Students for Fair Admissions, Inc. v. President and Fellows of Harvard College et al

Filing 509

MEMORANDUM in Opposition re 412 MOTION for Summary Judgment filed by M. B., K. C., Sarah Cole, Y. D., G. E., A. G., I. G., R. H., J. L., Fadhal Moore, Arjini Kumari Nawal, R. S., Itzel Vasquez-Rodriguez, Keyanna Wigglesworth. (Attachments: # 1 Exhibit Student Record)(Culleen, Lawrence)

UNITED STATES DISTRICT COURT FOR
THE DISTRICT COURT OF MASSACHUSETTS
BOSTON DIVISION

STUDENTS FOR FAIR ADMISSIONS, INC,

Plaintiff,
Civil Action No. 1:14-cv-14176-ADB
v.

PRESIDENT AND FELLOWS OF
HARVARD COLLEGE (HARVARD
CORPORATION),

Defendant.

MEMORANDUM OF AMICI CURIAE STUDENTS IN OPPOSITION TO PLAINTIFF’S
MOTION FOR SUMMARY JUDGMENT

TABLE OF CONTENTS
Page
INTRODUCTION AND SUMMARY OF ARGUMENT ............................................................. 1
DISTINCT INTEREST OF AMICI................................................................................................ 3
ARGUMENT .................................................................................................................................. 4
I.

Harvard’s race-conscious holistic admissions policy is narrowly tailored and its
flexible consideration of race does not unduly favor underrepresented ethno-racial
minorities at the expense of Asian Americans. ................................................................... 4
A.
B.

SFFA’s statistical analyses are predicated on false premises of merit and
do not show Harvard affords any undue preference for Black and Hispanic
students. .................................................................................................................. 8

C.

Harvard’s consideration of race is narrowly tailored because it has a
minimal effect on the selection rate of Asian Americans. .................................... 13

D.
II.

Harvard’s consideration of race is narrowly tailored because Asian
American applicants can also receive a “plus” due to their race. ........................... 7

Harvard’s race-conscious admissions policy is narrowly tailored because it
does not afford Black and Hispanic students an undue preference. ..................... 15

Any negative effect for Asian Americans observed in the admissions process is
caused by admissions practices that benefit white applicants and not raceconscious admissions. ....................................................................................................... 16
A.

The relevant comparison for an intentional discrimination claim is
between Asian Americans and similarly situated white applicants. ..................... 16

B.

SFFA must prove that Harvard’s current holistic process has a
discriminatory effect, is motivated by a discriminatory purpose, and that
there is some connection between the discriminatory motive and
challenged policy. ................................................................................................. 19

C.

Applying the Arlington Heights framework reveals that SFFA is not
entitled to summary judgment on its intentional discrimination claim................. 20
1.

There are genuine disputes over whether Harvard has any policies
advantaging white students over Asian Americans that have a
discriminatory effect. ................................................................................ 20

2.

SFFA has not demonstrated that any disparate effect in Asian
American admission rates comes from a discriminatory purpose as
set forth in the Arlington Heights factors. ................................................. 24
i

D.

A finding of intentional discrimination against Asian Americans would
not justify banning race-conscious admissions since such a ban would not
remediate the harm caused by a white admissions advantage. ............................. 27

CONCLUSION ............................................................................................................................. 30

ii

TABLE OF AUTHORITIES
Page(s)
Cases
Abbott v. Perez,
138 S. Ct. 2305 (2018) .............................................................................................................26
Anderson ex rel. Dowd v. City of Boston,
375 F.3d 71 (1st Cir. 2004) ......................................................................................................26
Fisher v. Univ. of Texas at Austin,
136 S. Ct. 2198 (2016) (Fisher II) .......................................................................................5, 15
Fisher v. Univ. of Texas at Austin,
570 U.S. 297 (2013) (Fisher I) ..................................................................................................5
Fudge v. City of Providence Fire Dept.,
766 F.2d 650 (1st Cir. 1985) ....................................................................................................21
Grutter v. Bollinger,
539 U.S. 306 (2003) ......................................................................................................... passim
Jones v. City of Boston,
752 F.3d 38 (2014) ...................................................................................................................21
McGuire v. Reilly,
386 F.3d 45 (1st Cir. 2004) ......................................................................................................20
Milliken v. Bradley,
433 U.S. 267 (1977) .................................................................................................................27
Mobile v. Bolden,
446 U.S. 55 (1980) ...................................................................................................................26
Regents of the Univ. of Cal. v. Bakke,
438 U.S. 265 (1978) ...................................................................................................4, 5, 15, 17
Village of Arlington Heights v. Metropolitan Housing Development Corp.,
429 U.S. 252 (1977) ......................................................................................................... passim
Other Authorities
Adam R. Pearson et al., The Nature of Contemporary Prejudice: Insights from
Aversive Racism, Soc. & Personality Psychol. Compass, no. 3, 2009.....................................28
Angela Duckworth et al., Grit: Perseverance and Passion for Long-Term Goals,
92 J. Personality & Soc. Psychol. 1087 (2007)........................................................................11
iii

Anthony P. Carnevale & Jeff Strohl, “How Increasing College Access is
Increasing Inequality, and What to Do About It,” in Rewarding Strivers:
Helping Low Income Students Succeed in College (Richard D. Kahlenberg
ed., 2010), https://www.luminafoundation.org/files/resources/increasinginequality.pdf. ..........................................................................................................................11
Bic Ngo & Stacey J. Lee, Complicating the Image of Model Minority Success: A
Review of Southeast Asian American Education, 77 Rev. of Educ. Res. 415
(2007) .......................................................................................................................................12
Brief of Columbia University, Harvard University, Stanford University and the
University of Pennsylvania as Amici Curiae, Regents of the Univ. of Cal. v.
Bakke, 438 U.S. 265 (1978) (No. 76-811), 1976 WL 181278 .................................................25
Christopher Wolsko et al., Framing Interethnic Ideology: Effects of Multicultural
and Color-Blind Perspectives on Judgments of Groups and Individuals, 78 J.
Personality & Soc. Psychol. 635 (2000) ..................................................................................28
Daniel Kahneman, Thinking, Fast and Slow (2011) ......................................................................29
Eric E. Noftle & Richard W. Robbins, Personality Predictors of Academic
Outcomes: Big Five Correlates of GPA and SAT Scores, 93 J. Personality &
Soc. Psychol. 116 (2007) .........................................................................................................11
Faye Crosby et al., Cognitive Biases in the Perception of Discrimination: The
Importance of Format, 14 Sex Roles 637 (1986) ....................................................................29
Gary Orfield et al., Alternative Paths to Diversity: Exploring and Implementing
Effective College Admissions Policies (2017) ........................................................................28
Geoffrey Beattie et al., An Exploration of Possible Unconscious Ethnic Biases in
Higher Education: The Role of Implicit Attitudes on Selection for University
Posts, 197 Semiotica 171 (2013) .............................................................................................29
Goodwin Liu, The Causation Fallacy: Bakke and the Basic Arithmetic of
Selective Admissions, 100 Mich. L. Rev. 1045 (2002) ............................................................14
Gregory M. Walton and Steven J. Spencer, Latent Ability: Grades and Test
Scores Systematically Underestimate the Intellectual Ability of Negatively
Stereotyped Students, 20 Psychological Sci. 1132 (2009) .......................................................11
J.W. Schofield, Causes and Consequences of the Colorblind Perspective in
Prejudice, Discrimination, and Racism 231-254 (J.F. Dovidio & S.L.
Gaertner, eds., 1986) ................................................................................................................28
James J. Heckman &Tim Kautz, Fostering and Measuring Skills: Interventions
that Improve Character and Cognition, National Bureau of Economic
Research 3-4 (Nov. 2013), http://www.nber.org/papers/w19656 ............................................11

iv

Jennifer A. Richeson & Richard J. Nussbaum, The Impact of Multiculturalism
Versus Color-Blindness on Racial Bias, 40 J. Experimental Soc. Psychol. 417
(2004) .......................................................................................................................................28
Jeremy Ashkenas, Haeyoun Park & Adam Pearce, Even With Affirmative Action,
Blacks and Hispanics Are More Underrepresented at Top Colleges Than 35
Years Ago, N.Y. Times (August 24, 2017) ..............................................................................17
Jerry Kang, Negative Action Against Asian Americans: The Internal Instability of
Dworkin’s Defense of Affirmative Action, 31 Harv. C.R.-C.L. L. Rev. 1 (1996) ....................17
Jerry Kang, Rachel D. Godsil & John V. Wintermute, Brief of Experimental
Psychologists as Amici Curiae in Support of Respondents, at 4, Fisher v.
Univ. of Texas at Austin, 136 S.Ct. 2198 (Nov. 2, 2015) (No. 14-981), 2015
WL 6774020 ................................................................................................................10, 11, 12
Kimberly West-Faulcon, Obscuring Asian Penalty with Illusions of Black Bonus,
64 UCLA L. Rev. 590 (2017) ............................................................................................14, 21
Laurie A. Rudman et al., “Unlearning” Automatic Biases: The Malleability of
Implicit Prejudice and Stereotypes, 81 J. Personality & Soc. Psychol. 856
(2001) .......................................................................................................................................28
Meike Bonefeld & Oliver Dickhäuser, (Biased) Grading of Students’
Performance: Students’ Names, Performance Level, and Implicit Attitudes,
Frontiers in Psychology (May 9, 2018),
https://www.frontiersin.org/articles/10.3389/fpsyg.2018.00481/full#B46 ..............................10
Patricia G. Device et al., A Gender Bias Habit-Breaking Intervention Led to
Increased Hiring of Female Faculty in STEMM Departments (2017),
https://psyarxiv.com/tdvy7.......................................................................................................29
Paul Jargowsky, Concentration of Poverty in the New Millennium: Changes in
Prevalence, Composition, and Location of High Poverty Neighborhoods, The
Century Foundation and Rutgers Center for Urban Research and Education 5
(Dec. 2013),
https://tcf.org/assets/downloads/Concentration_of_Poverty_in_the_New_Mill
ennium.pdf ...............................................................................................................................13
Phoebe C. Ellsworth & Samuel R. Sommers, How Much Do We Really Know
About Race and Juries? A Review of Social Science Theory and Research, 78
Chicago-Kent L. Rev. 997 (2003)............................................................................................28
Randall Kennedy, Blacks and the Race Question at Harvard in Blacks at
Harvard: A Documentary History of African-American Experience at
Harvard and Radcliffe (Werner Sollors et al., eds. 1993) .......................................................26

