Affirmative Action as Culture War
July 13, 2001
For publication in The Blackwell Companion to Racial and Ethnic Studies, ed. David Theo Goldberg and John Solomos (Blackwell Publishers, forthcoming 2002).
Revision of article by the same name in The Cultural Territories of Race: White and Black Boundaries, ed. Michèle Lamont (Russell Sage Foundation Press and University of Chicago Press, 1999).
My thanks for comments on an earlier version of this article to Marianne Engelman Lado, Amy Gutmann, Luke Harris, Ira Katznelson, Michèle Lamont, Dale Miller, John Skrentny, Alan Wolfe, and the participants in various seminars at Princeton University.
NOTE: not quite final version.
If one examines any collection of books or syllabi on the subject of affirmative action over the past twenty-five years, one observes two phenomena -- a huge outpouring of legal and philosophical analyses of its merits and a paucity of empirical examinations of its mechanisms and effects. The legal and philosophical analyses range from passionate assertions that quotas are essential in order to mitigate American racism to equally fervent arguments that any racial or gender-based preference violates core American values of equality of personhood and opportunity. One can even find a few carefully nuanced intermediate positions that subtly distinguish among recipients, procedures, triggering circumstances, and the like (e.g Appiah and Gutmann 1996). Without denigrating the energy and creativity of many of these efforts, I believe it is fair to say that the core legal and philosophical positions were laid out in the first few years of this debate. With the exception of the recent claim about the intrinsic benefits of diversity for an organization, the succeeding volumes have mostly developed or elaborated upon these original claims.
What We Don’t Know about Affirmative Action
There are, in contrast, huge holes in the corpus of what we know about how affirmative action actually works in practice. For example, we know very little about just how people are hired or admitted to most universities. When is race or gender a tie-breaker; when does minority status still count against the applicant; when are less-qualified African Americans or women hired/admitted over more qualified whites, Asians, or men; when does the reverse occur? Do certain kinds of professional schools or firms consistently treat affirmative action in ways different from that of other kinds of professional schools or firms? I know of virtually no research within and across nonelite university admissions offices, corporate personnel offices, professional schools’ admissions offices, or police or fire departments that carefully traces these processes and analyzes them comparatively or theoretically.
What happens after a person is hired or admitted in circumstances where affirmative action is presumed to have played a role? Do blacks or white women feel stigmatized, inferior, insecure? If so, do they overcompensate by rigidity or racial paranoia or timidity? Do they feel any more insecure than, for example, alumni children admitted to universities as legacies, or white working-class athletes admitted to bolster Big Ten football teams, or the boss’s nephew put in charge of the front office? Alternatively, if they do feel insecure or are stigmatized, are they able to overcome their initial obstacles and succeed at about the same rate as other workers or students? What are the processes by which people who are initially labeled as “affirmative action hires” move toward success or failure; do they have more to do with internal fortitude, organizational culture, structural opportunities, or what? Do some contexts facilitate success or reify stigma more than others? Again, there are very few careful and systematic studies of co-workers’ interactions and corporate or university practices in which affirmative action is a central part of the organizational context.
Another set of questions: How many whites or men are told by admissions officers or personnel directors that they would have been hired/admitted if it were not for affirmative action pressures? After all, that is an easy and mutually gratifying response from a gatekeeper to an angry or disappointed candidate -- and to many such candidates in a row, so long as each is addressed in the absence of the others. An example: one man reported on an e-mail list his stellar undergraduate record and his admission to several excellent law schools, but with little funding. “At least three admissions counselors stated outright, and others implied, that had I been anything but white male and had those numbers [GPA, LSAT scores etc.], I would’ve been immediately full-ride plus living expenses” (Finley 1996). If his experience occurs frequently, that would explain why so many more whites than is arithmetically possible believe that they or someone they know about was denied a job or promotion because of affirmative action. (See my discussion of survey results below.) But to my knowledge, no one has conducted research on how affirmative action is presented to non-minorities denied jobs or admission or promotion.
We know very little about how minority set-asides for contracts work. Conventional wisdom holds that the process is corrupt, with the most common allegation being that an African American (or woman of any race) is used as a “front” for predominantly white (or white male) owners of a firm (for example, see Oreskes 1984). How often does that occur? Why cannot those who grant contracts find out about it? Do they try? Conversely, how often do minority set-asides function as they were presumably intended to -- giving a start to a firm headed by a woman and/or person of color who would not otherwise be able to find clients or win contracts due to racial or gender stereotypes? How and when does that more positive dynamic work, and what distinguishes successful from unsuccessful firms once set-asides are in place? Some analysts have begun to answer these questions, but the disproportion between assertion and knowledge is enormous (House Government Operations Committee 1994; Bates and Williams 1996; Enchautegui et al. 1996; Myers and Chan 1996).
Broader political research would also be useful. Why did affirmative action surface as an especially “hot” political issue in 1995, given that white men (and to a lesser degree, white women) have always disliked strong versions of it (Bennett et al. 1995: tables 2, 3)? Conversely, why haven’t corporations, universities, most political candidates, and city governments jumped on the anti-affirmative action bandwagon? Do they see benefits in affirmative action policies that they would not be able to attain absent an apparently-coercive governmental mandate? (An analogy here might be school superintendents who welcomed judicial decisions mandating desegregation because the mandate allowed them to make changes in the school system that were otherwise politically too difficult.) Why has there been an anti-affirmative action referendum only in two states (California and Washington) and one city (Houston) so far?  How should we understand the conflicting views of affirmative action held by white women and Asians?
Broader historical research would similarly be illuminating. In the 1960s, as Stephen Carter (1991) reminds us, many on the left saw affirmative action as an individualistic sell-out. In their view, affirmative action encouraged personal mobility of the most energetic, articulate, and effective actors within a racial or gender group at the expense of structural transformation that would benefit the whole group. Regardless of whether it was right or wrong, what happened to that view? Why did support for affirmative action move from being a relatively right-wing position (President Nixon established the Philadelphia plan partly to give more blacks a stake in the extant economic system, and endorsed set-asides in order to encourage black capitalism) to being a relatively left-wing position over the past thirty years? Conversely, why hasn’t the political right embraced affirmative action as a classic case of sponsored mobility, a process of carefully and selectively admitting a few outsiders into the elite and defusing claims of the rest?
Who is eligible for affirmative action in various locations and venues, and why? In Richmond, Virginia, for example, the city’s set-aside program covered Hispanics, “Orientals,” American Indians, Eskimos, and Aleuts as well as blacks; as Justice O’Connor pointed out in Croson, “there is absolutely no evidence of past discrimination against” members of most of these groups; “it may well be that Richmond has never had an Aleut or Eskimo citizen.”  When are women treated as a distinct class, and why? Do they receive different (better? worse?) treatment under employers’ affirmative action guidelines than African Americans, or Hispanics, for example? Who decides about cases of mixed or ambiguous racial or ethnic identity, and to what effect?
