Diversity and Stereotyping
David Wasserman
There are, as my colleague Robert Fullinwider has pointed out, far more compelling rationales for affirmative action than the diversity of the workplace and classroom. Diversity, however construed, does not require proportionality, often regarded as a hallmark of affirmative action policies, or even the significant representation of any particular minority group. It lacks the moral urgency of arguments for corrective justice or social reconstruction. Moreover, one can reasonably argue that diversity is not as important in some contexts as in others.
Some of its advocates readily concede the limited scope of the diversity rationale. Akhil Amar and Neal Kumar Katyal, for example, believe that diversity has greater appeal in areas like education, where there is sustained interaction among the members of a community, than in areas like subcontracting, where interaction is limited. For this very reason, they argue, a Supreme Court that has struck down minority set-asides in government contracting may yet decide to uphold affirmative action in university admissions. A modest justification may also prove to be a resilient one.
There are other reasons to endorse the diversity rationale, despite its limitations. Even if diversity does not demand proportionality or confer enforceable rights on marginalized groups, it may increase opportunity and access in settings where even small gains in representation would constitute significant progress. Moreover, it appears to be a less divisive rationale than corrective justice or social reconstruction, especially where it emphasizes the benefits of greater minority participation to the larger society.
Acting on Generalizations
There is, however, another objection to the diversity rationale that would deny it even such a modest role. Critics argue that the very benefits it offers are predicated on objectionable forms of stereotyping.
It is readily apparent that most familiar uses of the diversity rationale involve generalizations from race, gender, or ethnicity. An urban police force wants to hire more black and Hispanic officers because it thinks that they are likely to have far better rapport with the disaffected and wary youth of those neighborhoods than their white counterparts. A corporation recruits women and minorities for its sales force on the assumption that they will generally be better than their white counterparts at pitching its products to female and minority customers, and that these customers are more likely to give their business to salespeople of their own race or gender. An urban school system wants more black and Hispanic teachers because it thinks they will generally be better than their white counterparts at spotting talent in, and motivating, alienated black and Hispanic students, as well as relating to parents and the broader community.
In each of these examples, the diversity rationale appeals to generalizations about the strength and influence of group loyalties, or about the degree of fellow-feeling and understanding between group members. Some of the generalizations concern the response of community members or clients to a diverse force of teachers or police officers or salespeople; others concern the likely attributes of the teachers, police officers, and salespeople themselves.
Such generalizations are not, of course, intended as universals. But even where they are carefully qualified, and exceptions duly noted, critics of the diversity rationale often find them troubling. In particular, they object to the use of race or gender as a proxy for skills, attitudes, and behavioral dispositions.
Certainly one can imagine instances where such use would be unacceptable. Suppose an organization hires more blacks in order to get a more athletic workforce. In that case, the underlying generalization is offensive, associated with a long history of invidious discrimination, and unnecessary, since the employer can test all applicants for the desired skills rather than relying on race as a proxy.
In the more familiar examples above, however, the underlying generalizations are less offensive, as well as less dispensable. Certain attitudes and behavioral dispositions, like rapport with a wary or alienated population, are difficult to test for, and their association with race and ethnicity does not have the invidious character of generalizations about talents. Still, generalizations about attitude and behavior, or so I will argue, can have significant moral and social costs. In contrast, generalizations about experience may be less troubling. I will explore these two kinds of generalizations as they apply to the pursuit of racial diversity in higher education -- the venue which, according to Amar and Katyal, is the most hospitable to any diversity rationale.
The Campus Mix
Consider a standard argument for diversity among students and faculty. Advocates claim that it is important for colleges and universities to increase the representation of blacks on the ground that they are likely to have attitudes, experiences, and values which it is desirable to include in the campus mix. Although those attitudes, experiences, and values are neither unique to blacks nor shared by all blacks, blacks are, by virtue of their upbringing and treatment, more likely than other people (specifically white males) to possess them.
Again, there are two grounds for opposing such generalizations. First, critics argue that the generalizations are dubious or unreliable. This complaint has some sting to it, since advocates of diversity are often eager to point out the implausible or poorly established claims embedded in other people's generalizations. Weighing the value of diversity against other considerations that enter into admissions decisions, Amar and Katyal note that "SAT scores and grades are at best a crude proxy for a student's potential to teach other students." But then, race or gender is also a proxy for that potential. How crude a proxy depends, as we will see, on what minority students are expected to teach their classmates.
The critics' second complaint is that racial generalizations are inherently objectionable, and that in endorsing them, proponents of the diversity rationale are guilty of a fatal inconsistency. Abigail Thernstrom makes the complaint in these terms:
Affirmative action proponents seem to want Americans to indulge in racial stereotyping for some purposes (the drawing of district lines, the classification of applicants into victim and nonvictim groups for purposes of admission to institutions of higher education, etc.), but violently object when they view such stereotyping as a danger in other contexts [such as news coverage reporting the race of crime perpetrators or suspects]. . . . One is tempted to ask, which way do you want it, folks? Is a high degree of race consciousness beneficial or pernicious?