v

Samuel D. Museus & Rican Vue, Socioeconomic Status and Asian American and
Pacific Islander Students’ Transition to College: A Structural Equation
Modeling Analysis, 37 The Rev. of Higher Educ. 45 (2013) ...................................................11
See Olga Khazan, Ending the Extracurricular Privilege, One Man’s Mission to
Make College Admissions Sane (and Fair) Again, The Atlantic (Dec. 21,
2016), https://www.theatlantic.com/education/archive/2016/12/endingextracurricular-privilege/511307/ ............................................................................................13
Sherick Hughes et al., Causation Fallacy 2.0: Revisiting the Myth and Math of
Affirmative Action, 30 Educ. Pol’y 63 (2016)..........................................................................14
Sylvia P. Perry et al., Modern Prejudice: Subtle, But Unconscious? The Role of
Bias Awareness in Whites’ Perceptions of Personal and Others’ Biases, 61 J.
Experimental Soc. Psychol. 64 (2015).....................................................................................29
William C. Kidder, How Workable Are Class-Based and Race-Neutral
Alternatives at Leading American Universities?, 64 UCLA L. Rev. Disc. 100
(2016) .......................................................................................................................................28
William Kidder & Angela Onwuachi-Willig, Still Hazy After All These Years: The
Data and Theory Behind Mismatch, 92 Tex. L. Rev. 895 (2014) ...........................................28

vi

INTRODUCTION AND SUMMARY OF ARGUMENT
SFFA’s strategy of using Asian Americans as a cover to eviscerate race-conscious
admissions is not only divisive, it is misleading both legally and factually and wholly insufficient
for summary judgment. SFFA blurs two entirely different claims: (1) that Harvard intentionally
discriminates against Asian Americans to the benefit of white students and (2) that Harvard’s
race-conscious policy admitting ethno-racial minorities unduly edges out Asian American
applicants. But these two claims must be de-coupled. They call for distinct legal standards,
involve different facts, and compel different remedial interventions.
Students reject SFFA’s attempt to conflate these two distinct claims—intentional
discrimination versus the validity of race-conscious admissions—and Students urge this Court to
distinguish them. This brief is structured to assist in that task.
Section I demonstrates that SFFA is not entitled to summary judgment on its claims that
Harvard’s race-conscious policy violates the narrow tailoring framework set by the Supreme
Court. Harvard more than satisfies this standard. Importantly, Harvard’s flexible consideration
of race does not unduly favor underrepresented minorities at the expense of Asian Americans.
SFFA’s arguments mischaracterize the record and misconstrue the meaning of merit. Students
concur with Harvard’s conclusion that race-conscious admissions is essential for assessing an
applicant’s true potential and for cultivating a campus community that facilitates “new ways of
understanding,” prevents “ongoing feelings of isolation,” and prepares students to “assume
leadership roles.”1 Asian Americans are not unduly harmed by such a consideration. In fact, the
record and research indicate that as a community Asian Americans benefit from it. Students
acknowledge that a policy that values greater ethno-racial diversity may impact admissions
1

Report of the Committee to Study the Importance of Student Body Diversity, Dkt. 419, Ex. 45 at 8, 23; Report of
The Committee to Study Race-Neutral Alternatives, Dkt. 419, Ex. 47 at 9.

1

decisions for a small number of applicants, just as any attribute valued by a university may tip
the balance in some applicants’ favor. As the Supreme Court has recognized, this type of limited
impact is not discriminatory so long as the process meets certain criteria: it does not insulate
individuals from comparison, it flexibly considers all pertinent elements of diversity, and it
ensures race does not become the defining feature of an application. The record indicates that
Harvard’s policy promoting ethno-racial diversity satisfies all of these elements.
Section II addresses SFFA’s intentional discrimination claim by clarifying the appropriate
analytical framework (which SFFA obscures) and highlighting factual disputes. For SFFA’s
intentional discrimination claim the only comparison that matters is between Asian Americans
and whites, not other ethno-racial groups.

The reason is simple: the Supreme Court has

permitted giving a “plus” to underrepresented racial minorities when such a process is narrowly
tailored. But giving white students a “plus” over Asian Americans based on race is never
constitutional since it does not promote racial diversity or reduce racial isolation. Instead, the
governing legal standard is one where SFFA must establish that Harvard intended to discriminate
against Asian Americans to the benefit of whites. Indeed, SFFA acknowledges that the relevant
comparison for intentional discrimination involves Asian Americans vis a vis whites, not other
minority groups. (Dkt. No. 413 at 7.) (“First, Harvard intentionally discriminates against AsianAmerican applicants. Incontrovertible evidence shows that Harvard’s admissions policy has a
disproportionately negative effect on Asian Americans vis-a-vis similarly-situated white
applicants . . .” ) (emphasis added).
Second, in Section II, Students identify numerous factual disputes which compel
proceeding to trial rather than granting SFFA summary judgment. SFFA has provided virtually
no material evidence on the Arlington Heights factors, except for discriminatory effect, which is

2

an important factor but not sufficient to establish discriminatory intent. As to discriminatory
effect, any bias is most attributable to preferences afforded to “ALDC” applicants—recruited
athletes, children of Harvard College or Radcliffe alumni, applicants on the Dean’s or Director’s
interest lists, and children of Harvard faculty and staff members—not race-conscious admissions
because when SFFA’s expert excludes those applicants, the data show there is little difference
between the white admission rate and the Asian-American admission rate. The record is clear
that any systemic disadvantage that Asian Americans do face is predominantly benefitting whites
in admissions, not ethno-racial minorities. In short, SFFA’s divisive strategy of trying to pit
marginalized groups against each other is not even borne out by its own data. Finally, any
finding of intentional discrimination does not call for the remedy SFFA proposes: prohibiting
any consideration of race in admissions. An outright ban would exceed this court’s remedial
authority since the violation would not relate to Harvard’s race-conscious policy.
DISTINCT INTEREST OF AMICI
As articulated in Students’ July brief and incorporated herein,2 Students stand together as
a broad cross-section of ethno-racial minorities—identifying as Asian American, Black, Latino,
Native American, and Pacific Islander—to defend Harvard’s right to cultivate racial diversity
and reduce racial isolation to the full extent allowed by law.3 Students are situated differently
than both SFFA and Harvard in this litigation in light of their lived experiences as ethno-racial

2

Dkt. No. 440
Students were permitted to participate as prospective, current, and now alumni students who identify as racial
minorities and who wish to defend Harvard’s right to consider race to the full extent allowed by law. (Memorandum
and Order on Proposed Defendant-Intervenors’ Motion to Intervene, Dkt. 52 at 23 (June 15, 2015).) At this stage in
the litigation, Students have submitted over a dozen declarations to support their position and assist the Court. (See
Dkt. 31, Ex. 1; Dkt. 230, Ex. 1, Ex. 2.; Dkt. 440, Ex. 1.) Numerical limits were never placed on Students’
participation, but Students remain mindful of the impending trial. Students note that they will soon be moving to
participate at trial, and pending this Court’s decision on trial participation, Students would be prepared to submit
declarations from additional students and/or several Students’ individual application files which reflect the readers’
comments and scores.
3

3

minorities living in a country that still reckons with its racist past and present. Importantly,
Students depart from both parties with regard to SFFA’s intentional discrimination claim.
Students condemn any discrimination that exists, but fundamentally disagree with the remedy
SFFA seeks: a blanket ban on “any use of race or ethnicity in the educational setting” in a
manner that “does not permit those engaged in the decisional process to be aware of or learn the
race or ethnicity of any applicant for admission.” (Dkt. 1 at 119.) Unlike Harvard, Students note
the data in the record that suggests “ALDC” preferences disadvantage Asian Americans (as well
as other ethno-racial groups) in the admissions process vis a vis whites.
Though varied in background, Students have a shared—and distinct—interest in ensuring
credible allegations of white advantage are thoroughly investigated, redressed, and not confused
with race-conscious policies promoting diversity.
ARGUMENT
I.

Harvard’s race-conscious holistic admissions policy is narrowly tailored and its
flexible consideration of race does not unduly favor underrepresented ethno-racial
minorities at the expense of Asian Americans.
SFFA contends that Harvard’s admissions policy—which the Supreme Court previously

cited as the model for constitutional, whole-person review4—is not narrowly tailored on three
grounds: (1) Harvard purportedly engages in racial balancing, (2) uses race as more than a “plus”
factor in admissions, and (3) could achieve racial diversity through race-neutral policies. (Dkt. 1
at 104-109, 112-114; see also Dkt. 413 at 7-8.) As articulated in Students’ July Brief, the record
compels summary judgment in Harvard’s favor on all counts challenging its race-conscious
holistic admissions policy,5 and Students incorporate their arguments herein. This brief offers

4

See Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 323 (1978); Grutter v. Bollinger, 539 U.S. 306, 334-35
(2003).
5
See generally Dkt. 440.

