus, even were it otherwise feasible for the University to implement some sort of high school-based percentage plan without fundamentally altering its very identity—which it plainly is not—the effects on diversity would be disastrous, particularly with regard to Hispanic students. The statistical reality is that infinitesimal numbers of minorities would be admitted through such a program, except through a small number of primarily African-American high schools in the urban Detroit area. Because virtually all of LS&A’s Hispanic students come from outside the State of Michigan, under a percentage plan, LS&A would be hamstrung in its ability to enroll such students. And its African-American student population would be dominated by students from a small geographical area, frustrating the interest in geographic and other diversity within all sectors of the student body.

This Court has taught that a race-neutral approach is "viable" for purposes of narrow tailoring only if that approach would be "comparably consistent" with, or better satisfy, the other traditional, non-racial goals also at stake. Easley v. Cromartie, 532 U.S. 234, 258 (2001); see also Wygant, 476 U.S. at 280 n.6 (an alternative must "promote the substantial interest about as well and at tolerable administrative expense"). That is not the case here. The University is not required to adopt a proposed race-neutral alternative that would undermine traditional admissions standards, depriving it of the ability to judge applicants as individuals and admit a class that, viewed collectively, brings to the table a stimulating mix of talents, perspectives, backgrounds, and academic interests.

[-50-] The University of Michigan has a substantial interest in creating and maintaining a world-class, selective institution of higher learning, seeking to enroll a diverse array of the best students from within the State and beyond its borders. The University’s current admissions system represents a careful and measured effort to tailor its admissions processes to its legitimate—indeed, unchallenged— educational objectives.


1. In 1997, when petitioners filed this lawsuit, for example, LS&A enrolled only 3,958 first-year students from more than 13,500 applicants. See Pet. App. 4a. Last year, LS&A received more than 17,000 applications. [return to text]

2. "Pet. App." refers to the Petition Appendix; "JA" refers to the Joint Appendix filed in this Court; "CAJA" refers to the Joint Appendix [return to text]

3. In 1995, when petitioner Gratz applied to LS&A, more than 1,400 white and Asian-American students with lower adjusted high school GPAs or test scores than hers were admitted, while more than 2,000 white and Asian-American students with higher adjusted GPAs or test scores were rejected. See CAJA 590. [return to text]

4. As used herein, the term "minority" refers to African-Americans, Hispanics, and Native Americans. See infra n.13. [return to text]

5. A race-neutral version of the current admissions system would reduce dramatically the number of minority students admitted to LS&A. See Pet. App. 40a-41a; CAJA 1901-02, 1874-76. In 1996, for example, under a race-blind system, the number of minority students admitted would have dropped from 1,335 to 269—out of 10,363 total admitted students. See CAJA 1874, 1881. This would have a devastating effect on integration in important learning contexts at the University. Under such a system, for example, the likelihood that a 60-person residence hall would include at least three African-Americans and at least three Hispanics would plummet from 92% to 19%. See id. at 1917. The likelihood of being the sole African-American student in that residence hall would jump from 2% to 31% under a race-blind admissions system, and the probability of being the sole African-American in a 30-person Introductory Psychology section would more than double, from 33% to 71%. See id. at 1921. [return to text]

6. Petitioners’ suggestion that the University has a "policy" of admitting all qualified minority applicants is incorrect. See Pet. Br. 4. There is no such policy. The fact that virtually all qualified minority applicants are admitted is simply a description of admissions outcomes given that there are few such applicants. Moreover, although prior versions of LS&A’s admissions guidelines encouraged counselors to admit qualified minority candidates as quickly as possible, without postponing decisions on their applications, it is undisputed that the guidelines from 1999 forward discontinued this practice. See Pet. App. 113a, 118a. [return to text]