I could list other arenas in which empirical research about the processes and effects of affirmative action are scarce or missing, but the point should be clear: Americans’ elaborate and sophisticated legal and normative debate about the legitimacy, desirability, and impact of affirmative action has until very recently taken place in something close to a factual vacuum.
That might not be surprising in the political arena -- after all, debates in Congress about abortion funding, illegal immigration, the effects of nuclear fallout, intervention in Kosovo, and other highly controversial issues often ignore what evidence exists or occur in the absence of much evidence at all. Such ignorance might not even be always inappropriate; elected officials and judges must sometimes make decisions regardless of whether they know enough about the choices confronting them. Similarly, the relative paucity of empirical knowledge might not be disconcerting to advocacy groups. Advocates of a particular policy position -- whether support or opposition to affirmative action, nuclear disarmament, abortion rights, or something else -- either are not primarily motivated by empirical considerations, or believe that they know enough facts or the right facts in order to espouse their position with confidence.
But the disproportion between legal and philosophical analysis on the one hand, and empirical analysis on the other, is inappropriate in the academy. Social scientists do, one normally assumes, believe that knowledge -- about how processes work, what effects policy innovations have, how historical and institutional contexts affect behaviors and outcomes -- matters. But except for a few crucial exceptions mostly buried in scholarly journals or legal briefs, they have not expended much effort on empirical analyses of affirmative action. Why not? And what does the relative paucity of research compared with argumentation tell us about the policy of affirmative action, the cultural context of American racial and gender politics, and -- most broadly -- the ideology of the American dream?
My argument, in brief, is the following: in the current American racial culture, affirmative action is more important to participants in the policy debate as a weapon with which to attack enemies in order to win some other battle than as an issue in and for itself. To be useful as a weapon, affirmative action must remain at the level of moral claims and single-dimensional outrage; the messy and complex realities that are likely to surface in careful empirical analyses do not help much in political and cultural warfare. This phenomenon is as true in the university as in the explicitly political realm, because most scholars do not pursue the traditional objective of scholarly neutrality in the arena of racial politics and policy choices. To put it most succinctly, the debate over affirmative action is predominantly a culture war over who is an American and what it means to be a good American, and only secondarily a dispute between political parties or policy analysts over how best to improve the status of African Americans or women.
I do not make this assertion only to deplore it, although I will do some of that below. Nor do I claim that the culture war over affirmative action is unique; on issues ranging from a national bank in the 1830s to Communists in the State Department in the 1950s, Americans have always become passionately concerned about social “crises” that in fact matter little to their daily lives. Instead, I want to explore the culture war over affirmative action for a more analytic reason: comparing what we believe about affirmative action with what we know and what we have chosen not to find out tells us a great deal about Americans’ hopes, fears, and self-images.
What Do We Know about Affirmative Action?
There has been, of course, some excellent empirical research on the subject of affirmative action (almost all of which has appeared only in esoteric professional journals or in unpublished legal briefs). A review of what it tells us will begin to substantiate my claim about the symbolic functions of the debate over affirmative action.
To begin with, a few solid historical analyses of the development and implementation of affirmative action policy have been published (Burstein 1985; Graham 1999; Skrentny 1996). With one clear exception (Skrentny), these analyses seldom address the issue of “why affirmative action?” and “why at a given historical moment?” Instead, their main burden has been to demonstrate that federal policy was in fact effective. That is, once federal officials, especially several presidents in a row beginning with Lyndon Johnson, decided to endorse affirmative action and develop an institutional structure to implement and enforce the policy, it happened. Graham (but not the other historians of affirmative action) even argues, roughly, that the federal government was too effective in the sense that affirmative action policy created an elaborate system of social regulation by bureaucrats and courts that is now unresponsive to public wishes, excessively interventionist, and counterproductive.
Thus affirmative action was encouraged, or at least not halted, by both conservative Republican presidents (Nixon, Reagan, Bush) and liberal Democratic ones (Johnson, Carter, Clinton). Federal laws and executive orders to foster or mandate affirmative action could have been reversed, but were not, when both houses of Congress were controlled by Republicans (or by Democrats); they could have been declared unconstitutional, but were not, when the Supreme Court was dominated by either liberal or conservative activists. The historical record is one of growth and persistence, with considerable trimming (especially of business set-asides) and usually more lip service than effort -- but without rejection of its core mission -- at the federal level (Rice and Mongkuo 1998; Leadership Conference on Civil Rights 1997: 2; Anderson 1996). Thus with a few exceptions, conservative opponents have not fought hard to eliminate affirmative action when they might have succeeded; that is the first suggestion of a gap between rhetoric and practice among policy actors and advocates purportedly on the same side of the issue.
A little research provides evidence about the economic effects of affirmative action policies on African Americans (and/or white women) and on the corporations that hire them (Leonard 1984a, 1984b, 1986, 1990; Heckman and Wolpin 1976; Heckman and Payner 1989; Badgett and Hartmann 1995; Holzer and Neumark 1999; Holzer 1998; Rodgers and Spriggs 1996). Aggregate analyses of employment suggest that affirmative action did target blacks during the 1970s, and that it contributed to the creation of a substantial black middle class that has persisted over the succeeding three decades. Thus federal regulations not only institutionalized the initial policy but also produced the desired outcomes, at least to some degree. Affirmative action did not, moreover, target only those already well-educated or in the middle income brackets; its initial effects were greatest for people in unskilled and semiskilled job categories (Leonard 1984b). There is no evidence that affirmative action policies have harmed productivity of participating firms (Leonard 1984c).
The same story appears to hold for university admissions. During the 1980s, the most selective four-year colleges were most likely to admit African American and Hispanic students preferentially. In nonelite schools, which 80% of college students attend, students of different races with similar characteristics were admitted at essentially the same rate (Kane 1998). Elite colleges were not harmed by their energetic affirmative action practices; by the end of the decade, their tuition had risen disproportionately compared with other universities’, as had the number of students applying for admission.
For universities as for firms, individual participants as well as the institution appear to gain from being associated with affirmative action. Attending a high-quality school yields higher earnings in adulthood, even after controlling for family background and school achievement; studies even show higher payoffs to blacks than to whites of attending an elite school (Brewer et al. 1999; Kane 1998; Daniel et al. 1995; Bowen and Bok 1998). Thus affirmative action beneficiaries at elite universities did indeed benefit, to no apparent detriment to their fellow white students.
Affirmative action was not, however, the most important factor in decreasing the racial wage gap between the 1960s and 1980s. Enforcement of laws against employment discrimination, as well as increasing educational attainment and achievement among blacks, did more (Heckman and Verkerke 1990; Leonard 1990; Smith and Welch 1984, 1989). During the 1980s, when enforcement of federal affirmative action regulations was all but halted and there was no demonstrable aggregate impact of affirmative action policy, the wage gap between fully-employed blacks and whites fluctuated, but did not fundamentally change (Farley 1996: 249; see also Leonard 1986).
Thus analyses of firms and universities, while thin, suggest several results: when implemented with at least a little pressure for compliance, affirmative action has had slight positive effects on beneficiary groups and no discernable negative effects on employers or colleges. When all pressure for compliance disappears, affirmative action mostly disappears. Even at its strongest, affirmative action has had less impact on racial wage inequality than have the much less controversial policies of improving educational attainment and achievement and enforcing the law against employment discrimination.