4

additional observations to further expose fallacies in SFFA’s assertions, specifically their claim
that race operates as a predominant factor in admissions. SFFA’s flawed allegations only serve
to incite divisions among ethno-racial minorities—they are far too deficient to warrant summary
judgment.
The constitutional inquiry for evaluating SFFA’s claim begins with whether “a university
may institute a race-conscious admissions program as a means of obtaining ‘the educational
benefits that flow from student body diversity.’” Fisher v. Univ. of Texas at Austin, 136 S. Ct.
2198, 2210 (2016) (Fisher II) (quoting Fisher v. Univ. of Texas at Austin, 570 U.S. 297, 310
(2013)) (Fisher I). Once a university offers a “reasoned, principled explanation” for its pursuit
of the educational benefits of diversity, that decision is entitled to deference. Fisher II, 136 S.
Ct. at 2208. Here, Harvard asserts that student body diversity, including racial diversity, is
“integral” to its pedagogical objectives and institutional mission.6

Such diversity exposes

Harvard’s students to “new ideas, new ways of understanding, and new ways of knowing,”
“prepares [them] to assume leadership roles in the increasingly pluralistic society,” and prevents
the “ongoing feelings of isolation and alienation among racial minorities in [its] community.”7
Harvard has clearly articulated a “constitutionally permissible goal,” Bakke, 438 U.S. at 311
(1978), that has resulted in an increasingly diverse campus even though the full benefits of
diversity have not been attained.8
Thus, the primary inquiry is whether Harvard’s consideration of race in its holistic
admissions policy is narrowly tailored to achieve the benefits of student diversity. Grutter, 539
U.S. at 334. The Supreme Court has indicated that an “admissions program cannot use a quota
6

Report of the Committee to Study the Importance of Student Body Diversity, Dkt. 419, Ex. 45 at 5.
Report of the Committee to Study the Importance of Student Body Diversity, Dkt. 419, Ex. 45 at 8, 23; Report of
The Committee to Study Race-Neutral Alternatives, Dkt. 419, Ex. 47 at 9.
8
See generally Dkt. 440.
7

5

system,” but it may consider race or ethnicity “as a ‘plus’ in a particular applicant’s file” as long
as the applicant is not “insulat[ed] . . . from comparison with all other candidates.” Id. at 334
(internal citations omitted).

The process should employ a “highly individualized, holistic

review” which flexibly considers “all pertinent elements of diversity . . . although not necessarily
according them the same weight.” Id. at 309. Individualized review ensures that an applicant’s
race is not the “defining feature of his or her application.” Id. at 337. Narrow tailoring also
requires that the university determines in “good faith” that “workable race-neutral alternatives”
would not suffice and undertakes “periodic reviews” to ensure racial preferences remain
necessary. Id. at 339, 342.
Harvard has submitted evidence demonstrating that its admissions process is an
individualized, whole person review where race is considered flexibly as one of many factors.9
Indeed, Students who reviewed their application file observed how their race was considered
alongside their class, other identity characteristics, academic and artistic accomplishments, and
extracurricular activities and leadership positions.10 Students submit the following arguments
below: (1) race is considered a plus factor for Asian American applicants, too; (2) SFFA’s
statistical arguments are based on faulty premises which incorrectly define merit; (3) Harvard’s
consideration of race has a notably small effect on the Asian American selection rate; and (4)
race is not the predominant reason as to why African American and Hispanic applicants are
admitted. Accordingly, SFFA is not entitled to summary judgment on Counts II, III, V.

9

See generally Dkt. 435 at 39-43.
See, e.g., Dkt. 440-1, Ex. 1.1 at 5 (Decl. of A.A.); Dkt 440, Ex. 1.3 at 17 (Decl. of D.I.); Dkt. 440-1, Ex. 1.6 at 30
(Decl. of S.C.) Dkt. 440-1, Ex. 1.8 at 37 (Decl. of T.D.).
10

6

A.

Harvard’s consideration of race is narrowly tailored because Asian
American applicants can also receive a “plus” due to their race.

SFFA asserts that the ethnicity of Asian American applicants is “rarely seen as a positive
factor in the chances of admissions” and such applicants are dismissed as “uninteresting and
indistinguishable” from other Asian American applicants. (Dkt. 413 at 20.) Yet these claims are
undermined by substantial evidence in the record.
Under Harvard’s policy, no applicant is excluded from discussing how race or ethnicity
has influenced their interests, goal or experiences.11 Testimony reveals that all applicants,
including Asian Americans (who, as a group, are represented at nearly twice the level of Black,
Latino, and Native American candidates12), are subject to Harvard’s individualized and holistic
consideration in which race may be deemed a positive factor. (Dkt. 435 at 29-30.) Therefore,
admission officers view each candidate as an individual and recognize, for example, that a
Vietnamese applicant is a “very deserving” first generation student; a Nepalese student would
offer a “fascinating perspective” on campus; and an Indian student could positively “add to the
mix” at Harvard. (Dkt. 435 at 29-30.) Admissions officers also remark on the racialized
experiences and activities of Asian American applicants in a positive way, which would be lost if
SFFA prevailed on its requested remedy. For example, admissions officers noted a student’s
essay on the plight of exiled Tibetans and her parents’ refugee status and highlighted another
applicant’s involvement in the Asian community as editor of a local journal. (Dkt. 435 at 29-30.)
Similarly, Students contend that Harvard viewed their racial and ethnic background as a
positive factor—including those students who identify as Asian American. For example, D.L.
identified himself as a Chinese American on his application to Harvard. His admission file
11

Dkt. 435 at 41-42 and n. 13-14.
See, e.g., Dkt. 419, Ex. 33 at 107 (showing racial composition of admitted class of 2019: 14% African American,
14% Hispanic or Other, 24% Asian American, and 40% white).
12

7

revealed that his interviewer was “particularly impressed” by his discussion of race and how the
“model minority myth perniciously uses Asian Americans as a measuring stick, which hurts all
people of color.” (Dkt. 440, Ex. 1.3 at 17.) The interviewer observed that this called for a
“sophisticated understanding of equality and discrimination,” and D.L. was ultimately offered
early admission to Harvard. (Dkt. 440, Ex. 1.8 at 17.)

T.D. is a first-generation Vietnamese

American who wrote extensively about his experience as an immigrant, the challenges he faced
as an English Learner, and his racial and ethnic identity in his personal statement. When he
reviewed his application file, he saw comments on his essay that he had a “strong sense of []self”
and saw that his immigrant Vietnamese identity was highlighted on his summary sheet. (Exhibit
1 at 2.) When S.C. viewed her application file, she discovered that her leadership abilities and
diverse academic interests, which drew largely from her Chinese heritage and low-income status,
weighed strongly in favor of admission as positive indications of her “humor, empathy, and
humility.” (Dkt. 440, Ex. 1.6 at 29.) She believes that she benefitted from Harvard’s admissions
policy which allowed Harvard to “look at [her] as a whole person and view [her] qualifications in
the context of both [her] class and race.” Id. These students’ collective experiences directly
contradict SFFA’s assertions that Asian Americans do not benefit from Harvard’s race conscious
admissions process and that Harvard’s holistic process systematically disadvantages Asian
American applicants as a group.
B.

SFFA’s statistical analyses are predicated on false premises of merit and do
not show Harvard affords any undue preference for Black and Hispanic
students.

Much of SFFA’s argument is premised on the view that academic scores—predominantly
SAT scores and grades—are the benchmark for determining whether an admission system is fair

8

to students of different races.13 Using Harvard’s academic index (not its academic rating), which
is 98% based on SAT scores and grades (Dkt. 421, Ex. 145 at 16), Dr. Arcidiacono pools
applicants for the six years he studied and then ranks them by academic decile from lowest to
highest. (Dkt 419-31 at 40-42.) Asian American students fare the best in the academic index,
followed by white students; Hispanic and African American students lag behind. (Id.) He then
performs a series of analyses showing that based on their academic decile Asian American
students have a low likelihood of admission compared to Hispanic and African American
students. (Id. at 44-53.) He proceeds to show some correlation between academic index deciles
and other factors Harvard uses. (Id.) He concludes this section by summarily arguing that there
is a “significant penalty against Asian-American students as compared to other racial groups,
including whites, and a significant preference given to African-American and Hispanic
applicants in both the personal and overall rating.” (Id. at 53) (emphasis in original).
This is the heart of SFFA’s argument in this case, and it fails because it is based on a
faulty premise. While grades and SAT scores are appropriate considerations in assessing merit,
they do not define merit in themselves. Though Dr. Arcidiacono and SFFA characterize these
factors as “objective,”14 they are subject to their own biases that tend to “penalize” African

13

As a corollary to SFFA’s argument regarding the academic index, SFFA claims that Asian American applicants
fair better on other “observable” measures, such as extracurricular ratings and alumni ratings. Setting aside the issue
that these factors are observable and therefore objective in the eye of SFFA, SFFA’s position is belied by their own
data. For example, SFFA misleadingly asserts that Asian American applicants have “higher extracurricular ratings
than any other racial group.” Dkt. 414 at 123. But notably, SFFA selectively chooses to ignore that African
Americans have the highest share of applicants with the top extracurricular score in the top decile. Dkt. 414 at 123.
SFFA is similarly misleading when it summarily asserts that Asian Americans receive “higher overall scores from
alumni interviewers than all other racial groups.” Dkt. 413 at 13. The data presents a much more complex picture.
SFFA’s own table shows that in the top five deciles of the academic index—which SFFA elsewhere claims include
the most competitive applicants, Dkt. 413 at 15—African Americans actually have the highest share of high
“overall” scores from alumni and Hispanics have the next highest share.
14
See, e.g., Dkt. 413 at 13.

