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

Filing 92

MOTION for Leave to File Reply Memorandum in Support of Motion to Compel by Students for Fair Admissions, Inc.. (Attachments: # 1 Proposed Reply Memo, # 2 Declaration of Patrick Strawbridge)(Caldwell, Benjamin)

UNITED STATES DISTRICT COURT
DISTRICT OF MASSACHUSETTS
BOSTON DIVISION
STUDENTS FOR FAIR ADMISSIONS, INC.
Plaintiff,
v.
PRESIDENT AND FELLOWS OF HARVARD
COLLEGE (HARVARD CORPORATION)
Defendant.

)
)
)
)
)
)
)
)
)
)
)

Civil Action No: 1:14-cv-14176

PROPOSED REPLY MEMORANDUM IN SUPPORT OF MOTION TO COMPEL
Harvard’s Opposition to SFFA’s Motion to Compel reveals its deep confusion about the
legal rules governing this case. Harvard has elected to use race as part of its admissions process.
Harvard must therefore accept the burden of proving that it can satisfy strict scrutiny. The
Supreme Court has made crystal clear its expectations about this analysis: “[s]trict scrutiny does
not permit a court to accept a school’s assertion that its admissions process uses race in a
permissible way without a court giving close analysis to the evidence of how the process works
in practice.” Fisher v. Univ. of Texas at Austin, 133 S. Ct. 2411, 2421 (2013) (emphasis added).
Indeed, this Court is obligated to ensure the record contains the evidence necessary to make a
“judicial determination that the admissions process meets strict scrutiny in its implementation.”
Id. at 2419-20 (emphasis added). Faithful application of strict scrutiny is essential to incentivize
universities “to make the existing minority admissions schemes transparent and protective of
individual review.” Grutter v. Bollinger, 539 U.S. 306, 394 (Kennedy, J., dissenting).
With these constitutional requirements in mind, Harvard’s Opposition should be seen for
what it is: an attempt to misdirect the parties and the Court, to avoid its obligations under the

1

Federal Rules of Civil Procedure, and to obstruct SFFA and this Court from obtaining the
discovery necessary to conduct the required “searching review” of how the admissions process
“works in practice.” More than five months after it answered the Complaint, Harvard continues
to delay its production of any responsive materials, essentially assuming for itself the power to
grant a stay. It continues to withhold from SFFA and the Court relevant information bearing
upon this motion. And it seeks to insulate important parts of its admissions process from any
review at all. Respectfully, this Court should grant SFFA’s motion, as modified below:
First, with respect to SFFA’s request that Harvard abide by its prior commitment to
produce a statistically significant sample of application files, Harvard does not challenge the
statistical and legal basis for SFFA’s proposed sample. Instead, Harvard blithely reneges on its
offer, asserting that sampling is unnecessary because SFFA’s expert can conduct his statistical
analysis with Harvard’s electronic admissions database. Harvard supports this view with a
declaration of its own expert, who reviewed “one year” of the admissions database, and
pronounced himself satisfied. But Harvard has refused to disclose the actual contents of its
admissions database to SFFA (or the Court)—despite its prior agreement that its database was
relevant and would be produced, and despite SFFA’s specific request for more information in
order to determine whether it may be able to adjourn or reduce its request for files. Essentially,
Harvard has said that SFFA and this Court should trust its expert’s opinion that all of the
information needed is in the database, but it refuses to provide even the most basic information
about what that database contains.
SFFA is willing to adjourn its motion to compel production of the application files if
Harvard promptly produces the databases responsive to its Request for Production No. 1. See
Strawbridge Decl. in Support of Opening Memo., Ex. C at 11-12. This would permit SFFA and