Another small arena of empirical research encompasses psychological experiments (Clayton and Crosby 1992; Crocker et al. 1991; Steele 1999; Nacoste 1985, 1994; Blanchard and Crosby 1989; Basic and Applied Social Psychology 1994). Results from these experiments are important, but so far do not move much beyond a demonstration of common sense. If people are told that affirmative action has influenced their attainment of a position, they devalue the position or their performance -- more so if they disapprove of affirmative action to begin with or if it was described as playing a central role, and less so if they see the apparent authority as racially biased. White women feel more demeaned by “unwarranted help” than do African Americans, who are more likely to see affirmative action as an entitlement. Coworkers may change their initial judgment that a new black manager is incompetent if he demonstrates that he is not or if they want to help him succeed. African Americans perform less well than they are capable of if they believe that they are underqualified compared with white students but must nevertheless uphold the honor of their race. In short, affirmative action may or may not have harmful psychological and interpersonal effects, depending on who is involved, how the policy is deployed, and the context within which it is deployed -- extremely useful knowledge, but hardly grounds for passion in either defense or opposition.
Another arena of scattered but important research findings treats the organizational effects of affirmative action (Kelly and Dobbin 1998; Dobbin et al. 1993; Edelman 1990, 1992; Sutton et al. 1994; Konrad and Linnehan 1995). Since the laws and executive orders creating affirmative action did not detail concrete compliance mechanisms, managers have constructed a wide variety of practices and structures. “Personnel professionals, in particular, saw opportunities in these new areas of law and promoted responses that would expand their power and numbers”(Dobbin 1996). The elaborate array of rules, offices, grievance procedures, plans, tests, and other mechanisms has come to have a life of its own; “over time, these structures, more than any particular substantive result, tend to be equated with compliance” (Edelman 1996). Forms of affirmative action that comport with the economic goals of a corporation -- such as “diversity” for certain firms or in certain markets -- become part of the corporate culture and are thereby domesticated. They neither disrupt conventional practices of hiring and promotion very much, nor do they generate much opposition. If anything, many executives and managers strongly support affirmative action as they understand it for several reasons: they believe that they are doing something socially productive, they believe that if properly handled “diversity” can increase their profits, and they are not paying high costs for endorsing affirmative action (Thomas and Ely 1996; Cox and Blake 1991; Badgett 1995). Again, few grounds for intense defense or opposition here.
The very small amount of aggregate data available show no evidence that affirmative action creates a sense of stigma or inferiority in its recipients. African Americans in firms that have affirmative action programs are at least as happy in their jobs as African Americans in firms that do not. The former demonstrate greater occupational ambition and are more likely to believe that people are helpful than the latter (Taylor 1994; Blanchard and Crosby 1989). Blacks who believe that affirmative action played a part in their hiring or promotion have no less confidence in their ability to do their job than do other workers (Hochschild 1995: 98-102, 290-292).
Surveys of public opinion offer the most fully developed research arena (Kinder and Sanders 1996; Pettigrew and Martin 1987; Bobo and Kluegel 1993; Bobo and Smith 1994; Gamson and Modigliani 1987; Sniderman and Piazza 1993; Kluegel and Smith 1983; Schuman et al.1997; Bobo 1998). Surveys provide several results crucial to my claim that the debate over affirmative action is more symbolic than substantive. About a third of white Americans cannot make any association with the phrase “affirmative action”(Steeh and Krysan 1996: 129). Those who do have an opinion on the topic (most of whom presumably have some idea of what it means), produce more consensus than one would imagine possible if one listened solely to the political activists and news media.
Although about three-fourths of white Americans consistently agree that blacks should “work their way up... without any special favors,” so do about half of black Americans. Although 85% or more of whites endorse “ability” rather than “preferential treatment” to determine who gets jobs and college slots, so do about three-fifths of blacks. Conversely, fully seven in ten whites (compared with over eight in ten African Americans) favor affirmative action programs “provided there are no rigid quotas.” Solid majorities in both races endorse special job training and educational assistance for women and people of color, extra efforts to identify and recruit qualified minorities, redrawing of voting districts to ensure minority representation, and other “soft” forms of affirmative action (Steeh and Krysan 1996; see also Gallup Organization 1995). One quarter of those who voted for California’s referendum banning affirmative action in 1996 would have preferred a “mend it, don’t end it” option (Lempinen 1996).
If one can trust survey data, then, it seems possible to develop a workable political consensus around some “soft” affirmative action policies. African Americans and Latinos will always support these policies more strongly than will whites, and women will support them somewhat more strongly than will men. But that is no different from the pattern of support for any policy measure that benefits some citizens more than or at the expense of others. The likelihood of different levels of enthusiasm does not stop politicians and policy planners from initiating new laws and regulations, so long as they can forge a set of rules that some people strongly endorse and most can tolerate. There might be such a set of rules to be found within the wide range of possibilities for affirmative action policy, if the troops would declare a truce in the culture war long enough to move warily toward a no-man’s-land in the center.
Other survey data, however, suggest caution; this no-man’s-land will be safely occupied only if perceptions of affirmative action are brought more closely into line with its practice. Relatively few whites (under 10%) claim to have been harmed by the workings of affirmative action policy. Nevertheless, more (up to 20%) claim to know someone who has been so harmed, and still more (30% to 40%) claim to have heard about someone who was so harmed. Up to 80% of whites believe it likely that a white will lose a job or promotion to a less-qualified black due to affirmative action. One third of whites think affirmative action programs frequently “deprive someone... of their rights,” and half of white youth (compared with one-fifth of black youth) think that more whites lose out to blacks due to “special preference” than blacks lose to whites due to prejudice (Hochschild 1995: 144, 308; Steeh and Krysan 1996: 139-140).
Unless they define affirmative action as any situation in which an African American bests a white, regardless of why, these survey respondents are wrong. In 1994, only 2% of 641 government contractors polled complained of quotas or reverse discrimination. Of the more than 3000 discrimination cases to reach courts between 1990 and 1994, fewer than 100 charged reverse discrimination; most of those, like most claims about any kind of discrimination, were dismissed due to lack of merit (Stephanopoulos and Edley 1995, Sec. 6.3; Blumrosen 1995). Possibly political correctness inhibits firms from complaining about quotas, and perhaps white men suffer under the same constraint when they consider bringing charges of reverse discrimination. But surely a strong legal case would sweep aside those hypothesized constraints at least sometimes. Thus it seems safe to conclude that many more people are exercised by fears of the policy of affirmative action than are harmed by its implementation. If views about affirmative action are subject to rational consideration (a large “if,” and one which I dispute over the next few pages), then wide dissemination of information about who actually benefits from or is harmed by affirmative action would be an essential component of finding a workable middle position.