7. LS&A implemented its current admissions system through guidelines adopted to govern admissions for the class entering in the fall of 1999. That system remains in place today. The admissions programs governed by the 1995-98 guidelines included three race-conscious practices that the University undisputedly has discontinued and disavowed: (1) the use of grids that take race into account by setting forth admissions options for applicants with various combinations of qualifications; (2) the exemption of minority students from the practice of rejecting candidates with very low grades and test scores without counselor review; and (3) a procedure known as "protected seats" that used projections of expected applications from groups known to apply late in the process (including minorities) to pace the rolling admissions process to permit consideration of such applications. See JA 275. The district court concluded that, while the use of grids, standing alone, was not necessarily unlawful, the combination of the three practices was impermissible. See Pet. App. 47a-48a. Petitioners devote much of their brief to attacking these abandoned admissions practices. However, because the University did not cross-petition to seek review of the district court’s determination that these practices, taken together, were impermissible, those practices are not properly before this Court. See, e.g., Trans World Airlines, Inc. v. Thurston, 469 U.S. 111, 119 n.14 (1985) (cross-petition is required to alter the judgment below). [return to text]

8. The volume of applications and the presentation of applicant information make it impractical for LS&A to use the same admissions system as the much smaller University of Michigan Law School. The large number of applications precludes LS&A from having a single decisionmaker. Also, applications to LS&A are not accompanied by comprehensive reports like those received by the Law School (from the Law School Data Assembly Service, a national clearinghouse), containing data on applicants’ grades and test scores and comparing each applicant to others from the same undergraduate institution. [return to text]

9. Contrary to petitioners’ assertion, there are no "written guidelines" for minority applicants that are "separate from . . . guidelines applicable to all other races and ethnicities." Pet. Br. 21. The document petitioners quote, see Pet. Br. 4, dated 1995, is not a guideline or policy and was never used to guide admissions decisions. It simply summarizes the typical admissions outcomes for minority applicants under LS&A’s prior guidelines. See JA 80-81.[return to text]

10. Petitioners inaccurately describe changes LS&A made to its admissions programs since 1995. First, the changes made between the 1997 and 1998 guidelines were not related to this litigation. See Pet. Br. 5. The 1995-1997 guidelines were in effect when petitioners applied, but at the time the lawsuit was filed, the 1998 guidelines were already in effect. Second, the University did not stipulate that all changes to its admissions guidelines since 1995 were non-substantive. The Joint Proposed Summary of Undisputed Facts states that "[t]he development of the selection index for admissions in 1998 changed only the mechanics, not the substance, of how race and ethnicity [were] considered in admissions." Pet. App. 116a. This statement addressed only the change from the grids to the selection index and did not address any other changes to the admissions process. The district court did not contradict that statement when it determined that, in the aggregate, all of the changes to the admissions system—particularly the abandonment of protected seats and automatic rejections—were constitutionally significant. [return to text]

11.Test scores are highly correlated with grades and therefore do not add much value in predicting an applicant’s future academic performance. See CAJA 1929. [return to text]

12. Counselors rate the high schools in their territory based on factors including college attendance rate, Advanced Placement courses offered, and average standardized test scores. See JA 226-28. Every applicant from the same secondary school receives the same number of points for this factor. See id. [return to text]

13. LS&A considers African-Americans, Hispanics, and Native Americans to be underrepresented minorities for purposes of considering race or ethnicity in admissions. See CAJA 1471. Because LS&A receives sufficient numbers of applications from qualified white and Asian-American students, it can enroll meaningful numbers of such students without consideration of race or ethnicity in admissions. See id. at 1445. [return to text]

14. While race therefore may be a factor in deciding which applicants receive this more searching review, petitioners are incorrect to suggest that all—or even close to all—minority applicants are flagged for this process, see Pet. Br. 7. See Pet. App. 39a-40a. Nor are all of the candidates who receive flags minorities. See id. [return to text]

* * * * * * *

32 Petitioners attack a straw man by characterizing the asserted interest as one in "academic freedom," standing alone. See Pet. Br. 13, 36. The University has never claimed that an interest in "academic freedom" is a sufficient justification for the consideration of race in admissions. Academic freedom certainly does not immunize a university’s conduct from constitutional scrutiny. At the same time, this Court has recognized the "countervailing constitutional interest," rooted in the First Amendment, that universities invoke in selecting students to fulfill their educational mission, Bakke, 438 U.S. at 313; the autonomy and deference afforded educators in making educational judgments, see id.; Regents of the Univ. of Mich. v. Ewing, 474 U.S. 214, 225 (1985); and the importance of "[an] atmosphere of ‘speculation, experiment and creation’" in the higher education context, Bakke, 438 U.S. at 312 (quoting Sweezy v. New Hampshire, 354 U.S. 234, 263 (1957) (Frankfurter, J., concurring)). These principles undergird the compelling nature of the interest in achieving the benefits of diversity in the context of higher education. [return to text]

