Diversity at Emory: An Open Letter
The university's response to SFFA v. Harvard raises questions about the extent of its racial discrimination, its plans to abide by the decision, and the meaning of and evidence for diversity.
Dear President Fenves,
The recent administrative messages related to the Supreme Court’s decision on affirmative action in college admissions—your message of June 29, Laney Graduate School Dean Arriola’s email of June 30, Candler Dean Jan Love’s message of July 3, and your 2022 essay for The Hill—show impressive unanimity on the issue of diversity, but they leave several questions.
1. Has Emory discriminated in the same way as Harvard?
As you know, the Supreme Court criticized Harvard University for having engaged in a “standardless” practice of using racial preferences in admissions with “no end in sight” in pursuit of “elusive” goals and “imponderable” interests while relying on “imprecise,” “arbitrary,” and “overbroad” categories, without articulating “a meaningful connection” between means and goals, in violation of legal equal-protection guarantees. In your public reaction, you do not directly address that critique, though you refer to the decrease in minority admissions at institutions that were already required to adopt race-neutral policies; in your earlier op-ed piece, you express the fear that, while “[r]ace is only one factor . . . without the ability to even consider it, the consequences are clear: Emory will become less diverse and minority enrollment will decline.” From these statements, I infer that our policy resembles Harvard’s. But how close is the similarity? Do all elements of the critique apply here?
Harvard put Asian-American applicants at a particular disadvantage, admitting far smaller proportions within any academically defined decile than of any other group and finding them lacking in personal qualities year after year; the Court also noted that, while Harvard claimed to judge applicants as individuals—a claim echoed in your message—this supposedly individualized process happened to produce roughly the same percentages in different racial groups year after year. Have Emory admissions and enrollments followed a similar consistent pattern, to the detriment of Asian applicants, or can we show variation to prove that we did not engage in equally purposeful discrimination?
I phrase my initial response in the form of questions, because even Emory’s undergraduate admissions process, more transparent than most, is quite opaque when it comes to diversity. Under Dean of Admission John Latting, Emory has involved some faculty in decision-making—I myself have participated—and given reporter Jeffrey Selingo uncommon access. In his book, Who Gets in and Why, Selingo reports that Emory rates applicants in four areas: “high school curriculum, extracurricular activities, recommendations, and intellectual curiosity”—the last of which he calls “a somewhat nebulous category determined by anything in the application from their essay to their activities.” Presumably, racial diversity can enter through this curiosity rating, but how is not clear. He also says that “Latting pays attention to numbers and he makes compromises—this many low-income students but that many legacies or this average SAT score but that much diversity” but without explaining how these “compromises” work or how much they compromise the formal ratings. At the same time, Latting stresses to his team that the top factors predicting first-year performance are recalculated GPA, followed by the rigor of the high school curriculum, so that they shouldn’t “waive off the numbers . . . under the name of holistic review”—causing Selingo to comment that it “sounded like another contradiction in a job full of them.” The severity of that contradiction again remains fuzzy, though if we can reasonably predict freshman grades from initial data, Emory should be able to estimate the academic impact of using race as “only one factor,” to use your phrasing. To my knowledge, such estimates have never been shared publicly, and therefore the weight of that “one factor” is uncertain as well. What is it, exactly?
2. How will Emory abide by the letter and spirit of the decision?
Judging by your public messages, Emory will have to make some major changes, but since our record is opaque, it is not clear how big those changes will need to be and how big an impact you expect. Could you be more specific?
The majority opinion cautions against trying to do indirectly what cannot be done directly and against the use of essays to circumvent its demand for equal treatment. That would require guardrails currently missing in the process Selingo describes. In your message, you note that the decision outlaws the use of race as an “explicit” factor in admissions, which could be read as signaling that you will encourage Emory officials to keep using race as an implicit factor to the greatest extent possible. Am I reading that signal correctly? The Court itself seems to open a backdoor to some implicit preferences by giving approval to “considering an applicant's discussion of how race affected his or her life, be it through discrimination, inspiration, or otherwise,” and I wonder if you and your colleagues aim to open that backdoor as wide as you can, for example by encouraging potential minority applicants to stress how they suffered from “systemic racism.” Of course, that might raise the prospect of what some cynical young people now call “trauma dumping,” as a way to game the system and appeal to administrators still pursuing diversity. Will we try to prevent that? Since you predict decline in URM college admissions if race is excluded, you in effect set up a criterion for judging compliance with the decision—if our numbers don’t change, outsiders might conclude that we are using race just as much as we did in the past. In your view, what reduction could they reasonably expect?
Without questioning the dedication of our admissions staff or John Latting’s professionalism in any way, I wonder if this might be an opportunity to rethink how we do admissions. Of course, elite colleges have spent much time and effort to “shape” a class, using detailed personal information, including on race, in recent years. Race aside, a university like Emory needs to consider at least academic talent and likelihood of attendance. Using such criteria to make an initial selection, one could imagine first constructing the relevant pool and then randomly selecting students for admission, thus maximizing diversity on a range of criteria while clearly satisfying the Court’s standard. Methods of this sort would provide clarity to applicants, advertise our commitment to fairness and transparency, and reduce the need for a cumbersome admissions bureaucracy. After selecting some applicants specifically for varsity sports, debate, and orchestra, why not randomize?