I have saved until last any discussion of the central exception to my complaint about the paucity of empirical studies of affirmative action. The Shape of the River is a very important work, both for its arguments themselves and because it is so unusual in this field. It confirms some of the critics’ fears; many African American students who attend elite colleges and universities do indeed enter with lower test scores and exit nearer the bottom than the top of their class. But it does more to confirm the supporters’ hopes: the more selective the school they attend, the better black students do in the short- and long-runs; white as well as black students value and benefit from racial diversity; black students are not paralyzed by insecurity, and black alumnae contribute importantly to their communities. The authors themselves, of course, are not neutral observers of the policy but they have striven – mostly successfully—to engage in serious social science research rather than in polemics. Responses have also been politically influenced, but they too have enhanced the discussion by reporting countervailing data of their own. Overall, The Shape of the River reinforces my basic points – that the consequences of affirmative action are mixed like those of most complex policies, that it is possible to do solid research on the subject, and that there are a huge number of unexamined questions remaining.
Affirmative Action as Cultural Warfare
Unfortunately, many people’s views about affirmative action are not subject to rational consideration, any more than people could be persuaded to think coolly about the “monster Bank” or Communism. If people did think about affirmative action by weighing its costs against its benefits, passions would not run so high. After all, the research record gives complete warrant to neither the hopes of supporters nor the fears of opponents. The policy has certainly aided specific individuals applying to elite colleges (Bowen and Bok 1998) and professional schools or to middle-level public service jobs (Collins 1983, 1997). It has arguably harmed those who would have been admitted or hired absent affirmative action (Lynch 1989; Taylor 1991). It may even have simultaneously helped and harmed a few people (Carter 1991; Steele 1999; Clayton and Crosby 1992). It has had great impact on police and fire departments, on a few law and medical schools, and on the Army. But compared to legislation and litigation against employment discrimination or barriers to voting and office-holding, or desegregation of schools, affirmative action has had relatively little aggregate impact. And it has, perhaps, benefited professional personnel officers and attorneys more than any other single group of people.
We face therefore two puzzles. Why are a lot of people so exercised over this particular policy, which distributes outcomes about as most other policies do and which has done less to change American racial hierarchies than have a variety of other less controversial policies? And why is the empirical base for understanding the practice and effects of affirmative action so thin and in most cases publicly invisible compared with the rich philosophical and legal arguments that the issue has evoked?
These two puzzles are both resolved by one answer: political actors find affirmative action an immensely valuable issue over which to debate, and therefore have little desire to figure out just how it operates. That is, affirmative action is too precious as a political weapon in a broader cultural war about what America stands for to be blunted by attention to real-life complexities.
Many opponents of affirmative action are less concerned with the policy per se than with a wider assertion that racial (or gender) discrimination no longer exists, and that African Americans’ continued claims of its persistence are merely whining or self-seeking. At their crudest, opponents are racially hostile. Less crudely, they are unable or unwilling to see structural barriers or institutional advantages that are independent of individual intentions or awareness. The most sophisticated opponents are more concerned about class or individual, rather than racial or gender, barriers (Lind 1995, Woodson 1996), or they judge the costs of a continued focus on racial differences to be too great now that the black middle class is fairly well-established (Sleeper 1997, Gitlin 1995), or they judge that laws and regulations against employment discrimination are strong enough to take care of the remaining racial bias in jobs (James Heckman, personal communication with the author, 1997). Alternatively, they espouse a principle of individual meritocracy which supersedes caveats or shadings (Glazer 1975; Eastland 1996; Thernstrom and Thernstrom 1997).
Many proponents of affirmative action demonstrate an equal but opposite dynamic. They are less concerned with the policy per se than with a wider assertion that racial (or gender) discrimination is just as virulent as it has always been, and that whites’ opposition is merely covert racism or inexcusable naiveté (Preston and Lai 1998; Motley 1998). At their crudest, proponents are paranoid or self-seeking. Less crudely they believe, as one of my students put it, that blacks have just as much right to a class structure as whites do and just as much right to use all means legally available to reach the top of it. The most sophisticated proponents see affirmative action as a means for individual blacks to overcome persistent racism and attain resources that will help the African American community and the nation as a whole to overcome its shameful past -- DuBois’s talented tenth (Rosenfeld 1991). Alternatively, they see employment discrimination (or biases in universities’ admissions policies) as sufficiently subtle as well as widespread that standard laws and litigation are insufficient. In this view, people of color and/or white women must be inserted into the core of the hiring/promotion/admissions processes in order to identify and bring to justice persistent hidden biases that work on behalf of white men (Bergmann 1996; Edley 1996; Reskin 1998).
Political elites within these groups do not talk to one another and have no electoral, social, or organizational incentive to do so. Legally, they each have a rich set of court cases, laws, and regulations to bolster their claims. Normatively, each group has available to it more philosophical justifications for its position than anyone can possibly read or use. Historically, each group can point to its preferred victories and defeats. Politically, each group has a core constituency and a wider set of citizens to whom it can turn for occasional support. Organizationally, each group has well-established but complex and constantly renegotiated (thus energy-draining) internal ties of communication, bargaining, and resource extraction to maintain. Socially, members of the two groups seldom encounter one another outside of formal, scripted disputes. Neither group has, in short, much reason or incentive to question its own position or give serious consideration to the other’s.
Academic research does not fit very clearly anywhere in this picture, which may explain why there has been so little of it compared with argumentation that is not empirically based. That is unfortunate for at least two reasons. First, some fascinating research questions are so far left untouched. Second, in my view only reasonably dispassionate analyses could provide the possibility of intellectual space for people to separate their broad beliefs about the role of race in America from their particular judgments about the efficacy of affirmative action compared with other possible routes to racial equality. Only if there is a cohort of people who can persuasively say, “it works in this regard but not in that one...” or “it works better than X but not as well as Y to achieve goal Z...” have we as a nation any chance to get past the shouting.
Why have advocates on all sides of the debate over American racial policy seized on affirmative action rather than, for example, wage discrimination or the quality of schooling in inner cities as the battleground for deciding what race means in the United States today? After all, affirmative action neither affects many whites nor comes close to solving the deepest problems of African Americans. So why is it “the highest pole in the storm”? Partly because opposition to affirmative action is one of the few remaining respectable vehicles for seeking to maintain white domination. No public figure can any longer argue, as one could fifty years ago, in favor of lesser schooling for black children or different wages for the same work based on one’s gender and race. In that sense, “ascriptive Americanism” lost the public debate to liberalism in the last third of the twentieth century (Smith 1997). But even if part of one’s motivation is to resist black competitive success, one can oppose affirmative action in the name of values that all Americans publicly claim to share (see Hochschild 1995: chapter 7, on white opposition to black competitive success).