2

its expert to review the data and determine whether the database material alone is sufficient to
conduct its analysis. To be clear, SFFA ultimately will require a statistically significant sample
of files—and certainly Harvard has not disagreed with the obvious proposition that SFFA cannot
be denied access to materials which Harvard intends to use to defend its admissions process. See
Opening Memo. at 16-17. But it makes little sense for SFFA and the Court to guess about the
completeness and sufficiency of the database that Harvard has shown to its expert but withheld
from SFFA. Instead, Harvard should produce that data promptly and permit SFFA to determine
if its request for a sample of files can be modified.1
Second, Harvard has failed to justify its intention to withhold the identities of its alumni
interviewers in this litigation. Harvard’s attempt to downplay the significance of these potential
witnesses is refuted by both public statements emphasizing the importance of the interview
process, and by Ms. McGrath’s prior testimony, which acknowledged that alumni interviewers
have raised questions about Harvard’s use of race in the interview process. Harvard’s speculation
about alleged future harassment in discovery are premature at best and patently insufficient to
resist discovery of this relevant and accessible information.
Third, Harvard attempts to minimize the importance of its transfer process. But Harvard
cannot dispute the fact that the Complaint raises issues relevant to the transfer admissions. Nor
can Harvard dispute that it uses race in that process. Evidence probative of racial discrimination
is no less relevant or permissible when it affects only a few hundred applicants competing for a
few dozen spots. SFFA’s motion to compel should be allowed.

1

To the extent the Court in this case orders a partial stay in response to Harvard’s
pending Motion to Stay, SFFA is willing to modify its prior submission to this Court (Dkt. No.
83) and forego the production of application files at this time.
3

I.

SFFA Is Willing To Adjourn Its Motion for Files If Harvard Timely Produces the
Database So SFFA’s Expert Can Examine Its Sufficiency.
At the beginning of this case, Harvard told the Court and SFFA, in what appeared to be

good faith, that it was willing to produce a statistically significant sample of files to augment
electronic data that all parties agreed was relevant and discoverable in this case. See Dkt No. 26,
at 13. SFFA accepted Harvard’s proposal, requesting a sample of approximately four percent
from the pool of relevant files. Opening Memo. at 7-8. SFFA’s approach to sampling, and its
requested sample size, was well within the range of sampling deployed in many other cases,
including cases as complex and with file populations as large as this one. Id. at 8-12. Harvard
responded by offering an unrepresentative sample that was statistically worthless. Id. at 13.
In Opposition to the pending Motion to Compel, Harvard has not challenged the
statistical or legal principles underlying SFFA’s sampling approach. Indeed, as expected,
Harvard does not bother to claim that 160 files, half of which are cherry-picked by Harvard, are
of any statistical use at all. See Opening Memo. ¶¶ 13-14. Nor does Harvard take issue with the
importance of stratification, or contest the fact that numerous courts have ordered comparable—
if not larger—samples in a variety of other cases.
Instead, Harvard’s primary argument with respect to application files is simply that the
undisclosed database alone is sufficient for SFFA’s purposes. Opposition at 3-7. To support this
argument, it offers the hearsay declaration of its own expert, Mr. McCrary, who has been
provided access to “one year” of this data, and contends it has all the information Dr.
Arcidiacono needs to conduct his analysis. Id., McCrary Decl. ¶¶ 18. But neither Mr. McCrary
nor Harvard describes the contents of this information beyond high-level generalities (and two
specifics directed at a couple of examples that Dr. Arcidiacono raised—without the benefit of
reviewing this database).
4

Mr. McCrary’s declaration raises many questions. Which specific year of application data
did he review? Does the same amount of data exist for the four cycles for which SFFA requested
this data? What, precisely, are the 900 fields that Mr. McCrary “understands” are available for
the year reviewed? Did this understanding come from something other than his personal review?
Do any of these fields include the text of, or excerpts from, the personal statements or teacher
recommendations? Is there any information that Harvard uses for its admissions decisions that is
not in the database? How is this data entered? Does it accurately reflect the paper submissions?
Because Harvard’s representations have been entirely opaque and uncooperative, SFFA
requested that Harvard provide basic information about the database’s contents. Strawbridge
Decl. Ex. A. This request was more than reasonable, given that Harvard agreed back on June 15,
2015, that it would produce relevant databases and the fields and variables contained therein. See
Strawbridge Decl. in Support of Opening Memo., Ex. C at 11-13 (Request Nos. 1 and 3).2 But, as
it has with respect to every other one of SFFA’s document requests, Harvard still refuses to
produce a single responsive document, notwithstanding that discovery has been pending for three
months, and its request for a stay has not been granted. Nonetheless, SFFA hoped that Harvard
would recognize the absurdity of the parties and Court guessing at the contents of a database and
produce the information necessary for SFFA to determine if it could narrow or forego, at least
for now, its request for application files.
Harvard, however, has refused to provide even the most basic information to answer
these obvious questions, even though it may have helped to resolve this dispute. Strawbridge
2