The reasoning in SFFA v. Harvard that invalidates race preferences in undergraduate admissions would seem to apply to employment as well. I therefore wonder if you and your colleagues think the decision has implications for faculty recruitment and hiring. Should we now be more cautious about “targeted” and “cluster” hires that explicitly take race into account as a way to diversify the faculty ranks? Will the university general counsel provide clearer guidance?
3. How does Emory weigh the relative merits of diversity and equality?
The Court decided that the UNC and Harvard admissions policies violated equal protection of the laws by discriminating in favor of some races over others. In your response, you express great concern about the decision’s effect on diversity at Emory, a priority you highlight and celebrate. But in your statements, and the deans’ messages, I did not see any affirmation of the principle at stake in the cases, i.e., equality under the law—which, unlike diversity, is actually articulated both in the Fourteenth Amendment and in the Civil Rights Act of 1964. If Emory still supports that principle, why did you and the deans not explicitly affirm that commitment? Should we understand your explicit endorsement of diversity without commensurate attention to racial equality under law as a preference for subordinating the latter to the former, in the way UNC and Harvard did? Though perhaps the Court moots the issue, in the sense that the institution is now legally bound to take non-discrimination seriously, it would still be useful to understand the university’s position.
4. What does Emory mean by diversity?
Assuming diversity will remain a high priority, it would also be good to clarify its meaning. The first paragraph in your message refers to the benefit of a student body with a “vast range of backgrounds and perspectives.” In your second paragraph, you express concern about institutions’ inability to admit a desirable percentage of “underrepresented minority students” if race cannot be used as a selection factor. This leaves your main argument for diversity unclear—does it mainly enrich education or remedy underrepresentation? If these rationales are connected, as they appear to be, do you consider the presence of racially underrepresented students the primary yardstick for judging the “range of backgrounds and perspectives” among our students?
In speaking of “underrepresented” students, what is the standard of representation you use, or that Emory would prefer to apply? Does underrepresentation only refer to certain racial groups, or does it also encompass other social categories—for example, would you consider men, black and white, to be currently underrepresented at Emory? If some groups are underrepresented, presumably we have too many members of other groups—which ones? Do we, or should we, judge underrepresentation among students in relation to certain groups’ proportion of the American population or of recent high school graduates or of young people who possess a certain kind of academic ability? In previous litigation, institutions invoked the notion of “critical mass”—perhaps an unfortunate metaphor— though UNC and Harvard seem to have jettisoned the concept; does Emory still assume that certain numbers of students with certain backgrounds are needed to achieve adequate representation?
In his recent book, Classified, legal scholar David Bernstein explains the very ad-hoc, historically contingent origins of the conventional racial categories used in the pursuit of diversity. Because “Hispanic” and “Asian” are broad and ill-defined, he finds those categories quite arbitrary, as did the Supreme Court. From his analysis I infer that is difficult, apart from mere numbers, to associate “Asian” or “Hispanic” with a particular contribution to diversity. Especially since we now face legal restrictions on the use of such categories, how can Emory pursue diversity in the future while taking the Bernstein critique seriously?
Your own message refers to diversity of perspectives and Dean Arriola says even more emphatically that “At Laney, we believe that innovation and excellence come from embracing the broadest definitions of diversity and inclusion.” I had not realized that diversity of perspectives and aiming for the “broadest definitions” of diversity was part of the institutional self-conception. Since most local and national discussion had centered on race and ethnicity, and to some extent on gender and sexual orientation, Emory’s definition of diversity had always seemed somewhat restrictive. In recruiting students and faculty, we have not, to my knowledge, made diversity of actual “perspectives” a high priority, so in that sense our working definition never seemed the “broadest” possible. If national trends among faculty and graduate students are any indication, an increasing proportion of current and aspiring academics lean left or far left, and many fields now feature few, if any, Republicans. After diversifying for decades, Harvard has only a handful conservatives on the faculty. Of course, that trend has many causes, but if it is unfolding at Emory while we strive to enact the broadest definition of diversity, it would seem concerning that the views of roughly half the country are underrepresented here. Do you consider it a problem that Emory is not as vastly diverse in political and ideological perspectives as the country at large?
Perhaps you don’t. On many public issues, the administration appears to speak with one voice, expressing one perspective. Of course, from a leadership standpoint, it makes good sense to have a unified team working toward shared goals, but its range of views seems far less vast than what you consider desirable in the institution as a whole. Do any of your colleagues ever dissent from the consistently progressive stances you take? Even if consensus generally prevails at the top, do you also promote diversity of perspectives in some way? Are you in fact trying to lead by example on this issue?