Opposition to affirmative action remains respectable because both sides to this dispute call on Americans’ deepest and most cherished values (often the same value, in fact). The set of values most closely associated with passion about affirmative action can be characterized as the American dream. The ideology of the American dream is the promise that all Americans, regardless of their race, sex, or background, can reasonably anticipate the attainment of some success, if they use means under their own control such as talent, ambition, and hard work. The pursuit of success is associated with virtue, and so can legitimately be seen as noble and elevating, rather than merely materialistic or selfish (Hochschild 1995). The ideology implies a balanced contract between the public sector and private individuals. The government is expected to provide equality of opportunity – thus no discrimination by race, gender, class, or religion – and a structure that makes anticipation of success reasonable – thus a strong system of public education, a flourishing economy with plenty of jobs, physical security from enemies without and crime within the nation. Citizens, in turn, are expected to act within that framework to take care of themselves – to attain skills, support themselves and their families, refrain from discrimination or false claims of victimization, and follow a moral code.
Most Americans believe in each component of the American dream and most share an implicit conviction that the balance between governmental and personal responsibility is about right (Hochschild 1995: chapter 1). Hence the United States lacks strong libertarian and Marxist political parties, as well as widespread white racist or black nationalist social movements. But, partly because their shared convictions are so strong, Americans contend fiercely over just how to translate those convictions into concrete practice.
To some, the American dream is necessarily individualist. The ideology is intended to create a structure within which each person can rise or fall according to his or her merit alone. Some individualists reluctantly support affirmative action on the grounds that it is a necessary way station on the path from racial domination to race-blind individualism. But most see affirmative action as a denial of the deepest tenets of the American dream, since in their eyes it gives special privileges to some at the expense of others. The government formerly discriminated in favor of whites, and now it discriminates in favor of blacks -- those are equal violations of the ideology of the American dream and they should be equally prohibited.
To others, the American dream is not necessarily individualist. The ideology can be just as well interpreted to mean that groups have the right to pursue success collectively. Until a group succeeds according to its members’ own shared vision, the individuals within it cannot attain their dreams (alternatively, the pursuit of group success shapes and directs individuals’ dreams). In the eyes of those committed to group identity as a defining characteristic of social engagement, prior (or current) discrimination in favor of whites is not symmetrical with current proactive efforts to benefit blacks. Until the two races are equal in political power, economic means, social status, and cultural autonomy, equality of opportunity does not exist in the United States and calls for symmetry are a mere pretence for continued domination.
Just as principled opposition to affirmative action may reinforce or be a cover for a desire to retain white racial domination, principled support for affirmative action may similarly reinforce a desire to use public policies to benefit oneself or one’s group. After all, Americans (like most other humans, probably) have historically shown themselves adept at making normative arguments that coincide with their self-interest. There is no reason to suppose that middle-class African Americans are any different from anyone else in this regard.
Thus some Americans see strong forms of affirmative action as violating the values of individualism, equal opportunity, and meritocracy that instantiate the American dream. Some of them may have less respectable grounds for opposition that merge with those values. Others see strong forms of affirmative action as the only lever available to pry open the hypocritical claim of purported equality but actual racial domination that characterizes “Amerikka.” Some of them too may have other, less publicly compelling, reasons for support that merge with their understanding of the American dream. Where self-interest and ideology coincide so powerfully and in two such directly opposed camps, nuanced views do not flourish.
The clash between contending sets of interests and values is exacerbated by a growing perceptual gap between the races. On the one hand, the best-off African Americans -- those best poised to take advantage of the opportunities that affirmative action offers -- increasingly distrust white Americans’ racial values and practices. In the 1960s, poorly educated blacks were more likely than well-educated blacks to agree that “whites want to keep blacks down;” by the 1980s, the positions of the two groups were reversed (Hochschild 1995: 74). In 1990 and again in 1996, well-educated blacks agreed more than did poorly-educated blacks that “the Government deliberately makes sure that drugs are easily available in poor black neighborhoods in order to harm black people” (Hochschild 1995: 74 and analyses of New York Times/CBS News survey of October 13, 1996, in possession of author). Given a perception of intransigent and even growing white racism in the United States, affirmative action becomes a crucial weapon in well-off African Americans’ arsenal (Hochschild 1995, chapters 4-7; Bositis 1997; Gallup Organization 1997).
On the other hand, whites are increasingly convinced that racial discrimination is declining, and that blacks no longer suffer much from the effects of previous discrimination. In 1995, 55% of whites (compared with 29% of blacks) mistakenly agreed that “the average African American” is as well off as or better off than “the average white person” in terms of jobs and education. Over 40% of whites (and about 20% of blacks) held the same mistaken view with regard to housing and income (Washington Post et al. 1995). By the 1990s a majority of whites typically agreed that blacks have equal or greater opportunities than do whites to get ahead generally, to attain an education, to be admitted to college, and to get a job (Hochschild 1995: 60-64, Appendix B). They see the existence of the new black middle class as evidence to support that perception. And given that 15 % of whites agree with the very strong statement that “almost all of the gains made by blacks in recent years have come at the expense of whites,” it is especially striking that the “average American” estimates up to one‑third of Americans to be black and one‑fifth to be Latino (the real figures are 12 percent and 9 percent) (Hochschild 1995: 143; Gallup and Newport 1990; Nadeau et al. 1993; Washington Post et al. 1995: table 1.1).
Thus for some whites as well as for some middle-class blacks, perceptions unite with interests and values to create passionate convictions about affirmative action, regardless of whether it “works” or not. No wonder there is a culture war over an issue that directly affects only a small fraction of the American population.
Is There an Escape from the Impasse?
I noted earlier a broad band of rough consensus in the survey data -- quotas or preferences are bad (the Supreme Court agrees) but extra efforts to identify and train people who have been disadvantaged by race or gender is good (the Supreme Court agrees here also). That is a workable starting point for designing a policy that could achieve many of the purposes of affirmative action without generating so much hostility -- if people are susceptible to compromise on this issue.
Even those who voted for California’s Proposition 209 in the November 1996 election were not all strongly opposed to affirmative action, and even those who voted against it were not all strongly supportive. A Field Poll conducted several days after the vote found that fewer than half of opponents agreed that “affirmative action policies should not be changed.” A third of the opponents further agreed that “affirmative action policies should be relaxed somewhat.” Conversely, about a quarter of the proposition’s supporters agreed that affirmative action policies should not be changed, or should be relaxed but not eliminated (Lempinen 1996). Some of these voters may simply have been confused about what they were voting for. But many apparently would have preferred a middle ground to the two stark alternatives they were offered. Similarly, 65% of respondents to a national survey the day of the 1996 presidential election hoped that President Clinton would “put more emphasis on affirmative action to improve educational and job opportunities for women and minorities” in a second term (“Clinton’s Second Term” 1996: 1). In Houston, Texas, 55% of the voters in November 1997 rejected a proposition that would have banned “affirmative action” in city contracting and hiring (Verhovek 1997b). In all of these cases, we see the glimmering of a political context in which citizens’ opinions on affirmative action could come to stand for something other than racism or denial of meritocracy.