9

American, Hispanic, and other groups of students with less privilege and educational opportunity
(including certain Asian American students).
As discussed extensively in the Amicus Brief by Harvard Student Associations,
standardized tests are “tainted by racial bias” and “artificially depress[]” the scores of people of
color without adding predictive value.15 Moreover, high school GPA is subject to grade inflation
and also can be affected by racial bias.16

Much of the test bias and grade inflation that

systematically disadvantages underrepresented ethno-racial minorities can be explained by a
specific, well-documented phenomenon called stereotype threat, which is the pressure that
people feel when they fear that their performance could confirm a negative stereotype about a
group with which they identify or belong.17 Because intellectual stereotypes are ubiquitous,
relying on standardized test scores and grades as an indicator of academic achievement and
capacity will “systematically underestimate” the ability of many stigmatized minorities by nearly
two-tenths of a standard deviation.18 Racial disparities in achievement are “not the consequence
of the inherent capacities of students; rather, this is a consequence of the educational
environment.”19
Further undermining the claim that test scores and high school grades are equivalent to
merit, studies show these academic metrics are not an accurate predictor of college performance.
One study found that “[t]est scores and high school grades explain less than half the differences

15

Dkt. No. 455-1 at 16-23.
Meike Bonefeld & Oliver Dickhäuser, (Biased) Grading of Students’ Performance: Students’ Names,
Performance Level, and Implicit Attitudes, Frontiers in Psychology (May 9, 2018),
https://www.frontiersin.org/articles/10.3389/fpsyg.2018.00481/full#B46.
17
Jerry Kang, Rachel D. Godsil & John V. Wintermute, Brief of Experimental Psychologists as Amici Curiae in
Support of Respondents, at 4, Fisher v. Univ. of Texas at Austin, 136 S.Ct. 2198 (Nov. 2, 2015) (No. 14-981), 2015
WL 6774020, at **7-8.
18
Kang, supra, 2015 WL 6774020, at *5.
19
Kang, supra, 2015 WL 6774020, at *22.
16

10

in college freshman grades, and are even weaker determinants of the likelihood of graduation.”20
Another found that achievement tests “predict only a small fraction of the variance in later-life
success” and may not adequately capture the character skills valued in school and other
domains.21 Moreover, a growing body of empirical research shows that character skills rival
cognition in predicting educational attainment22 and such skills do not necessarily move in the
same direction as academic factors. For example, a 2007 study found that a lower SAT score is
associated with a greater degree of grit, which in turn, is associated with a higher GPA at an elite
institution.23 A related study found grit predicted completion of a “rigorous” summer training
program for incoming freshman at West Point better than any other measure, including high
school class rank and SAT scores.24 Grit has also been found to predict success better than
involvement with extracurricular activities.25 Another study concluded that conscientiousness
and a willingness to work hard matter more than SAT scores and high school GPA in
determining college GPA.26
In sum, research confirms that the most promising students are not always the ones with
the highest SAT scores or the best high school grades.27 While the result of holistic admissions

20

Anthony P. Carnevale & Jeff Strohl, “How Increasing College Access is Increasing Inequality, and What to Do
About It,” in Rewarding Strivers: Helping Low Income Students Succeed in College, 71, 110 (Richard D.
Kahlenberg ed., 2010), https://www.luminafoundation.org/files/resources/increasing-inequality.pdf.
21
James J. Heckman &Tim Kautz, Fostering and Measuring Skills: Interventions that Improve Character and
Cognition, National Bureau of Economic Research 3-4 (Nov. 2013), http://www.nber.org/papers/w19656.
22
Heckman & Kautz, at 4.
23
Angela Duckworth et al., Grit: Perseverance and Passion for Long-Term Goals, 92 J. Personality & Soc. Psychol.
1087, 1093 (2007).
24
Duckworth, at 1095.
25
Duckworth, at 1099.
26
Eric E. Noftle & Richard W. Robbins, Personality Predictors of Academic Outcomes: Big Five Correlates of
GPA and SAT Scores, 93 J. Personality & Soc. Psychol. 116, 126 (2007).
27
Kang, supra, 2015 WL 6774020, at *24. See Gregory M. Walton and Steven J. Spencer, Latent Ability: Grades
and Test Scores Systematically Underestimate the Intellectual Ability of Negatively Stereotyped Students, 20
Psychological Sci. 1132, 1337 (2009) (finding that both test scores and classroom grades are influenced by bias
associated with psychological threat such that these measures of academic performance underestimate the true
intellectual ability and potential of ethnic minority students); Samuel D. Museus & Rican Vue, Socioeconomic

11

may be an incoming class with seemingly lower academic measures, accounting for stereotype
threat by considering race as a plus factor for ethno-racial minorities (including underrepresented
Asian American ethno-racial groups) permits universities to implement an admissions process
that more closely represents a true meritocracy.28
Fadhal Moore, an African American graduate of Harvard (class of 2015), can testify to
this fact through his own experience:
“When I applied [to Harvard] some seven years ago, I had a great SAT score but it was
below the average Harvard freshman’s. My application, just like everybody else’s, had to
be looked at holistically. That includes my race, which has unfortunately been the source
of many of the hardships my family has endured. For those who say I did not ‘deserve’
my spot, I would encourage them to look at the A’s I earned while I was there.”29
Harvard recognizes that its academic and extracurricular scores alone do not adequately
capture an applicant’s strengths or future capacity for success.30 Accordingly, Harvard assigns
applicants a personal rating that summarizes all of the different aspects of the application,
including essays, letters of recommendation, the alumni interview report, personal and family
hardship, and any other relevant information in the application, many of which are not
considered in any other rating.

(Dkt. 420 at 11.)

It represents an admissions officer’s

assessment of a student’s qualities and character, such as “humor, sensitivity, grit, leadership,
integrity, helpfulness, courage, kindness and many other qualities.”

(Dkt. 420 at 11.)

Admissions officers do not specifically consider race when assigning the personal rating. (Dkt.
420 at 11.)

Status and Asian American and Pacific Islander Students’ Transition to College: A Structural Equation Modeling
Analysis, 37 The Rev. of Higher Educ. 45, 50 (2013) (noting “research on AAPIs suggests that even positive racial
stereotypes can pose a threat and lead to negative outcomes because they can result in anxiety, fear of failure to
conform to that stereotype [of the model minority]. . .”); Bic Ngo & Stacey J. Lee, Complicating the Image of Model
Minority Success: A Review of Southeast Asian American Education, 77 Rev. of Educ. Res. 415, 443 (2007).
28
Kang, supra, 2015 WL 6774020, at *24-25.
29
Dkt. 440, Ex. 1.11, at ¶8 (Decl. of Fadhal Moore).
30
See Dkt. 418 at 10.

12

The personal rating may be vulnerable to implicit bias and other distortions, but it is a
mistake to assume that these ratings are biased just because they do not correlate with academic
and extracurricular scores.31 As discussed above, academic characteristics may run counter to
personal ones. Moreover, grades, standardized test scores, and extracurricular involvement32
tend to be inherently biased against Hispanic and African American students. Just as academic
and extracurricular measures provide some information about merit, so do personal factors. The
solution is not to exclude the personal rating or the overall rating, but to monitor for bias.
C.

Harvard’s consideration of race is narrowly tailored because it has a
minimal effect on the selection rate of Asian Americans.

Harvard’s race-conscious admissions policy has a notably small impact on the percentage
of Asian American applicants admitted to Harvard. This is not only because Harvard’s use of
race is limited—it is also simple math. Harvard does not admit enough African American and
Hispanic students to have much impact on Asian American students as a whole.
One way of determining whether a particular admission policy has a significant effect is
to examine that policy’s effect on selection rates. A selection rate is the number of successful

31

Relatedly, Students note that any admissions system can contain elements that are vulnerable to bias. This is true
of both “subjective” as well as “objective” criteria. Institutions, including Harvard, have an obligation to critically
examine their own practices in case they reveal indications of bias or lack a sufficient nexus to legitimate missiondriven goals.
32
Participation in extracurricular activities is influenced by privilege and should not be viewed as an objective
measure of merit. Richard Weissbourd, a senior lecturer at Harvard, proposes a “healthy and fair admissions
process” that redefines achievement in ways that create “greater equity and access for economically diverse
students” who may vary widely by race, culture and class. See Olga Khazan, Ending the Extracurricular Privilege,
One Man’s Mission to Make College Admissions Sane (and Fair) Again, The Atlantic (Dec. 21, 2016),
https://www.theatlantic.com/education/archive/2016/12/ending-extracurricular-privilege/511307/. Since Black and
Latino families are more likely to live in neighborhoods with concentrated poverty, it logically follows that they face
even higher hurdles in acquiring what Weissbourd describes as “laundry lists of achievements and activities” which
tend to track social privilege. Id.; Paul Jargowsky, Concentration of Poverty in the New Millennium: Changes in
Prevalence, Composition, and Location of High Poverty Neighborhoods, The Century Foundation and Rutgers
Center for Urban Research and Education 5 (Dec. 2013),
https://tcf.org/assets/downloads/Concentration_of_Poverty_in_the_New_Millennium.pdf.