As it is framed, Thernstrom's challenge might seem easy to meet. A high degree of "race consciousness" is pernicious when it hurts the members of stigmatized groups; it is more defensible when it helps them. A negative generalization about the violent or criminal behavior of young black men is objectionable in part because the burdens of its overbreadth fall so clearly on the innocent, law-abiding members of a vulnerable and disadvantaged community. It is true that the overbreadth of positive generalizations -- for example, that black college applicants have shown perseverance and resilience in the face of pervasive bias -- will confer a competitive disadvantage on non-minority applicants who do not enjoy a similar presumption. But this might reasonably be regarded as a less egregious injustice.
Such a response to Thernstrom, however, overlooks the less obvious burdens that even the most favorable racial generalizations may impose on blacks themselves. Some critics of the diversity rationale contend that generalizations regarding race, however positive, harm their subjects by perpetuating one of the most oppressive features of their stigmatization: to be seen primarily as representatives of a group rather than as individuals.
A Burden of Expectations
Jim Chen, for example, argues that generalizations about the experience or perspective of minority candidates for faculty positions function as ideological straitjackets. "Under affirmative action," Chen writes, "the mind of the minority professor becomes res universitatis, something belonging not only to the academic community that she has voluntarily chosen, but also to an external, race-based community to which she has been ascribed. Her mind is no longer her own, having been conscripted in large measure for service to both of these communities." Of course, any successful candidate, minority or not, may be measured against the expectations under which she was chosen. But the burden of such expectations is greater for minority candidates, since the contribution they are expected to make to diversity is understood not with reference to their individual talents or interests, but rather to their membership in a particular group.
Chen may be justified in claiming that the diversity commonly sought by universities pressures those hired under its rubric to adopt minority views, pursue minority research, and engage in minority advocacy. The standard terms used by proponents of diversity, such as "viewpoint" and "perspective," are ambiguous, covering, on the one hand, the experiences an individual has had and the culture she has absorbed; on the other, the positions and opinions she has adopted or is likely to adopt, and the interests and commitments she has acquired or is likely to acquire. The latter understanding of "viewpoint" or "perspective" diversity, which emphasizes belief and behavior, may well be the one that informs most academic and corporate policies. And it is easy to see why such expectations would be terribly constricting.
It may be, however, that valuable kinds of diversity can be pursued with less offensive generalizations. I want to suggest that generalizations concerning background and experience are less constricting and oppressive than those about behavior or attitude. The case for diversity becomes less problematic when it focuses on what a candidate has experienced rather than on what she has done or is likely to do.
Experience and Background
In an academic setting, diversity does not require us to favor minority candidates because they are likely to express acceptably unorthodox views, or to engage in approved forms of activism. Rather, the preference for minority candidates is based on an expectation that they will bring to the community important types of experience to which most of its members have very little exposure. These types of experience may include the candidate's firsthand encounters with certain social facts, such as poverty or exclusion, and her knowledge of a culture which exposed her to a broad range of such experiences and gave her a variety of ways of understanding and coping with them. A preference for diversity in life experience and culture would favor candidates not only from "Title VII minorities," but also from insular Appalachian and Amish communities, as well as Islamic and formerly Communist countries. It would overlap with a preference for geographical diversity to the extent that geography shaped the candidates' upbringing and experience.
The pursuit of this sort of diversity is not premised on the expectations about opinions, interests, and commitments which Chen finds so objectionable. Far from relying on the "direct equati[on of] race with belief and behavior" denounced by Justice O'Connor in Metro Broadcasting, it may well challenge any such equation. Part of the educational value in such diversity comes precisely from seeing the complexity and indeterminacy of the relationship between experience and culture, on one hand, and beliefs and commitments, on the other.
Of course, the extent to which race or ethnicity is associated with distinctive experiences and culture will depend on how much commonality there is to the life experiences and culture of group members, and this will obviously vary with time and place -- Jews in late 20th-century America, for example, undoubtedly share far fewer significant experiences than did Jews in 17th-century Poland. There is certainly room for disagreement about the commonalities in the experiences of African-Americans, women, and other underrepresented groups. Conservatives and optimists, for example, tend to think that the end of legal segregation and the increase in economic opportunity has created a black middle class that has much more in common with its white counterpart than it does with the poorer blacks left behind in the inner city. Many middle-class blacks, like Ellis Cose in The Rage of a Privileged Class, would argue that race continues to be a dominant and pervasive factor in their lives.