SFFA’s Request No. 1 sought “All Electronic Databases from any time that include
information concerning early action applications, early action admissions, other freshman
applications, other freshman admissions, transfer applications, transfer admissions, freshman
enrollment, and total undergraduate enrollment.”
Request No. 3 sought “Documents sufficient to show the fields or variables included
within the Electronic Databases responsive to Request No. 1.”
5

Decl., Ex. A. Instead, Harvard doubled down on its prior position that it would not produce any
documents at all pending a decision on its request for a stay—even though it agreed from day
one of this case that these databases are responsive and must be produced. Id. For all practical
purposes, Harvard has granted itself a stay. There is no reasonable excuse why, five months after
its filing of the Answer in this case, Harvard has refused to exchange a single responsive
document with SFFA, despite SFFA’s repeated requests that the parties exchange discovery.
Despite Harvard’s obstinance, SFFA believes that it is appropriate to sequence its request
for application files in light of the new information about the contents of the database (which
Harvard could have provided months ago). If Harvard produces the database for the four years
requested by SFFA (encompassing the four admissions cycles from 2011-12 to 2014-15), then
SFFA will review the database and determine whether it needs the files after all, and, if so,
whether it can make do with fewer files than it initially thought. The sooner this information is
produced, the sooner the parties can resolve these differences and make meaningful progress in
this case.
Again, SFFA believes that its originally requested sample of application files may
ultimately be necessary. But in light of Harvard’s new claims about the alleged breadth and
depth of data, SFFA believes the sampling question can be litigated with a better factual
understanding of what the data does, and does not, include. Because Harvard has refused to
provide this data to SFFA or the Court, SFFA respectfully requests that the Court order its
production within 30 days.3 This timeline is more than reasonable, given the electronic and
readily accessible nature of the data and the limited need for any individual redaction effort.

3

Harvard raises a number of arguments about the alleged burden of the Family
Educational Rights and Privacy Act (“FERPA”). SFFA strongly disagrees with Harvard’s
claims about the requirements and burden of FERPA, especially given the Protective Order and
6

II.

Harvard’s Alumni Interviewers Are Witnesses with Discoverable Information.
SFFA is entitled to discover the “identity and location of persons who know of any

discoverable matter.” Fed. R. Civ. P. 26(b)(1). This straightforward principle has been repeatedly
upheld by this Court. See, e.g., Massachusetts v. First Nat’l Supermarkets, Inc., 112 F.R.D. 149,
152 (D. Mass. 1986) (“It is clear that the plaintiff is entitled to discover the … identity and
location of persons having knowledge of any discoverable matter.”) (quoting Rule 26(b)(1));
Clark v. General Motors Corp., 1975 U.S. Dist. LEXIS 12095, at *4 (D. Mass. 1975).
Harvard tries to convince this Court that information possessed by its alumni interviewers
is of “limited relevance” (Opposition at 13) and that these reports have “only a very small effect
on the ultimate composition of a Harvard class” (Opposition at 14). This Court should not be
fooled. Ms. McGrath’s deposition made it crystal clear that a significant number of alumni
interviewers are likely to have information directly relevant to this case. For example, Ms.
McGrath testified that alumni interviewers “have certainly had questions” and are “often asking
us questions about race in the process.” Strawbridge Decl. Ex. B 124:2-4, 13-18. Harvard’s
alumni interviewers “often” tell Harvard admissions personnel that “[y]ou admitted too many of
this” or “too few of that. This could be any kind of characteristic, and it certainly has been
applied to race.” Id. at 124:22-125:7. Moreover, Ms. McGrath recalled some specific occasions
where alumni interviewers raised specific concerns about discrimination against AsianAmericans. Id. at 125:12-126:18. And contrary to Harvard’s suggestions, Ms. McGrath and other
Harvard personnel have previously indicated that alumni interviews, in fact, played an important