Similarly, many public officials and corporate officers seem wiser than either the activists or the academics. They focus more on the actual workings of affirmative action than the latter, and they find affirmative action more manipulable and less revolutionary (for better or for worse) than the former (Wolfe 1996; Hochschild 1998). Local public officials find that appointing a few well-connected black advocates has symbolic as well as substantive pay-offs in the next election. Corporate managers find that affirmative action has shifted from a pesky problem to a core management tool. A decade ago, problems of “equal employment opportunity” came in almost last (just above sexual harassment) on a list of executives' “human‑resource management issues;” by 1992, “cultural diversity” led corporate executives' list of “workforce concerns.” Almost two-thirds of companies surveyed by the Conference Board in 1991 offered diversity training to their employees, and most of the rest planned to do so soon (Schein 1986; Towers Perrin 1992:3; Wheeler 1994: 9; Glater and Hamilton 1995). The Conference Board now promotes conferences and publishes reports on “Managing Diversity for Sustaining Competitiveness” (Conference Board 1997).
We have, then, an electorate and many public and private officials who apparently seek a middle ground of “extra help” but not “reverse discrimination,” faced with a paired set of fiercely vocal activists who portray their opposites as either racist or unAmerican. For the former group, affirmative action is a policy with virtues and defects, appropriate interpretations and outrageous distortions -- much like any other policy. Its members have little use for academic analyses of the policy, but for the same reason that they have little use for academic analyses of virtually all policies; scholars are too slow, too methodologically driven, or too attuned to deep structures rather than useable facts. For the latter group, affirmative action is a symbol of deep racial claims and anxieties. Its members too have no use for empirical evidence, not because they find scholars irrelevant (often they are themselves scholars), but because the most careful analyses show that affirmative action seldom has dramatic impact, whether positive or negative, and that it frequently has unintended and ironic consequences.
At the most general level, affirmative action serves as a litmus test for anxieties about the meaning and validity of the American dream. Many African Americans fear, perhaps rightly, that the American dream was never meant to include them and can never be wrenched from its historical role of enabling some white men to legitimately dominate all other residents of the United States. In their eyes affirmative action instantiates a new and more sympathetic interpretation of the nation’s core ideology. It is also one of the only levers available to pry apart the tightly linked chains of meritocratic beliefs, personalistic practices, and structural biases in favor of the well-off. Many whites (especially but not only men) fear, perhaps rightly, that the American dream is losing its hold on Americans’ moral imagination. They see affirmative action as the tempter leading people to believe that what they do does not matter as much as who they are. For both groups, the devil is at the gate, and affirmative action is implicated in the defense of the city. Under these circumstances, we should not be surprised that evidence on how it actually works is irrelevant to all except those who have to put the policy into practice and live with the results.
Anderson, Bernard. 1996. “The Ebb and Flow of Enforcing Executive Order 11246.” American Economic Review. 86: 298-301.
Appiah, K. Anthony and Amy Gutmann. 1996. Color Conscious. Princeton: Princeton University Press.
Badgett, M.V. Lee. 1995. “Affirmative Action in a Changing Legal and Economic Environment.” Industrial Relations. 34: 489-506.
Badgett, M.V. Lee and Heidi Hartmann. 1995. The Effectiveness of Equal Employment Opportunity Policies. Washington D.C.: Joint Center for Political and Economic Research.
Basic and Applied Social Psychology. 1994 Special Issue on Affirmative Action. 15 (1 and 2).
Bates, Timothy and Darrell Williams. 1996. “Do Preferential Procurement Programs Benefit Minority Businesses?” American Economic Association Papers and Proceedings. 86: 294-7.
Bennett, Stephen Earl, Alfred Tuchfarber, Andrew Smith and Erick W. Rademacher. 1995. “Americans' Opinions of Affirmative Action.” Institute for Policy Research, University of Cincinnati. Unpublished paper.
Bergmann, Barbara.1996. In Defense Of Affirmative Action. New York: Basic Books.
Blanchard, P. A. and Faye J. Crosby, eds. 1989. Affirmative Action in Perspective. New York: Springer‑Verlag.
Blumrosen, Alfred W. 1995. Draft Report on Reverse Discrimination Commissioned by Labor Department: How Courts Are Handling Reverse Discrimination Cases. Daily Labor Report, March 23. Washington, D.C.: Bureau of National Affairs.
Bobo, Lawrence and James R. Kluegel. 1993. "Opposition to Race Targeting." American Sociological Review 58 (4): 443‑64.
Bobo, Lawrence and Ryan A. Smith. 1994. "Antipoverty Policy, Affirmative Action, and Racial Attitudes" in Confronting Poverty: Prescriptions for Change, edited by Sheldon Danziger, Gary Sandefur, and Daniel Weinberg. New York: Russell Sage Foundation and Harvard Univ. Press: 365‑95.
Bobo, Lawrence. 1998. “Race, Interests, and Beliefs about Affirmative Action,” American Behavioral Scientist. 41 (7): 985-1003.
Bositis, David. 1997. “1997 National Opinion Poll: Race Relations.” Washington D.C.: Joint Center for Political and Economic Studies.
Bowen, William and Derek Bok. 1998 The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions. Princeton: Princeton University Press.
Brewer, Dominic, Eric Eide, and Ronald Ehrenberg. 1999. “Does It Pay to Attend an Elite Private College?” Journal of Human Resources 34 (1): 104-123.
Broder, David and Robert Barnes. 1995. “Few Governors Join Attack on Racial Policies.” Washington Post, Aug. 2: A01ff.
Burstein, Paul. 1985. Discrimination, Jobs, and Politics. Chicago: University of Chicago Press.
Carter, Stephen. 1991. Reflections of an Affirmative Action Baby. New York: Basic Books.
Clayton, Susan D. and Faye J. Crosby. 1992. Justice, Gender, and Affirmative Action. Ann Arbor: University of Michigan Press.
“Clinton’s Second Term.” 1996. The Polling Report 12 (Nov.18): 1, 6.
Clotfelter, Charles. 1996. Buying the Best: Cost Escalation in Elite Higher Education. Princeton: Princeton University Press.
Collins, Sharon M. 1983. "The Making of the Black Middle Class." Social Problems 30: 369‑82.
---- 1997. Black Corporate Executives: The Making and Breaking of a Black Middle Class. Philadelphia: Temple University Press.
Conference Board. 1997. “Managing Diversity for Sustaining Competitiveness.” New York: Conference Board.
Cox, Taylor H. and Stacy Blake. 1991. "Managing Cultural Diversity: Implications for Organizational Competitiveness." Academy of Management Executive 5: 45‑56.
Crocker, Jennifer, Kristin Voellel, Maria Testa, and Brenda Major. 1991. “Social Stigma: the Affective Consequences of Attributional Ambiguity,” Journal of Personality and Social Psychology. 60 (2): 218-228.
Daniel, Kermit, Dan Black, and Jeffrey Smith, 1995. “College Quality and the Wages of Young Men,” University of Pennsylvania, Wharton School. Unpublished paper.
DelVecchio, Rick. 1996. “Cities Trying to Detour Past Prop. 209.” San Francisco Chronicle. Nov. 25: A1ff.
Dobbin, Frank, John Sutton, John Meyer, and W.R. Scott. 1993. "Equal Opportunity Law and the Construction of Internal Labor Markets." American Journal of Sociology 99 (2): 396‑427.