33 Justice Powell also grounded his conclusion in social science evidence. See Bakke, 438 U.S. at 315 n.50 & 318 n.51 (citing Manning, The Pursuit of Fairness in Admissions to Higher Education, in Carnegie Council on Policy Studies in Higher Education, Selective Admissions in Higher Education (1977)). [return to text]

34 James B. Angell, the third president of the University of Michigan, emphasized a similar theme in his 1879 commencement address, not long after the Civil War. See Nancy Cantor, A Michigan Legacy: Ensuring Diversity and Democracy on Campus, Mich. Alumnus (Summer 1998). [return to text]

35 This Court has long recognized the "special role" and significant mission of educational institutions in our democracy, Board of Educ. v. Pico, 457 U.S. 853, 921 (1982) (O’Connor, J., dissenting). See, e.g., Board of Regents of Univ. of Wisc. v. Southworth, 529 U.S. 217, 233 (2000); Plyler v. Doe, 457 U.S. 202, 221 (1982); Brown, 347 U.S. at 493. In enacting measures to reduce "racial isolation" in elementary and secondary schools, Congress has also repeatedly recognized the educational value of racially diverse educational settings. See Brief for Respondents, Grutter v. Bollinger, et al., No-02-241 at 21-22. [return to text]

36 Statement of President George W. Bush, Jan. 15, 2003, available at http://www.whitehouse.gov/
news/releases/2003/01/print/ 20030115-7. html. See also Statement of Mary Sue Coleman, President of the University of Michigan, Jan. 15, 2003, available at http://www.umich.edu/%7Enewsinfo/ Releases/
[return to text]

37 See also Erica Frankenburg, Chungmei Lee & Gary Orfield, A Multiracial Society with Segregated Schools: Are We Losing the Dream? (2003), available at http://www.civilrightsproject.harvard.edu. [return to text]

38 While several Justices of this Court have recognized that including persons of different races and sexes on juries may bring valuable diversity of perspective and experience, the Court has rejected the notion that jurors can be excluded on the basis of race or sex alone. See Batson v. Kentucky, 476 U.S. 79 (1986); J.E.B. v. Alabama, 511 U.S. 127 (1994). This rejection stems from the special function of the jury, where the guiding principle is impartiality. "Without denying the possibility that race, especially as an imperfect proxy for experience, makes a difference in jury decisionmaking (and, in some cases, legitimately so)," race-neutrality is the better course in the deliberately neutral context of a jury. Bush v. Vera, 517 U.S. 952, 1051 n.5 (1996) (Souter, J., dissenting). The "cost of the alternative is simply too great." Id. Those concerns have no place in the educational context where the institutional interest is not impartiality, but an "atmosphere of speculation," Bakke, 438 U.S. at 312, best achieved by the inclusion of the widest variety of perspectives. [return to text]

39 Justice Powell recognized that distinction in Bakke. In concluding that Davis had failed to provide empirical support for its asserted compelling interest in providing doctors to underserved areas, Justice Powell rejected the unproven assumption that minority graduates would be more likely to practice in underserved communities. See Bakke, 438 U.S. at 310-11. In contrast, because he concluded that a university’s interest in obtaining the educational benefits of diversity was compelling, Justice Powell evidently determined that that interest did not depend on an improper, stereotyping assumption. Dissenting in Metro Broadcasting, Justice O’Connor made the same point, noting that the link the majority drew between "race and behavior, especially when mediated by market forces" is precisely "the assumption that Justice Powell rejected" in Bakke. Metro Broad., 497 U.S. at 619. [return to text]