13

applicants by race divided by the number of total applicants from that race. This measure is used
by SFFA’s expert Dr. Arcidiacono as well as several academics.33
In a 2002 law review article, then-Professor and now-California Supreme Court Justice
Goodwin Liu, discussed what he called Causation Fallacy where he debunked the myth that
affirmative action materially impacts white applicants’ likelihood of admission.34 Subsequent
academics have built on Justice Liu’s research and applied it to the admissions rates of Asian
students as well as white students.35 Indeed, the authors of Causation Fallacy 2.0 provided
updated data to show that if no African American or Latino students were admitted to Harvard,
the admission rate for all other students would only increase by one percentage point: from
5.84% to 6.84%.36 Using the data from the experts in the case, the effect of Harvard’s raceconscious admissions policy on Asian American applicants is probably less than that. Harvard’s
expert Dr. Card estimates that if Harvard removed consideration of race, the Asian percentage of
Harvard’s class would increase by 3 percentage points37 which, given Harvard’s typical number
of admits (about 2,000),38 would increase the number of Asian American admits by around 60
students. Applying this to the class of 2018 admissions data for Asian Americans (where 400
Asian Americans were admitted out of 7,80639), the number of admitted Asian Americans would

33

Dkt. 419-31 at 109-10, Tables B.2-1 and B. 2-2; Goodwin Liu, The Causation Fallacy: Bakke and the Basic
Arithmetic of Selective Admissions, 100 Mich. L. Rev. 1045, 1075 (2002); Sherick Hughes et al., Causation Fallacy
2.0: Revisiting the Myth and Math of Affirmative Action, 30 Educ. Pol’y 63, 81 (2016); Kimberly West-Faulcon,
Obscuring Asian Penalty with Illusions of Black Bonus, 64 UCLA L. Rev. 590, 600 (2017); Dkt.415, Ex.1 at 112.
34
Goodwin Liu, The Causation Fallacy: Bakke and the Basic Arithmetic of Selective Admissions, 100 Mich. L. Rev.
1045, 1095 (2002).
35
See, e.g., Sherick Hughes et al., Causation Fallacy 2.0: Revisiting the Myth and Math of Affirmative Action, 30
Educ. Pol’y 63 (2016); Kimberly West-Faulcon, Obscuring Asian Penalty with Illusions of Black Bonus 64 UCLA
L. Rev. Disc. 590 (2017).
36
Sherick Hughes et al., Causation Fallacy 2.0: Revisiting the Myth and Math of Affirmative Action, 30 Educ. Pol’y
63, 82 (2016).
37
Dkt. 419, Ex. 33 at 110.
38
See Dkt. 420 at 1-2.
39
HARV0023177.

14

increase from 400 to 460 and correspondingly increase the selection rate of admitted Asian
American applicants from 5.1% to 5.9%.
This impact is small and also entirely consistent with constitutional principles which
permit a flexible “plus” for racial minorities to achieve sufficient representation on campus.
Established Supreme Court precedent recognizes that race—like any other attribute valued by a
university—may tip the scale in favor of admission among qualified candidates, but this does not
render the policy unconstitutional. See, e.g., Bakke, 438 U.S. at 317-18. Narrow tailoring is
especially evident when, as here, the number of impacted seats is notably small. As Fisher II
reflected: “The fact that race consciousness played a role in only a small portion of admissions
decisions should be a hallmark of narrow tailoring, not evidence of unconstitutionality.” Fisher
II, 136 S. Ct. at 2212.
D.

Harvard’s race-conscious admissions policy is narrowly tailored because it
does not afford Black and Hispanic students an undue preference.

The constitutional rationale for implementing a race-conscious admissions program is to
achieve “the educational benefits that flow from student body diversity.” Fisher II, 136 S. Ct.
at 2210.

As such, an effective program will necessarily increase ethno-racial diversity on

campus. The question is not whether race has an effect in some admissions decisions. It will.
Rather, the question is whether race is the “predominant” or “defining” feature of an application.
Grutter, 538 U.S. at 320, 337. Here, it is not. Beyond the fact that Harvard’s consideration of
race for ethno-racial minorities minimally affects Asian Americans’ selection rate (by one
percentage point or less), those who do receive a “tip” from race-conscious admissions do not
receive an undue preference.
The data shows that Harvard is not admitting African American and Hispanic applicants
solely because of their race. Rather, the data shows Harvard is admitting highly qualified

15

Hispanic and African American students, and in some cases, giving them a nudge because of
their race.

Harvard’s expert Dr. Card analyzed the probability of admissions for African

American and other underrepresented minority students (including Hispanic applicants, but
excluding Asian Americans). Ranking such students by probability of admission (considering all
relevant factors, not just academics), Dr. Card considered how these underrepresented applicants
compared to white candidates with similar qualifications. (Dkt. 419, Ex. 33 at 84.) His analysis
showed that for eight of the ten deciles, the positive effect of race was minimal. (Dkt. 419, Ex.
33 at 86.) It was only in the top two most competitive deciles that race had a notable positive
effect on admissions for African American or other minority students. (Dkt. 419, Ex. 33 at 8384.) He further found that for 94% of the African American candidates and 96% of the other
underrepresented minority candidates, admissions decisions were more attributable to
“unobserved characteristics” (i.e. unquantifiable factors) than to race. (Id. at 85-86.) In other
words, race mattered less than other factors in the admissions process. SFFA’s claim that race is
the predominant factor in admissions is clearly refuted by the record. 40
II.

Any negative effect for Asian Americans observed in the admissions process is
caused by admissions practices that benefit white applicants and not race-conscious
admissions.
A.

The relevant comparison for an intentional discrimination claim is between
Asian Americans and similarly situated white applicants.

As discussed above, the Supreme Court has permitted universities to consider race as a
“plus factor” so long as the process is narrowly tailored to harness the benefits of student body
diversity.

See, e.g., Grutter, 539 U.S. at 334.

40

Accordingly, there is nothing inherently

Students note that SFFA engages in a series of dubious race “changing” exercises to suggest Asian American
applicants would have a higher rate of admissions if treated like Black or Hispanic applicants. See, e.g., Dkt. 413 at
46-47. Students restate the arguments made in their July brief that this analysis misconstrues the proper legal
standard and is unreliable due to the significant limitations of Dr. Arcidiacono’s modeling decisions. See generally
Dkt. 435 at 10-15.

16

unconstitutional at play if a university admits a highly qualified, but lesser-represented minority
student—commonly Black, Latino, and Native American students—over a similarly situated
Asian American student. See id.; Bakke, 438 U.S. 265, 317–18. The constitutionality of such an
action must be evaluated under the narrow tailoring standard set by Bakke, Grutter, and Fisher.
However, the Supreme Court has not permitted universities to give a relative “plus” to
white applicants over similarly situated non-white applicants, based on race. The reason is
simple: at today’s elite institutions, giving white students a “plus” based on race would not
further racial diversity.41 White students are also the most relevant comparison for exploring
whether a university’s policy intentionally acts as an admission “ceiling.” The logic is once
again straightforward: at top universities such as Harvard, white students comprise the
overwhelming majority of applicants.42 Correspondingly, the proportion of ethno-racial minority
students is notably small.43 As such, any intentional scheme to effectively limit the number of
Asian Americans on campus would predominantly preserve more seats for white students. For
all these reasons, the relevant comparison for the intentional discrimination claim is between
Asian Americans and white students, not other ethno-racial minority students.44

41

For example, a recent New York times analysis revealed that white students continue to make up the
overwhelming plural majority of all eight ivy league schools. For example, the data shows that in 2015 white
students comprised the following shares of the incoming freshman class: 52% at Brown University, 40% at
Columbia University, 45% at Cornell University, 56% at Dartmouth University, 47% at Harvard University, 49% at
Princeton University, 50% at University of Pennsylvania, and 51% at Yale University. The 2017 article also noted
that “black and Hispanic students are more underrepresented at the nation’s top colleges and universities than they
were 35 years ago.” Indeed, the data showed that in 2015, the Black share of the incoming freshman class remained
below 10% at all eight school and the Hispanic share remained below 20% despite their increasing share of the
college-age population. Jeremy Ashkenas, Haeyoun Park & Adam Pearce, Even With Affirmative Action, Blacks and
Hispanics Are More Underrepresented at Top Colleges Than 35 Years Ago, N.Y. Times (August 24, 2017),
https://www.nytimes.com/interactive/2017/08/24/us/affirmative-action.html.
42
See Jeremy Ashkenas, Haeyoun Park & Adam Pearce, supra; see also Dkt. 415, Ex. 1 at 112 (showing that over
six years of applicant data, white applicants are the overwhelming plurality comprising 40% of the applicant pool)
43
Dkt. 415, Ex. 1 at 112 (showing that over six years of applicant data, Black applicants make up approximately
11% of the pool and Hispanic applicants make up 12% of the pool).
44
See also generally Jerry Kang, Negative Action Against Asian Americans: The Internal Instability of Dworkin’s
Defense of Affirmative Action, 31 Harv. C.R.-C.L. L. Rev. 1, 3 (1996) (explaining that claims of discrimination

17

SFFA styles its motion to confuse the relevant racial comparison by presenting data
across all racial groups together.45 Such confusion is heightened by SFFA’s inflammatory and
grossly misleading suggestion that Harvard expresses favoritism towards Black, Latino, and
Native American students over Asian-American students.46 All of Harvard’s non-white students
fight tirelessly to have their voices heard and their needs met on Harvard’s campus. (See, e.g.,
Dkt. 440-1, Ex. 1.10 at ¶ 7. (Decl. of Sarah Cole).) Any step forward made by minority students
results from longstanding efforts, not favoritism. (Id., Ex. 1.8 at ¶ 10 (Decl. of T.D.); Ex. 1.12,
at ¶¶ 14-15 (Decl. of Vasquez-Rodriguez .)) Moreover, gains made by one ethno-racial group
often extend to others due to minority students’ coalitional work and intersectional interests. (Id.;
see also id., Ex. 1.4 at ¶ 8 ( Decl. of J.L.); Ex. 1.8 at ¶ 8 (Decl. of A.Z.).) But SFFA’s passing
references to other ethno-racial minorities simply serves its divisive strategy of pitting
marginalized groups against each other. It bears no relevance on its intentional discrimination
claim.
Even SFFA admits that white applicants are the only pertinent benchmark for its
intentional discrimination claim.