People may disagree about not only the extent but also the value of the experiences and culture shared by members of a particular group. Army brats may well share a lot of experience associated with transience and dislocation, but we may not feel that it is critical to include people with such experiences in our academic community. In contrast, we may regard an academic community as impoverished if it does not include people who have experienced certain kinds of exclusion or stigmatization. This kind of diversity may be especially valuable in a community whose members have largely led sheltered, privileged lives, lives that may incline them to moral complacency.
A pair of epistemological assumptions lies behind a preference for diversity of this kind. The first is that the actual experience of exclusion and stigmatization (mediated by the culture of the excluded and stigmatized group) yields knowledge and insight of a kind rarely obtainable by other means. The second assumption is that sustained personal interaction -- rather than, say, reading books or watching movies -- offers the best chance to convey something of this knowledge and insight, however imperfectly, to others. If the first assumption were false, the community would not need firsthand accounts of exclusion and stigmatization; if the second were false, it could get them from books.
Although the first assumption seems plausible, it is still an empirical generalization with notable exceptions. As Claudia Mills points out, individuals who are not members of minority groups can sometimes achieve, through their own powers of empathy and imagination, a vicarious understanding of the experience of group members. The second assumption is also plausible, but it may seem a peculiar one for a university to make. University education is premised on the effectiveness of books and other comparatively impersonal, non-interactive forms of communication in giving students insight into things they will never directly experience. While a university can also recognize the educational benefits of sustained personal interaction, its commitment to those benefits may be suspect. It is belied not only by the official tolerance of self-segregated dorms and classes, as Amar and Katyal point out, but also by an increasing and uncritical reliance on less personal (and social) educational media, like the Internet.
Burdens and Opportunities
Whatever criticism we may raise against the generalizations that sponsor the pursuit of experiential diversity, it seems clear that they do not straitjacket minority candidates as severely as generalizations about beliefs, opinions, and commitments. They do not involve treating individuals as members of groups from which, in David Bromwich's words, "all one's relevant supposed interests and opinions can be projected." Nonetheless, they may still have psychological and moral costs.
In the first place, being valued for one's group-specific experiences can be awkward or demeaning. It is something of an insult to have a host or friend turn to you and ask how you feel about some recent event as a black, a Jew, or a woman. This may be true even if the query assumes not that blacks, Jews, or women have a single view of that event, but merely that your reaction to it will be influenced by your being black, Jewish or female. The second assumption, I would argue, is less offensive than the first. But the distinction between them is hard to maintain, especially if you are the only black, Jew, or woman in a dorm, class or department. A minority of one is more likely to be treated as a representative or spokesperson for her group.
Second, even if minority students are recruited in sufficient numbers to discourage their typecasting, the pursuit of experiential diversity appears to assign them an educational responsibility not shared by other students. While they might ideally see this more as an opportunity than a burden -- a chance to make their classmates less insular and complacent -- such an educational process can be quite irksome: Minority students may feel that they are expected to remedy the ignorance, or gratify the curiosity, of people who ought to know better. In practice, the commitment to diversity may degenerate into an interest in the exotic. Moreover, those minority students who have led lives of inclusion and privilege may resent the expectation, however innocent, that they have unusual tribulations to share.
Finally, there is a danger that educational institutions -- buffeted by competing pressures from federal regulators, alumni, and their own faculty and students -- will be neither willing nor able to limit their recruitment, admissions, and hiring policies to the experiential generalizations I have tried to defend. Given the difficulties in distinguishing acceptable from unacceptable generalizations, there is reason to fear that the distinctions will be obscured in practice, if they are ever made. And even if conscientious administrators attempt to maintain them, these distinctions may well be ignored or rejected by the people who are affected by university admissions policies, from the minority students and faculty selected under them to the university community at large. If diversity will inevitably be seen as a rationale that supports the recruitment of minority candidates as representatives of, or advocates for, their groups, or as a smoke screen for other controversial agendas, its advertised benefit as a less divisive rationale for affirmative action may prove illusory.
Sources: Akhil Reed Amar and Neal Kumar Katyal, "Bakke's Fate," UCLA Law Review, vol. 43 (1996) Abigail Thernstrom, "Voting Rights: Another Affirmative Action Mess," UCLA Law Review, vol. 43 (1996); Jim Chen, "Diversity as Damnation," UCLA Law Review, vol. 43 (1996); Metro Broadcasting, Inc. v. FCC, 497 U.S. 547 (1990); Ellis Cose, The Rage of a Privileged Class (HarperCollins, 1993); Claudia Mills, "Multiculturalism and Cultural Authenticity," Report from the Institute for Philosophy and Public Policy, vol. 14, no. 1/2 (Winter/Spring 1994); David Bromwich, Politics by Other Means (Yale University Press, 1992); Patricia Elam Ruff, "Private School, Private Pain," Washington Post (February 23, 1997).