Dobbin, Frank. 1996. “Organizational Response to Affirmative Action and Equal Employment Law.” Memo to Conference on Social Science Perspectives on Affirmative Action, American Sociological Association, Washington D.C.
Eastland, Terry. 1996. Ending Affirmative Action: The Case for Colorblind Justice. New York: Basic Books.
Edelman, Lauren. 1990. "Legal Environments and Organizational Governance: The Expansion of Due Process in the Workplace." American Journal of Sociology 95 (6): 140-40
---- 1992. "Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights Law." American Journal of Sociology 97 (6): 1531‑76.
---- 1996. “Affirmative Action in Employment.” Memo to Conference on Social Science Perspectives on Affirmative Action, American Sociological Association, Washington D.C.
Edley, Christopher Jr. 1996. Not All Black and White: Affirmative Action and American Values. New York: Hill and Wang.
Enchautegui, Maria, Michael Fix, Pamela Loprest, Sarah von der Lippe, and Douglas Wissoker. 1996. “Do Minority-Owned Businesses Get a Fair Share of Government Contracts?” Policy and Research Report of the Urban Institute. Summer/Fall: 4-7.
“Excerpts from Round Table with Opponents of Racial Preferences.” 1997. New York Times. December 22: A24.
Farley, Reynolds. 1996. The New American Reality. New York: Russell Sage Foundation.
Finley, Steve. 1996. “Re: Being a Privileged White Male.” on Listserv Athenaemail@example.com. June 5.
Fried, Albert. 1997. McCarthyism : The Great American Red Scare. New York: Oxford University Press.
Gallup Organization. 1995. Affirmative Action: A Gallup Poll Special Report. Princeton: Gallup Organization.
---- 1997. The Gallup Poll Social Audit on Black/White Relations in the United States. Princeton: Gallup Organization.
Gallup, George Jr. and Frank Newport. 1990. "Americans Ignorant of Basic Census Facts." Gallup Poll Monthly, No. 294: 2-5.
Gamson, William. 1990. The Strategy Of Social Protest, 2nd ed. Belmont, CA.: Wadsworth Publishers
Gamson, William and Andre Modigliani. 1987. "The Changing Culture of Affirmative Action." Research in Political Sociology 3: 137‑77.
Gitlin, Todd. 1995. The Twilight of Common Dreams. New York: Metropolitan Books.
Glater, Jonathan and Martha Hamilton. 1995. “Affirmative Action’s Corporate Converts.” Washington Post. March 19: H1, H6
Glazer, Nathan. 1975. Affirmative Discrimination: Ethnic Inequality and Public Policy. New York: Basic Books.
Graham, Hugh Davis. 1999. “Since 1964: The Paradox of American Civil Rights Regulation,” in Taking Stock: American Government in the Twentieth Century, edited by Morton Keller and R. Shep Melnick. New York: Cambridge University Press: 187-218.
Griffin, Peter, Arthus Getis, and Ernst Griffin. 1996. “Regional Patterns of Affirmative Action Compliance Costs.” Annals of Regional Science. 30: 321-40.
Heckman, James and Brook Payner. 1989. “Determining the Impact of Federal Antidiscrimination Policy on the Economic Status of Blacks: A Study of South Carolina.” American Economic Review.79: 138-77.
Heckman, James and J. Hoult Verkerke. 1990. “Racial Disparity and Economic Discrimination Law: An Economic Perspective.” Yale Law and Policy Review. 8: 276-98.
Heckman, James and Kenneth Wolpin. 1976. "Does the Contract Compliance Program Work? An Analysis of Chicago Data." Industrial and Labor Relations Review 29: 544‑64.
Hochschild, Jennifer. 1995. Facing Up to the American Dream: Race. Class, and the Soul of the Nation. Princeton NJ: Princeton University Press.
--- 1998. “The Strange Career of Affirmative Action.” Ohio State Law Journal 59 (3): 997-1037.
Holmes, Steven. 1997. “Thinking About Race With a One-Track Mind.” New York Times. December 21, p. 1 sec. 4
Holzer, Harry 1998. “Employer Hiring Decisions and Antidiscrimination Policy.” In Generating Jobs: How to Increase Demand for Less-Skilled Workers, edited by Richard Freeman and Peter Gottschalk. New York: Russell Sage Foundation: 223-257.
Holzer, Harry and David Neumark. 1999. "Are Affirmative Action Hires Less Qualified? Evidence from Employer‑Employee Data on New Hires." Journal of Labor Economics 17 (3): 534-569.
Holzer, Harry and David Neumark. 2000. "Assessing Affirmative Action.” Journal of Economic Literature 38: 483-568.
House of Representatives, Government Operations Committee, U.S. Congress. 1994. Problems Facing Minority and Women-Owned Small Business Including SBA Section 8(a) Firms in Procuring U.S. Government Contracts: An Interim Report. HRPT. NO. 103-870. Washington D.C.: U.S. Government Printing Office.
Kahlenberg, Richard. 1996. “Bob Dole’s Colorblind Injustice.” Washington Post. June 2: C1, C4.
Kane, Thomas. 1998. “Racial and Ethnic Preferences in College Admissions.” In The Black-White Test Score Gap, edited by Christopher Jencks and Meredith Phillips. Washington D.C.: Brookings Institution: 431-56.
Karabel, Jerome 1993. “Berkeley and Beyond.” American Prospect 12: 156-60.
Kelly, Erin and Frank Dobbin. 1998 “How Affirmative Action Became Diversity Management.” American Behavioral Scientist 41 (7): 960-84
Kinder, Donald and Lynn Sanders. 1996. Divided by Color. Chicago: University of Chicago Press.
Kluegel, James and Eliot Smith. 1983. "Affirmative Action Attitudes: Effects of Self‑interest, Racial Affect, and Stratification Beliefs on Whites' Views." Social Forces 61: 797‑825.
Konrad, Alison and Frank Linnehan. 1995a. "Formalized HRM Structures: Coordinating Equal Employment Opportunity or Concealing Organizational Practices?" Academy of Management Journal 38: 787‑820.
Ladd, Everett, ed. 1998. America at the Polls 1998. Storrs CT: University of Connecticut, Roper Center.
Langfitt, Frank. 1995. “GOP Abandons Effort on Minority Set-Asides.” Baltimore Sun. May 23: 1B.
Larana, Enrique, Hank Johnston, and Joseph Gusfield, eds. 1994. New Social Movements: From Ideology To Identity. Philadelphia PA: Temple University Press.
Leadership Conference on Civil Rights. 1997. Affirmative Action for Women and Minorities: It Works Well, It Is Necessary, and It Benefits America. Washington D.C.: Leadership Conference on Civil Rights.
Lempinen, Edward. 1996. “Affirmative Action Foes Span Spectrum.” San Francisco Chronicle.
Nov. 22: B1ff.
Leonard, Jonathan 1984a. "The Impact of Affirmative Action on Employment." Journal of Labor Economics 2: 439‑63.
---- 1984b. "Employment and Occupational Advance under Affirmative Action." Review of Economics and Statistics 66: 377‑85.