SFFA opens its summary judgment motion by alleging

“Harvard’s admissions policy has a disproportionately negative effect on Asian Americans visa-vis similarly-situated white applicants that cannot be explained on non-discriminatory
grounds.” (Dkt. No. 413 at 7) (emphasis added.) SFFA’s calculation of the “Asian American
penalty” similarly recognizes that whites are the relevant comparison group. SFFA’s expert Dr.

against Asian Americans in university admissions should “us[e] the treatment of Whites as a basis for comparison . .
. . To be clear, Whites, not any other race, are used as the baseline.”)
45
See, e.g., Dkt. 413 at 14 (comparing personal scores of Asian Americans across the academic deciles as compared
to white, Hispanic, and African American applicants).
46
For example, SFFA makes the bald assertion that “Harvard’s reaction to claims of discrimination . . . against
Asian Americans contrasts starkly with how it responds to complaints from other minority groups. When Native
Americans raised concerns about their representation on campus, they were taken seriously. As were the concerns
that Latinos and Latinas [and African Americans] raised. . . .”. Dkt. 413 at 27.

18

Arcidiacono “remov[ed] penalties against Asian-American applicants” by calculating admissions
outcomes based on if Harvard “had treated Asian-American applicants identically to their
white counterparts.” (Dkt. 414 at 143) (emphasis added.) While Students disagree with SFFA
on many other points, Students agree that evaluating whether a discriminatory “penalty” exists
requires focusing on the treatment of Asian Americans vis a vis whites.
B.

SFFA must prove that Harvard’s current holistic process has a
discriminatory effect, is motivated by a discriminatory purpose, and that
there is some connection between the discriminatory motive and challenged
policy.

SFFA’s motion for summary judgment makes clear that its intentional discrimination
claim does not challenge Harvard’s race-conscious policy, per se, but rather attacks Harvard’s
holistic admissions process in its entirety. (See Dkt. 413 at 5-6.) Since the policy does not
discriminate against Asian Americans “on its face” SFFA contends that the governing
framework is laid out by Village of Arlington Heights v. Metropolitan Housing Development
Corp., 429 U.S. 252, 265-68 (1977).

However, SFFA fails to follow the Arlington Heights

standard and its proffered evidence falls far short of satisfying it.47
SFFA claims that it has met its burden of showing intentional discrimination primarily
through its allegation of discriminatory effect.

(Dkt. 413 at 12.)

The legal standard for

evaluating intentional discrimination claims, however, requires a showing of both discriminatory
effect and discriminatory purpose. Arlington Heights, 429 U.S. 252, 264-65 (1977). The burden
is on the plaintiffs to show that “discriminatory purpose was a motivating factor” in a policy.
Arlington Heights, 429 U.S. at 270-71. This requires demonstrating more than mere awareness.
“[Intent] means more than mere knowledge by the [defendant] that a policy has a discriminatory

47

Students do not agree that this is the appropriate standard for analyzing any differential treatment vis a vis other
ethno-racial minorities, which should be reviewed under the narrow tailoring standard discussed above.

19

effect; the [defendant’s] agent must have adopted the policy because of, not despite, its
discriminatory impact.” McGuire v. Reilly, 386 F.3d 45, 63 (1st Cir. 2004) (internal citations
omitted) (emphasis added).
Under Arlington Heights, the court evaluates whether plaintiffs have carried their burden
of proving discriminatory purpose and effect by considering various factors. “The impact of the
official action [and] whether it bears more heavily on one race than another . . . may provide an
important starting point.” Arlington Heights, 429 U.S. at 266 (internal citations and quotation
marks omitted). Other relevant factors include: “[t]he historical background of the decision . . . ,
particularly if it reveals a series of official actions taken for invidious purposes;” “[d]epartures
from the normal procedural sequence;” “[s]ubstantive departures [from the norm] . . .” and “the
legislative or administrative history . . . , especially where there are contemporary statements by
members of the decisionmaking body, minutes of its meetings, or reports.” Id. at 267-68.
C.

Applying the Arlington Heights framework reveals that SFFA is not entitled
to summary judgment on its intentional discrimination claim.
1.

There are genuine disputes over whether Harvard has any policies
advantaging white students over Asian Americans that have a
discriminatory effect.

Harvard’s Opposition identifies numerous material disputes with regard to SFFA’s proof
of discriminatory effect. (Dkt. 435 at 5-27.)

Students do not repeat Harvard’s criticisms here

but agree that Dr. Arcidiacono’s modeling choices and data manipulations are far too suspect to
warrant summary judgment in SFFA’s favor. In particular, Dr. Arcidiacono’s choice to remove
all “special category” (ALDC) applicants from his data set, rather than controlling for such
characteristics, reflects a deeply flawed approach which renders the conclusions that he draws
unreliable. (Dkt. 435 at 11-13.) He also erroneously combines all six years of data into one data
set instead of analyzing each year separately. (Dkt. 435 at 13-14, n. 3).

20

These issues aside, examining differences in selection rates offers a starting point for
determining whether Harvard’s admissions system has a disparate effect on Asian Americans as
compared to whites. This is what is done in employment discrimination cases.48 The four-fifths
test is one measure that is used to determine whether the discrepancy is substantial. The First
Circuit has described the four-fifths test as a “helpful benchmark,” that should not be treated as
decisive. Jones v. City of Boston, 752 F.3d 38, 51 (2014). The benefit of the four-fifths test is
that it is easy to calculate: the selection rate of the group claiming disparate impact is divided by
the selection rate of the group that is allegedly favored. A result that falls below four-fifths (i.e.
less than 0.8) serves as some evidence of disparate impact. Id. at 49.

One scholar studying

college admissions has suggested that the four-fifths test could be used to evaluate claims of a
disparate effect between Asian American and white admissions.49
Based on Dr. Arcidiacono’s analysis, SFFA has claimed “Harvard’s admissions system
discriminates against Asian American applicants in at least three respects”: (i) the personal
rating, (ii) the overall score and (iii) admissions decisions.

(See Dkt. No. 413 at 13-16.)

Applying the four-fifths test to Dr. Arcidiacono’s preferred dataset yields the following results:

48

See e.g., Fudge v. City of Providence Fire Dept.,766 F.2d 650, 657 (1st Cir. 1985) (“[I]n a case involving a claim
that a screening test for admission to employment imposes a disparate and adverse impact on blacks, the initial
inquiry must be whether there is a discrepancy in the rate of passage as between whites and blacks. If so, an intuitive
judicial judgment must next be made whether the discrepancy is substantial.”)
49
Kimberly West-Faulcon, Obscuring Asian Penalty with Illusions of Black Bonus, 64 UCLA L. Rev. Disc. 590,
615-16 (2017).

21

Personal rating by admissions
officers: share receiving a two
or higher50
Overall rating by admissions
officers: share receiving a two
or higher51
Admissions rate52

Asian American (Dr.
Arcidiacono’s
dataset excludes
(ALDC) applicants)
17.65%

White (Dr.
Ratio
Arcidiacono’s
dataset excludes
(ALDC) applicants)
21.29%
0.83

4/5 test
violated?

4.85%

4.44%

1.09

No

4.0%

4.2%

0.95

No

No

Notably, none of these rates violate the four-fifths test. Thus, the differentials are not
large enough to suggest racial unfairness under traditional disparate impact analysis. Students’
use of Dr. Arcidiacono’s data should not be interpreted as an endorsement of his method, which
Students find highly problematic. Rather, Students use Dr. Arcidiacono’s data to illustrate that
SFFA is not entitled to summary judgment because its own data tables are open to divergent
interpretations which should be resolved at trial.
Moreover, Students highlight another aspect of the data which neither party raises: when
“special category” (ALDC) applicants are included—athletes, legacy applicants, dean/director’s
list, children of faculty/staff—the gap significantly widens between Asian Americans’ and
whites’ admissions rates.