the statute’s authorization to produce even unredacted student records so long as notice is given
to students. See 20 U.S.C. § 1232g(b)(2)(B); 34 C.F.R. § 99.31(a)(9); see also Doe v. Ohio,
2013 WL 2145594, at *5 (S.D. Ohio May 15, 2013). Because Harvard’s arguments are all
directed at the application files, and not the databases, see Opposition 9-12, there is no need to
address the FERPA issues in detail at this time.
7

role in the admissions process. See Opening Memo. at 18.4 In light of this testimony, it is
disingenuous for Harvard to claim that it “cannot conceive of a reason why” alumni interviewers
might have relevant information in this case. Opposition at 14.
Learning the mere “identity and location” of a witness who may have discoverable
information that is in Harvard’s possession does not subject that witness to “annoyance,
embarrassment, or oppression.” The one case Harvard relies upon for that strained proposition—
Cusumano v. Microsoft Corp., 162 F.3d 708 (1st Cir. 1998)—is wholly misplaced. That decision
involved a subpoena raising “fundamental First Amendment value[s],” in which a party sought
production of research materials (including notes, tapes and transcripts) from third-party
academic investigators for use in an antitrust case (i.e., far more than just the identities of
witnesses SFFA now seeks). Id. at 712, 717. Harvard cannot preemptively wall off an entire
category of people with relevant information by asserting that they are “volunteers” and
speculating about the potential effects if a handful of relevant witnesses are ultimately deposed.5
The interviewers can resist discovery if and when requests are made, and the fact that they are
volunteers does not insulate them from inquiry into their role in a racially discriminatory process.

4

See also William R. Fitzsimmons, Guidance Office: Answers from Harvard’s Dean,
Part 1, N.Y. Times, Sept. 10, 2009, at http://thechoice.blogs.nytimes.com/2009/09/10/ harvarddeanpart1/ (noting that “[s]tudents’ intellectual imagination, strength of character, and their ability to
exercise good judgment” are the “critical factors in the admissions process” and are “revealed
not by test scores” but, in part, by “alumni/ae and staff interview reports”).
5

Harvard’s suggestion that SFFA intends to depose thousands of interviewers is absurd.
In any event, Harvard will receive notice before SFFA sends any third-party subpoenas, and will
thus have the opportunity to object or seek relief in this Court if it has any concerns about the
scope of that discovery. But Harvard cannot withhold information in its own possession at this
stage based on its rank speculation about what might happen in the future, and the effect it might
have on some of the recipients. Even more absurd is Harvard’s claim that it can conceal the most
basic information about the interview process, but place the burden on SFFA to come back with
“with a targeted request” for interviewer discovery and “explanation of that need.”
8

More importantly, this Court has an obligation to ensure the record permits thorough
examination of Harvard’s admissions process, in practice. Fisher, 133 S. Ct. at 2421. Alumni
interviewers are a key part of that process. Preventing SFFA from learning the names of a subset
of those personnel most likely to have relevant information is inconsistent with the responsibility
to provide a full record from which to determine whether Harvard can satisfy strict scrutiny.
Again, it is Harvard’s decision to use race that has rendered this discovery essential. Any chilling
effect on alumni interviewers is secondary to the Supreme Court’s clear requirement of close
scrutiny when a university makes its decision in part on the race of its applicants.
III.