40 Justice Powell recognized that actual interaction with diverse peers is the vehicle by which diversity benefits students. In acknowledging the widely-held consensus that a vibrant education is linked to a diverse student body, Justice Powell quoted the statement of then-President of Princeton University, William Bowen, that "a great deal of learning occurs informally . . . through interactions among students of . . . different races." Bakke, 438 U.S. at 313 n.48 (citing William G. Bowen, Admissions and the Relevance of Race, Princeton Alumni Weekly 7, 9 (Sept. 26, 1977)). See Patricia Gurin, Evidence for the Educational Benefits of Diversity in Higher Education: Response to the Wood & Sherman Critique by the National Association of Scholars (May 30, 2001), available at http://www.umich.edu/~urel/ admissions/research/. [return to text]

41 This is also true for Hispanics, a population encompassing significant differences in both background and experience. "A diversity of views exists among Latinos, and the differences between the foreign born, regardless of their country of origin, and the native born and those between the English dominant and the Spanish dominant are most notable." See Pew Hispanic Center/Kaiser Family Foundation, 2002 National Survey of Latinos, Executive Summary 6 (2002), available at http://www.pewhispanic.org/
[return to text]

42 See generally Diversity Challenged: Evidence on the Impact of Affirmative Action (Gary Orfield & Michal Kurlaender eds. 2001) ("Diversity Challenged"); Sylvia Hurtado, et al., Enacting Diverse Learning Environments: Improving the Climate for Racial/Ethnic Diversity in Higher Education (1999); see also Sylvia Hurtado, Linking Diversity and Educational Purpose: How Diversity Impacts the Classroom Environment and Student Development, in Diversity Challenged at 187, 198. [return to text]

43 See Roxanne Harvey Gudeman, Faculty Experience with Diversity: A Case Study of Macalester College, in Diversity Challenged at 251, 271; Anthony Lising Antonio, et al., Effects of Racial Diversity on Complex Thinking in College Students (2003), available at http:// www.siher.stanford.edu; Patricia Marin, The Educational Possibility of Multi-Racial/Multi-Ethnic College Classrooms, in American Council on Education & American Association of University Professors, Does Diversity Make a Difference? Three Research Studies on Diversity in College Classrooms 61, 69 (2000). [return to text]

44 Amicus NAS’ brief misconstrues Bowen & Bok’s findings, see Brief for National Association of Scholars as Amicus Curiae 25 n.25, by omitting their conclusion that any "costs" of race-sensitive admissions are outweighed by the benefits: "In the eyes of those best positioned to know, any putative costs of race-based policies have been overwhelmed by the benefits gained through enhanced access to excellent educational opportunities." Shape of the River at 265. [return to text]

45 In Bakke, Justice Powell considered the possibility that "preferential programs may only reinforce common stereotypes holding that certain groups are unable to achieve success without special protection," but concluded that this risk was outweighed by the significant educational benefits of diversity to all students. 438 U.S. at 298-99. Bowen & Bok’s findings confirm that Justice Powell was right to discount the possibility of stigmatic harm resulting from considering race in admissions. [return to text]


53 An academically selective college can expect that only one in six qualified low-income applicants will be either African-American or Hispanic. See Thomas J. Kane, Racial and Ethnic Preferences in College Admissions, in The Black-White Test Score Gap 431, 450 (Christopher Jencks & Meredith Phillips, eds., 1998); Shape of the River at 46-50. [return to text]

54 The United States briefly alludes to other possible "race-neutral alternatives," suggesting that the University might actively seek other sorts of diversity, such as geographical diversity; "modify or discard facially neutral admissions criteria that tend to skew admissions results" against minorities; or open its doors "to the best students from throughout the State or Nation." U.S. Br. (Grutter) 13-14; see also U.S. Br. (Gratz) 14-15. As to the first suggestion, it is uncontroverted that LS&A already pursues such a policy, as it grants selection index points to, for example, students from underrepresented areas, from underprivileged socioeconomic backgrounds, and with extraordinary indicia of personal achievement and leadership potential. The United States’ second point actually supports LS&A’s educationally sound decision to consider non-academic factors in combination with standardized testing and to place more emphasis on high school grades than on test scores. Incomprehensibly, petitioners and the United States urge the Court to force LS&A to rely more heavily on standardized test scores, U.S. Br. (Gratz) 24 (urging reliance on "objective qualifications"); see also Pet. Br. 25, even as the United States promises that diversity could be better pursued were such data disregarded. Finally, as to the third suggestion, it is undisputed that LS&A already "opens its doors" to the best students from throughout Michigan, the other 49 States, and dozens of countries. [return to text]