Applying the four-fifths test to the admissions rates53 of Asian

Americans and whites across the entire pool of applicants for six years yields the following:

50

Dkt. 414 at 127.
Dkt. 414 at 130.
52
Dkt. 419, Ex. 31 at 35. Students note that Dr. Arcidiacono offers inconsistent calculations of the average
admissions rate for non-ALDC applicants. His report lists the admissions rates referenced in the table above (4.0%
for non-ALDC Asian Americans, and 4.2% for non-ALDC whites). But elsewhere, Dr. Arcidiacono’s analysis
reflects a different “average” admissions rate for Asian American and white students, appearing as 5.14% and 4.9%
respectively (resulting in a 4/5 ratio of 1.04, similarly presenting no violation of the 4/5 test). Dkt. 414 at 134.
Lacking the underlying data, Students are unable to determine what factors explain how Dr. Arcidiacono arrived at
different results. But these inconsistencies in Dr. Arcidiacono’s rates of admission provides further support for
denying summary judgment so that factual disputes can be properly resolved at trial.
53
Students only had access to the raw admissions numbers of the full applicant pool over six years, disaggregated by
race. Dkt. 415, Ex. 1 at 112. This allowed them to calculate the admissions rates by racial group. But Students did
51

22

Asian American applicants
(all applicants,
6 years of data)
Admissions rate54

5.9%

White
applicants
(all applicants,
6 years of data)
7.9%

Ratio

4/5 test
violated?

0.74%

Yes

The fact that a violation of the four-fifths test appears once “ALDC” applicants are
included strongly suggests that it is preferences given to “ALDC” applicants that cause any
observed disadvantage for Asian Americans in the admissions process—not race-conscious
admissions. Harvard’s internal OIR report made a similar observation when it noted that:
“athletes and legacies explain the difference in raw admit rates for Asian and White applicants.”
(Dkt. 421, Ex. 145 at 3).
The raw numbers help explain why any disadvantage faced by Asian Americans face is
most attributable to “ALDC” preferences. Whites overwhelmingly outnumber any other ethnic
group in these “ALDC” categories.

According to Dr. Arcidiacono’s table, the admission

numbers over a six year period are as follows55:

Athlete
Legacy
Faculty child
Staff child
Dean / Director's List
Total Admits

White
(admits)
817
1080
33
47
701
5,020

Asian American
(admits)
101
163
13
26
133
2,459

Black
(admits)
124
67
0
2
29
1,400

Hispanic
(admits)
54
90
2
6
59
1,293

Total
(admits)
1179
1541
60
89
1034
11,068

not have access to raw data on the personal and overall scores of the full applicant pool over six years, disaggregated
by race. As such, Students could not perform a similar 4/5 test for personal and overall scores across the full
applicant pool.
54
Dkt. 415, Ex. 1 at 112. Admissions rates were calculated by dividing the total number of admitted applicants for a
given racial group by the total number of applicants for a given racial group. This ratio converted to a percentage by
multiplying it by 100. For example, for Asian Americans the following calculation was performed: (2459 [total
Asian American admits] ÷ 41369 [total Asian American applicants]) x 100.
55
Dkt. 415, Ex. 1 at 112.

23

The sheer number of white admission offers associated with these special categories
(approximately 2,67756) roughly equals all admissions offers given to Asian Americans (2,45957)
and also roughly equals all admissions offers given to Black and Hispanic students combined
(1,400 and 1,293 respectively58). These numbers reflect how any displacement faced by Asian
Americans is much more likely attributable to “ALDC” applicants than ethno-racial minorities.
Even ignoring the impact of “ALDC” preferences, SFFA fastens onto the incorrect
culprit for any observed bias against Asian Americans: Harvard’s race-conscious policy
promoting diversity. But the evidence cannot support such a proposition. Indeed, SFFA’s own
expert Dr. Arcidiacono explained that removing the alleged “Asian American penalty” would
most substantially reduce admissions offers given to whites in terms of raw numbers, not Black
and Hispanic applicants. (Dkt. 419-35 at 122.) Specifically, 147 fewer white applicants would
be admitted over a six-year period as opposed to 42 and 51 fewer admission offers to Blacks and
Hispanics respectively. (Id.) A penalty that greatly advantages white applicants hardly seems
tethered to a policy designed to increase racial diversity and the presence of underrepresented
ethno-racial minorities.
2.

SFFA has not demonstrated that any disparate effect in Asian
American admission rates comes from a discriminatory purpose as set
forth in the Arlington Heights factors.

SFFA made no serious effort to develop the other Arlington Heights factors, other than
historical background of the decision, and it fails to show how that historical background relates
to the challenged policy. Arlington Heights teaches that “the historical background of the

56

Dkt. 415, Ex. 1 at 112. This calculation may include duplicate applicants. For example, if an applicant qualifies
as both a legacy and an athlete, they would appear twice in this calculation. Students did not have access to the raw
data files and so were limited to using preexisting tables in the parties’ expert reports.
57
Dkt. 415, Ex. 1 at 112.
58
Dkt. 415, Ex. 1 at 112.

24

attacked decision,” 429 U.S. at 267, can shed light on whether a defendant harbored
discriminatory animus. The policy attacked here—Harvard’s holistic process that considers race
to promote diversity—has no historical connection to the history discussed at length in SFFA’s
motion: Harvard’s 1920s admissions policy that limited Jewish students and other racial
minorities. (Dkt. 413 at 29-32; Dkt. 421, Ex. 240 at 3.)
President Lowell’s exclusionary practices are historically distinct and diametrically
opposed to Harvard’s current diversity-driven admissions policy. As a basic matter, it is not
clear to Students how discrimination against Jews in the 1920s proves discrimination against
Asian Americans today, a different racial group and different era. But more importantly, the
current holistic policy (namely, “the diversity rationale”) grew out of the Civil Rights Movement
in the 1960-1970s, not the 1920s.59 This admissions policy has been driven by a motivation to
create “a greater degree of openness and inclusiveness” at Harvard (Dkt. 419, Ex. 41 at 35), and
by a recognition that “Harvard remained, for three centuries, committed primarily to educating
the sons of New England’s elite.” (Dkt. 419, Ex. 44 at 8.) Harvard’s current admissions process
is grounded in the belief that students’ different backgrounds and personal traits, including their
ethno-racial identities, enhances their contributions and achievements and engenders a more
diverse and vibrant academic community to the benefit of all students. (Dkt. 419, Ex. 44 at 9.)
Unlike during the era under Lowell when there was abundant documentation of racial exclusion,
there is no evidence that any Harvard employees seek to limit the number of Asian American
students on campus. Rather, the evidence produced suggests that Harvard admissions officers
view applicants’ ethno-racial identities as possibly enhancing their contributions and
achievements, including those of Asian-American heritage. (See Dkt. 435 at 35.)
59

See Dkt. 419, Ex. 41 at 42; Brief of Columbia University, Harvard University, Stanford University and the
University of Pennsylvania as Amici Curiae, Bakke, 438 U.S. 265 (1978) (No. 76-811), 1976 WL 181278, **3-4.

25

It is plain that Harvard’s current inclusionary policy—which views ethno-racial diversity
as an asset not a hindrance—does not resemble Lowell’s exclusionary practices. As a legal
matter, this historical disconnect nullifies the relevance of the history discussed by SFFA. Where
there has been a clear break in the historical context of a challenged program, a court will not
infer that the new program was adopted with a discriminatory intent. See, e.g., Anderson ex rel.
Dowd v. City of Boston, 375 F.3d 71, 83 (1st Cir. 2004). The Supreme Court recently reaffirmed
that “[t]he allocation of the burden of proof and the presumption of . . . good faith are not
changed by a finding of past discrimination. ‘[P]ast discrimination cannot, in the manner of
original sin, condemn . . . action that is not itself unlawful.’ The ‘ultimate question remains
whether a discriminatory intent has been proved in a given case.’” Abbott v. Perez, 138 S. Ct.
2305, 2324-2325 (2018) (quoting Mobile v. Bolden, 446 U.S. 55, 74 (1980)) (citation omitted).
SFFA has failed to show that Harvard’s race-conscious admissions policy is motivated by an
intent to discriminate against Asian Americans.
If anything, the ethno-racial exclusion under President Lowell is further reason to
preserve holistic, race-conscious admissions, not ban it. Of note, President Lowell’s antagonism
towards ethno-racial minorities was not limited to Jews. He also fostered prejudice against
African Americans and recent immigrants.60 (Dkt. 421, Ex. 240 at 3.) Harvard’s exclusionary
history erects even greater barriers to cultivating a welcoming, inclusive campus. Race-conscious
admissions can assist in overcoming this obstacle by increasing the number of underrepresented
students on campus and signaling a value for racial diversity and inclusion. For the reasons
stated above, SFFA is not entitled to summary judgment on its intentional discrimination claim.

60

See generally Randall Kennedy, Blacks and the Race Question at Harvard in Blacks at Harvard: A Documentary
History of African-American Experience at Harvard and Radcliffe (Werner Sollors et al., eds. 1993).

26

D.

A finding of intentional discrimination against Asian Americans would not
justify banning race-conscious admissions since such a ban would not
remediate the harm caused by a white admissions advantage.

Even if SFFA were to ultimately prevail on its intentional discrimination claim, SFFA
still would not be entitled to the remedy it seeks which is drastic and inapposite. Specifically,
SFFA asks this Court to “prohibit[] Harvard from using race as a factor in future undergraduate
admissions decisions” and require “Harvard to conduct all admissions in a manner that does not
permit those engaged in the decisional process to be aware of or learn the race or ethnicity of any
applicant for admission.” (Dkt. No. 1 at 119.) The factual record does not justify this remedy
and prevailing law would not permit it.
When fashioning a remedy, courts are guided by general equitable principles centered on
three primary considerations: the nature and scope of the violation, restoring the victims of
discriminatory conduct to the position they would have occupied in the absence of such conduct,
and reconciling public and private needs. Milliken v. Bradley, 433 U.S. 267, 280–81, 288 (1977).
“[F]ederal court decrees exceed appropriate limits if they are aimed at eliminating a condition
that does not violate the Constitution or does not flow from such a violation.” Id. at 282 (1977).
An admissions practice that advantages white applicants clearly “does not flow” from a policy
designed to increase the admissions chances of ethno-racial minorities.