Harvard’s Transfer Student Admissions Process Is Encompassed by and
Relevant to This Litigation.
Any matters “relevant to any party’s claim or defense” are within the scope of party-

controlled discovery. Fed. R. Civ. P. 26(b)(1). Harvard pretends that its transfer student
admissions process is not encompassed within SFFA’s Complaint, but that document speaks for
itself. See Complaint ¶ 24 (describing SFFA members who “seek to transfer to Harvard when it
ceases the use of race or ethnicity as an admissions preference”); id. ¶ 300 (alleging that
Harvard’s transfer process may be used to further improper racial balancing and preserve a
specific racial composition of students); ¶¶ 334-38 (describing how Harvard’s transfer student
process could constitute a race-neutral alternative to its race-based admissions process). The
Complaint puts the transfer admissions process directly at issue, and that is all that is required to
bring it within the realm of discovery under Fed. R. Civ. P. 26(b).
Missing from Harvard’s opposition is any mention of the conceded fact that its transfer
student admissions process uses the same “holistic” approach, including the use of race, that
applies to freshman applicants. See Opening Memo. at 19-20 (quoting Answer ¶ 186).

9

Information concerning the transfer student admissions process is just as relevant to SFFA’s
claims as is its identical freshman admissions process.6
Harvard further contends that transfer admissions “play a tiny role in the composition of
any Harvard undergraduate class.” Opposition at 15. But Harvard’s own witness concedes that
more than a thousand students compete for one or two dozen slots each year. The unlawful use of
race is not excused simply because the competition for these slots is substantial.7 Because
Harvard’s transfer process is derivative of its freshman application process and because both
processes use race in identical fashion, discovery on this topic is likely to uncover information
that will assist in illuminating and clarifying the fundamental issues in this case. SFFA’s motion
to compel information about the transfer admissions process should be allowed.
CONCLUSION
For the foregoing reasons and those expressed in SFFA’s Opening Memorandum, SFFA
respectfully requests that the Court: (a) order Harvard to comply with SFFA’s Requests for
Production Nos. 19 and 24; (b) order Harvard to produce documents regarding its transfer
admissions process on the same terms it produces information regarding its freshman admission
process in response to SFFA’s discovery request; and (c) grant further relief as the Court sees fit.
6

Harvard appears to believe that SFFA’s membership must include a current applicant
who already has applied to transfer in order to challenge its use of race in that context. See
Opposition at 14-15. But an applicant may challenge a university’s admissions process if she
demonstrates that she “was ‘able and ready’ to apply as a transfer student should the University
cease to use race in undergraduate admissions.” Gratz v. Bollinger, 539 U.S. 224, 262 (2003).
Gratz rejected the argument that a party had to submit a transfer application, especially where
the “use of race in undergraduate transfer admissions does not implicate a significantly different
set of concerns than does its use of race in undergraduate freshman admissions.” Id. at 264.
7

Harvard’s admissions rate on transfer students ranges between 1 and 2 percent, see
McGrath Decl. ¶¶ 7-8, while its overall freshman admissions rate is roughly 5.5 percent. See,
e.g., https://college.harvard.edu/admissions/admissions-statistics. In both cases, Harvard admits
“relatively few” applicants; and it overstates the differences when it claims its transfer
admissions rate is “not comparable to the rate for freshman applications.” Opposition at 14.
10

Respectfully submitted,
By: /s/ William S. Consovoy
Paul M. Sanford BBO #566318
Benjamin C. Caldwell BBO #675061
BURNS & LEVINSON LLP
One Citizens Plaza, Suite 1100
Providence, RI 02903
Tel: [phone]
Fax: [phone]
[e-mail]
[e-mail]

Dated: August 6, 2015

William S. Consovoy
Thomas R. McCarthy
J. Michael Connolly
CONSOVOY MCCARTHY PARK PLLC
3033 Wilson Boulevard, Suite 700
Arlington, Virginia 22201
Tel: [phone]
Fax: [phone]
[e-mail]
[e-mail]
[e-mail]
Patrick Strawbridge
BBO #678274
CONSOVOY MCCARTHY PARK PLLC
Ten Post Office Square
8th Floor South PMB #706
Boston, MA 02109
Tel: [phone]
[e-mail]
Counsel for Plaintiff Students for Fair
Admissions, Inc.

CERTIFICATE OF SERVICE
I hereby certify that this document filed through the ECF system will be sent
electronically to the registered participants as identified on the Notice of Electronic Filing (NEF).

/s/ Benjamin C. Caldwell

4846-3986-1286.1

11