55 Texas offers students graduating in the top 10% of any Texas high school admission to the public institution of their choice. California offers admission to the state system, though not to the school of one’s choice, to the top 4% of each high school’s graduates, provided they have completed requisite coursework. Florida guarantees admission, though not choice of school, to the top 20% of each high school’s graduates who have completed a prescribed curriculum. See generally Catherine L. Horn & Stella M. Flores, Percent Plans in College Admissions: A Comparative Analysis of Three States’ Experiences 19-22 (2003), available at http://www. civilrightsproject.harvard.edu/research/
affirmativeaction/tristate.php#full report; U.S. Comm’n on Civil Rights, Beyond Percentage Plans: The Challenge of Equal Opportunity in Higher Education vii-x (2002) ("Beyond Percentage Plans").
[return to text]

56 Even the authors of Texas’ percentage plan confirm that it has been ineffective in increasing diversity, is not a national model, and cannot substitute for direct, competitive consideration of race in admissions. See Brief for Authors of the Texas Ten Percent Plan as Amicus Curiae. See also Brief for Professors Glenn C. Loury, Nathan Glazer, John F. Kain, Thomas J. Kane, Douglas Massey, and Marta Tienda as Amici Curiae; Brief for the University of Pittsburgh, et al. as Amici Curiae. [return to text]

57 The district court considered and rejected the viability of a percentage plan like the one in Texas. See Pet. App. 40a-41a (crediting unrebutted expert testimony that a percentage plan would damage LS&A’s academic selectivity and result in a "spurious form of equality"). [return to text]

58 See Petition for Writ of Certiorari by the State of Texas et al. at 18-19, Hopwood v. Texas, 236 F.3d 256 (5th Cir. 2000), cert. denied, 533 U.S. 929 (2001) (conceding as much); Horn & Flores, supra, at 15-19; Patricia Marin & Edgar K. Lee, Appearance and Reality in the Sunshine State: The Talented 20 Program in Florida 11 (2003); John F. Kain & Daniel M. O’Brien, Hopwood and the Top 10 Percent Law: How They Have Affected the College Enrollment Decisions of Texas High School Graduates 3-4 (2003), available at http://www/utdallas.edu/research/ greenctr/; Beyond Percentage Plans, supra, at 16 (California), 53 (Florida). [return to text]

59 See U.S. Br. (Gratz) 14; U.S. Br. (Grutter) 14-17 and sources cited therein; Brief for the State of Florida as Amicus Curiae at 4, 8-10; Kain & O’Brien, supra, at 4-5.[return to text]

60 See Marta Tienda et al., Closing the Gap? Admissions & Enrollments at the Texas Public Flagships Before and After Affirmative Action 7 (2003), available at http://www.texastop10.princeton.edu/ publications/tienda012103.pdf; Kain & O’Brien, supra, at 4; see also Jeffrey Selingo, What States Aren’t Saying About the ‘X-Percent Solution,’ Chron. Higher Ed., June 2, 2000, at A43 ("Aides to Gov. Jeb Bush of Florida admit they settled on a 20-percent standard after computer models of 10-percent and 15-percent policies failed to produce enough black and Hispanic students."). [return to text]

61 "[T]he top ten percent admission policy is not an alternative to affirmative action and by itself can only achieve minimal campus diversity, even in the presence of high levels of [high] school segregation." Tienda, et al., supra, at 41 (emphases omitted); see also Beyond Percentage Plans, supra, at 19-24, 57-60, 65-66, 116; Marin & Lee, supra, at 23, 27-31, 34-36; Kain & O’Brien, supra, at 29-32; Horn & Flores, supra, at 50-51, 58-59 ("percent plans seem to have the least impact on the most competitive campuses, which have persisting losses in spite of many levels of efforts to make up for affirmative action").