Stated simply, the

“nature and scope of the violation” would not warrant banning race-conscious admissions.
It is clear that SFFA’s proposed remedy would not place Asian Americans in the position
they would have been absent the illegal conduct. As a practical matter, a white advantage such
as legacy preferences could persist irrespective of race-conscious admissions. Moreover, experts
in the field of countering racial bias advocate for fostering greater self-awareness and reflection

27

around race, not “blindness” to it in world where racial inequities are pervasive.61 As multiple
researchers have found, “[i]nterventions that rely on ‘color-blind’ strategies, in which people are
encouraged to suppress their category-based stereotypes in favor of more personalized
judgments, appear to be particularly ineffective.”62 In some cases, researchers have found that a
forced “color-blind” strategy may even backfire.63
In further contravention of equitable principles, such a remedy would fail to reconcile
public and private needs in at least three respects. First, such a ban carries a substantial threat of
exacerbating racial isolation and suppressing the benefits of a diverse campus.64

Second,

SFFA’s proposed injunction poses significant practical barriers that would render meaningful
individualized review impossible, potentially preventing Harvard from: conducting interviews,
recruiting applicants in person, and viewing drama, music, or other performance on video or live.
It would censor student expression by jeopardizing students’ ability to: submit essays and
recommendations discussing how race or ethnicity has impacted their lives; list awards and
activities indicating race or ethnicity; or write about their immigrant stories regardless of their
61

See Jennifer A. Richeson & Richard J. Nussbaum, The Impact of Multiculturalism Versus Color-Blindness on
Racial Bias, 40 J. Experimental Soc. Psychol. 417, 421 (2004) (“[E]xplicit racial bias was relatively greater after
exposure to the color-blind perspective on interethnic relations, compared to the multicultural perspective….”);
Phoebe C. Ellsworth & Samuel R. Sommers, How Much Do We Really Know About Race and Juries? A Review of
Social Science Theory and Research, 78 Chicago-Kent L. Rev. 997, 1010-1016 (2003); Adam R. Pearson et al., The
Nature of Contemporary Prejudice: Insights from Aversive Racism, Soc. & Personality Psychol. Compass, no. 3,
2009, at 13.(“[E]fforts to be colorblind can sometimes produce ‘rebound effects,’ causing biases to become activated
even more.”).
62
Laurie A. Rudman et al., “Unlearning” Automatic Biases: The Malleability of Implicit Prejudice and Stereotypes,
81 J. Personality & Soc. Psychol. 856, 857 (2001) (citing Christopher Wolsko et al., Framing Interethnic Ideology:
Effects of Multicultural and Color-Blind Perspectives on Judgments of Groups and Individuals, 78 J. Personality &
Soc. Psychol. 635 (2000)).
63
Rudman, supra, at 857 (citing J.W. Schofield, Causes and Consequences of the Colorblind Perspective in
Prejudice, Discrimination, and Racism 231-254 (J.F. Dovidio & S.L. Gaertner, eds., 1986)).
64
See Dkt. 440 at 17-23, 29-38; see also, e.g., Dkt. 435 at 47-48; Gary Orfield et al., Alternative Paths to Diversity:
Exploring and Implementing Effective College Admissions Policies (2017) (collection of ETS research papers);
William C. Kidder, How Workable Are Class-Based and Race-Neutral Alternatives at Leading American
Universities?, 64 UCLA L. Rev. Disc. 100, 121-26 (2016) (synthesizing numerous studies and documents and
stating that SFFA’s expert Dr. Richard Kahlenberg, makes empirically unfounded and misleading claims about race
neutral alternatives); William Kidder & Angela Onwuachi-Willig, Still Hazy After All These Years: The Data and
Theory Behind Mismatch, 92 Tex. L. Rev. 895, 930-34 (2014).

28

country of origin. Third, and relatedly, purging all references to race would ignore a factor that
is often inextricable from an applicant’s formative life experiences, perversely penalizing some
applicants in the name of equal protection (especially non-white applicants who
disproportionately face racial barriers).65 As A.A., who identifies as Asian American and queer,
aptly expressed: “the application reflected who I am as a person, and part of that includes my
ethnic and racial identity.” (Dkt. 440-1, Ex. 1.1 at ¶ 8 (Decl. of A.A.).)
Equitable considerations plainly preclude banning race-conscious admissions. However,
the same legal principles and research can provide insight into appropriate remedies if a trial
does uncover discrimination against Asian Americans vis a vis white applicants.

Such tested

strategies for countering racial bias include: conducting trainings around diversity and the
psychological processes in decision making;66 engaging in effortful, deliberative processing
which prevents snap judgments;67 and instituting feedback mechanisms and accountability
measures, such as notetaking and tracking metrics.68 Harvard engages in many of these practices
already,69 but safeguards could be specifically tailored to address any undue bias faced by Asian
Americans.70 No matter how the remedy is fashioned, SFFA’s intentional discrimination claim

65

See Dkt. 440 at 24-27.
Patricia G. Device et al., A Gender Bias Habit-Breaking Intervention Led to Increased Hiring of Female Faculty
in STEMM Departments (2017), https://psyarxiv.com/tdvy7. For more research testing effective trainings, see Sylvia
P. Perry et al., Modern Prejudice: Subtle, But Unconscious? The Role of Bias Awareness in Whites’ Perceptions of
Personal and Others’ Biases, 61 J. Experimental Soc. Psychol. 64 (2015).
67
Geoffrey Beattie et al., An Exploration of Possible Unconscious Ethnic Biases in Higher Education: The Role of
Implicit Attitudes on Selection for University Posts, 197 Semiotica 171, 195-96 (2013). See also Daniel Kahneman,
Thinking, Fast and Slow (2011).
68
See Faye Crosby et al., Cognitive Biases in the Perception of Discrimination: The Importance of Format, 14 Sex
Roles 637 (1986).
69
See, e.g., Ellsworth Ex. 127 (training presentation urging admissions officers to “be wary of aggregated data for
Asian American student populations” and to “[h]onor the nuance of both identity and context”).
70
For example, trainings could be developed around common stereotypes harbored towards Asian Americans which
differ in distinct ways from those of other ethno-racial minorities; data feedback loops could be implemented to
track the relative personal scores of Asian Americans vis a vis whites throughout the admissions process making the
Committee more sensitive to notable trends; and accountability pressures could be applied by conducting random
66

29

definitively does not warrant ending race-conscious admissions. Such an injunction would be
both factually illogical and legally impermissible.
CONCLUSION
As a broad cross-section of ethno-racial minorities who are prospective, current, and
former students of Harvard, Students firmly assert that our elite institutions cannot be colorblind
in a world full of color, nor may they be indifferent to a world full of difference.

The

constitution recognizes this need, permitting race-conscious admissions when certain
requirements are met. Here, Harvard meets such requirements for narrow tailoring and certainly
has placed SFFA’s facts in dispute. For the foregoing reasons and those articulated in Students’
July brief, this Court should affirm the constitutionality of Harvard’s race-conscious admissions
policy that promotes diversity. Even if trial is appropriate for SFFA’s claim of intentional
discrimination given the factual disputes in the record and indication that preferences for legacy
applicants impact Asian Americans in admission, any such claim must be limited to a
comparison of Asian American and white applicants.
Respectfully Submitted,
/s/ Oren M. Sellstrom
Oren M. Sellstrom (BBO #569045)
LAWYERS’ COMMITTEE FOR CIVIL RIGHTS
AND ECONOMIC JUSTICE
61 Batterymarch Street, Fifth Floor
Boston, MA 02110
Tel: [phone]
[e-mail]
/s/ Genevieve Bonadies Torres
Genevieve Bonadies Torres (pro hac vice)
Kristen Clarke
audits of admissions officers’ files to identify any existing biases and provide feedback. These are just a few
examples drawn from prevailing research on best practices.

30

Jon M. Greenbaum (pro hac vice)
Brenda Shum (pro hac vice)
LAWYERS’ COMMITTEE FOR CIVIL RIGHTS
UNDER LAW
1500 K Street, NW
Washington, DC 20005
Tel: [phone]
[e-mail]
/s/ Nicole K. Ochi
Nicole K. Ochi (pro hac vice)
ASIAN AMERICANS ADVANCING JUSTICE
1145 Wilshire Boulevards
Los Angeles, CA 90017
Tel: [phone]
[e-mail]
/s/ Lawrence Culleen
Lawrence Culleen (pro hac vice)
Nancy Perkins (pro hac vice)
Steven Mayer (pro hac vice)
Emma Dinan (pro hac vice)
ARNOLD & PORTER KAYE SCHOLER LLP
601 Massachusetts Ave., NW
Washington, DC 20001
Tel: [phone]
[e-mail]

Dated: August 30, 2018

COUNSEL FOR AMICI CURIAE

31

CERTIFICATE OF SERVICE
In accordance with Local Rule 5.2(b), I hereby certify that this document filed through
the ECF system on August 30, 2018 will be sent electronically to the registered participants as
identified on the Notice of Electronic Filing.
/s/ Lawrence Culleen
Lawrence Culleen

32