5. How does Emory gauge the benefits of diversity?
If Emory intends to pursue diversity as much as possible in the future, it would be good to understand the evidence for it. Like many official statements, your messages were light on facts. Regarding diversity, institutional rhetoric tends to focus on intentions and inputs, not data and outputs. Of course, Emory reports the demographic profile of faculty and students, and perhaps that is all we need, but your statements indicate that you expect demography to pay off in actual learning. Anecdotally, many of us have a sense of how students with diverse backgrounds can enrich class discussion on some subjects and challenge their peers constructively, which we welcome. Assuming that, in a scientifically oriented institution, diversity is not a dogma exempt from empirical assessment, what kind of evidence has Emory collected to bolster its commitment to diversity as integral to education? To my knowledge, which may be faulty, we have not seriously studied the issue locally, and we would not be able to satisfy the Court’s frustration about lack of evidence on the link between means and goals, but I would be happy to stand corrected on that score.
Of course, this is more than a local issue. With apologies for not doing a thorough literature review, I am also not aware of broader studies, such as experimental comparisons of more and less diverse educational settings, or before-and-after studies that show how learning at Emory or its peers improved in measurable ways when they increased the kind of diversity they pursued, or studies demonstrating that students at race-conscious affirmative-action institutions like UNC enjoyed a superior education compared with legally race-neutral institutions like UC Berkeley, or studies diagnosing the presumed disadvantages that score-oriented CalTech or racially homogeneous HBCUs or top universities in China suffer due to their lack of diversity.
In the litigation leading up to the Supreme Court decision, UNC and Harvard appear not to have supplied any evidence on diversity’s practical payoff, as the majority opinion noted. Even the 2011 Wabash National Study report on college experience, covering thousands of students over several years, does not directly address diversity’s effects, but offers mixed evidence on student attitudinal change, noting a modest increase during college in “universality-diversity awareness,” but a modest decline in “openness to diversity and challenge” (and a larger decline in “academic motivation”!). If I read these sources correctly, neither litigant advocacy nor large-scale studies supply much evidence for the claimed benefits of diversity.
Gauging the benefits of diversity strikes me, a sociologist without specific expertise in this area, as a hard problem in any case: how should we measure and model it, and how can we control for relevant confounding variables? To take a simple example related to (inter)national diversity: how do we assess the diversity benefit brought by Chinese students while disentangling it from the impact of their intellectual strengths? If there are any solid studies that answer such questions and undergird Emory’s diversity policies, I would be willing to do any homework you might assign.
Raising these issues assumes, as the Court did, that to evaluate success we need empirical assessment. But from your fervent statements one might get the impression that you view commitment to diversity as an absolute value to cherish rather than as (implying) a hypothesis subject to refutation or modification. Would that impression be incorrect?
6. How does Emory gauge the costs of diversity?
In your recent statements, you present diversity as an unalloyed good. I do not recall Emory administrators commenting on possible trade-offs. But have you considered such trade-offs? Have costs associated with diversity been part of your decision-making, and if so, which do you find most relevant? For example, if an institution pursues a certain type of student, presumably it is prepared to fund the effort through extra tuition support and offers of financial aid. In recent years, what were the additional expenditures needed, if any, to secure a certain percentage of URM enrollees? There might be an academic cost as well. For example, if admission on only academic grounds would lead to a reduction in URM student percentages, it would appear to follow that the academic credentials of a portion of recent URM admittees were not on par with those of other students. Roughly how large is the gap and any related disparity in grades? Some Supreme Court justices debated the relevance of so-called “mismatch”; has Emory investigated whether that occurs here?
In principle, it seems, you think minority representation and overall range of perspectives are positively correlated. But do all elements of diversity hang together so nicely? For example, some scholars of higher education have found that attitudes toward public issues vary by race and gender within academia, with women and African Americans leaning more liberal. If that finding is confirmed, the pursuit of more people with the right background in race or gender may reinforce the liberal-progressive atmosphere at most universities rather than expanding the range of perspectives. In other words, it may not be possible to maximize all dimensions of the “broadest” definition of diversity at the same time. Would you consider that a problem, or would ideological conformity be a small price to pay for greater racial diversity, if we can somehow keep promoting it under current legal constraints?
For what it is worth, in my view the liberal principles of equal justice under law and equal treatment according to universal standards outweigh racial discrimination in pursuit of diversity. In fields where quality matters—the NBA, science, higher education—the logic of “representation” has no place. Based on my own college experience abroad, and for the sake of basic fairness, I prefer simple and transparent, academically focused admissions procedures that treat everyone alike, for example in the form of random selection above a certain academic threshold. Though Title VI of the Civil Rights Act, not mentioned in your statements, rather clearly prohibits racial discrimination in federally supported programs, I also believe private institutions should have more leeway to discriminate than federal law or the Supreme Court currently allow. If leaders like you want to put their thumb on the scale in favor of black and Hispanic applicants, or commit an institution in the South to racial reparations for the sake of social justice, or for that matter limit admission to Methodists only, they should be able to do so openly and honestly, without legal contortions and without losing federal funds for research.
For now, however, the law fits neither your view nor mine. I will therefore follow with interest how you will steer Emory toward full and transparent compliance.