Further, to the extent that some percentage plans have contributed to the presence of minorities in higher education, that limited success is dependent on race-conscious recruitment, financial aid, and support programs. See Horn & Flores, supra, at 58-59; Marin & Lee, supra, at 31- 33, 35-37 (Florida’s Talented 20 program expressly retained explicitly race-conscious scholarship, recruitment, and support programs). The legality of these efforts may well be at issue herein. See, e.g., Podberesky v. Kirwan, 38 F.3d 147 (4th Cir. 1994). [return to text]

62 The position of the United States is in direct tension with national efforts to desegregate secondary education and encourage mobility out of low-performing schools, for example, through the No Child Left Behind Act, Pub. L. No. 107-110, § 5301(a)(4) (A), 5301(a)(5), 115 Stat. 1425, 1806 (2002) (codified at 20 U.S.C. § 7231). [return to text]

63 See Selingo, supra, at A31 (noting that Florida’s percentage plan penalizes students at more competitive high schools and generally excludes strong minority students in integrated schools and magnet programs). [return to text]

64 It is telling, therefore, that Berkeley’s and UCLA’s ability to attract, enroll, and retain African-Americans, Hispanics, and Native Americans has been especially hindered by enforced race-neutrality and imposition of a percentage plan. See Horn & Flores, supra, at 39, 48; Beyond Percentage Plans, supra, at 18. [return to text]

65 While LS&A draws more than half its applicants and one-third of its student body from out-of-state, the same is not true of the flagship institutions in States with percentage plans. In 2001, 14% of University of Texas-Austin applicants and 9.5% of Texas A&M applicants were non-Texans, and out-of-state students comprised only around 7% and 5%, respectively, of those schools’ freshman classes. See Texas Higher Education Coordinating Board, First-Time Undergraduate Applicant, Acceptance and Enrollment Information for Summer/Fall 2001, 66-67, 42-43 (2002), available at http://www.thecb.state.
tx. us/ reports/ pdf/0515. pdf. Similarly, Florida’s flagship campus, the University of Florida, draws only 13% of its students from outside Florida. See Marin & Lee, supra, at 28. The University of California at Berkeley has an out-of-state undergraduate enrollment of just under 10%, see http://osr4.berkeley.edu/, and at UCLA such enrollment is a mere 6%, see http://www.ucla.edu.
[return to text]

66 Texas is projected to become a "majority-minority" state by 2005, and whites already are a minority among those of college age. See Tienda, et al., supra, at 6. Indeed, this is now the case in California, where non-Hispanic whites are already the minority.
http://quickfacts. census.gov/ qfd/states /06000.html. Looking specifically to potential college applicants, the demographics of these three states are telling. Among Texans between 15 and 19 years of age as of the 2000 census, 44% were white, 13% African-American, and 39% Hispanic. Among Californians of the same age cohort, 34% were white, 7% African-American, and 39% Hispanic. In Florida, that same youth population was 55% white, 21% African-American, and 20% Hispanic. See Horn & Flores, supra, at 26. Indeed, more than three-quarters of the Nation’s Hispanic population lives in those three states plus New York and Illinois; more than half live in California and Texas. See CAJA 1959.
[return to text]

67 See Horn & Flores, supra, at 27. For example, in Florida’s populous Miami-Dade County, 60% of the population is Hispanic and 21% is African-American. See http://factfinder.census.gov.[return to text]

68 According to the 2000 U.S. Census, among Michigan residents of 15-19 years of age, 75.7% were white, 15.2% African-American, 4.3% Hispanic, .6% Native American, and 1.8% Asian-American. See http://factfinder.census.gov/servlet. The pool of potential minority college applicants is even smaller: among Michigan students attending the 12th grade of high school in the 2000-2001 school year, 83% were white, 12% African-American, 2% Hispanic, 2% Asian-American, and 1% Native American. See http://nces.ed.gov/ccd/bat. [return to text]