---- 1984c. "Anti‑discrimination or Reverse Discrimination: The Impact of Changing Demographics, Title VII, and Affirmative Action on Productivity.” Journal of
Human Resources 19 (2): 145‑74.
---- 1986. “What Was Affirmative Action?” American Economic Review 76: 359‑63.
---- 1990. "The Impact of Affirmative Action Regulation and Equal Employment Law on Black Employment." Journal of Economic Perspectives 4: 47‑63.
---- 1996. "Wage Disparities and Affirmative Action in the 1980s." American Economic Association Papers and Proceedings 86: 285‑9.
Lind, Michael. 1995. The Next American Nation: The New Nationalism and the Fourth
American Revolution. New York: Free Press.
Lynch, Frederick R. 1989. Invisible Victims: White Males and the Crisis of Affirmative Action. New York: Greenwood.
Meyers, Marvin. 1957. The Jacksonian Persuasion: Politics and Belief. Stanford CA: Stanford University Press.
Morin, Richard. 1995. “Affirmative Action for White Guys?” Washington Post. Oct. 22: C5.
Motley, Constance Baker. 1998. Equal Justice Under Law. New York: Farrar, Straus, and Giroux.
Myers, Samuel and Tsze Chan. 1996. “Who Benefits from Minority Business Set-Asides? The Case of New Jersey.” Journal of Policy Analysis and Management 15 (2): 202-26.
Nacoste, Rupert. 1985. “Selection Procedure and Responses to Affirmative Action,” Law and Human Behavior 9 (3): 225‑42.
---- 1994. “If Empowerment Is the Goal...: Affirmative Action and Social Interaction,” Basic and Applied Social Psychology 15: 87‑112.
Nadeau, Richard, Richard Niemi, and Jeffrey Levine. 1993. "Innumeracy About Minority Populations." Public Opinion Quarterly 57 (3): 332-47.
National Council for Research on Women. 1996. “Affirming Diversity: Building a National Community that Works.” Issues Quarterly. 1: 1-28.
Ness, Carol. 1996. “Prop. 209 Wins, Heads for Courts.” San Francisco Examiner. Nov. 6: A1ff.
Onishi, Norimitsu. 1996. “Affirmative Action: Choosing Sides.” New York Times, Section on “Education Life.” March 31: 26-35.
Oreskes, Michael. 1984. “The Set-Aside Scam.” New Republic. Dec. 24: 17-20.
Pettigrew, Thomas and Joanne Martin. 1987. "Shaping the Organizational Context for Black American Inclusion," Journal of Social Issues 43 (1): 41‑78.
Ponessa, Jeanne. 1997. “Higher Ed. Outreach Plan Targets At-Risk Calif. Youths.” Education Week. June 4: 5.
Preston, Michael and James Lai. 1998. “The Symbolic Politics of Affirmative Action.” In Racial and Ethnic Politics in California, vol. 2, edited by Michael Preston, Bruce Cain, and Sandra Bass. Berkeley CA: University of California, Institute of Governmental Studies Press: 161-98.
Pulley, Brett. 1996. “Affirmative Action: Can Whitman Stand Firm?” New York Times. Nov. 24: New Jersey section, p. 2
Reskin, Barbara. 1998. The Realities of Affirmative Action in Employment. Washington D.C: American Sociological Association.
Rice, Mitchell and Maurice Mongkuo. 1998. “Did Adarand Kill Minority Set-Asides?” Public Administration Review. 58: 82-86
Rodgers, William III and William Spriggs, 1996. “The Effect of Federal Contractor Status on Racial Differences in Establishment-Level Employment Shares: 1979-1992.” American Economic Review. 86: 290-93.
Rosenfeld, Michel. 1991. Affirmative Action and Justice: A Philosophical and Constitutional Inquiry. New Haven CT: Yale University Press.
Schein, Lawrence. 1986. “Current Issues in Human‑Resource Management.” Research Bulletin (of The Conference Board), no. 190.
Schrecker, Ellen. 1986. No Ivory Tower: McCarthyism and the Universities. New York: Oxford University Press.
Schuman, Howard, Charlotte Steeh, Larry Bobo, and Maria Krysan. 1997. Racial Attitudes in America, revised ed. Cambridge: Harvard University Press.
Skerry, Peter. 1997. “The Strange Politics of Affirmative Action.” Wilson Quarterly. 21 (winter): 39-46.
Skrentny, John. 1996. The Ironies of Affirmative Action. Chicago: University of Chicago Press.
Sleeper, Jim. 1997. Liberal Racism. New York: Viking Press.
Smith, James and Finis Welch. 1984. "Affirmative Action and Labor Markets." Journal of Labor Economics 2: 269‑301.
--- 1989. “Black Economic Progress after Myrdal.” Journal of Economic Literature. 27: 519-64.
Smith, Rogers. 1997. Civic Ideals: Conflicting Visions of Citizenship in U.S. History. New Haven CT: Yale University Press.
Sniderman, Paul and Thomas Piazza. 1993. The Scar of Race. Cambridge MA: Harvard University Press.
Steeh, Charlotte and Maria Krysan. 1996. "The Polls‑‑Trends. Affirmative Action and the Public, 1970‑1995." Public Opinion Quarterly 60 (1): 128‑58.
Steele, Claude. 1999. “A Threat in the Air: How Stereotypes Shape Intellectual Identity and Performance.” In Promise and Dilemma: Perspectives on Racial Diversity and Higher Education, edited by Eugene Lowe. Princeton: Princeton University Press: 92-128.
Stephanopoulos, George and Christopher Edley, Jr. 1995. Affirmative Action Review. Unpublished report to the President. Washington, D.C.
Sutton, John, Frank Dobbin, John Meyer and W. Richard Scott. 1994. "The Legalization of the Workplace." American Journal of Sociology 99 (4): 944-71.
Taylor, Bron. 1991. Affirmative Action at Work. Pittsburgh PA: University of Pittsburgh Press.
Taylor, Marylee C. 1994. "Impact of Affirmative Action on Beneficiary Groups: Evidence from the 1990 General Social Survey." Basic and Applied Social Psychology 15: 143‑78.
Thernstrom, Stephan and Abigail Thernstrom (1997) America in Black and White: One Nation, Indivisible. New York: Simon and Schuster.
Thomas, David and Robin Ely. 1996. “Making Differences Matter: A New Paradigm for Managing Diversity.” Harvard Business Review. Sept/Oct.: 79-90.
Tien, Chang-Lin. 1999. “What a University Can Learn and Teach about Conflict and Difference.” In Promise and Dilemma: Perspectives on Racial Diversity and Higher Education, edited by Eugene Lowe. Princeton: Princeton University Press: 193-198
Towers Perrin.1992. Workforce 2000 Today: A Bottom‑Line Concern. New York: Towers Perrin.
“Tuition Trends.” 1997. Chronicle of Higher Education. May 30: A11.
Verhovek, Sam. 1997a. “Houston to Vote on Repeal of Affirmative Action,” New York Times. November 2, p. 28.
--- 1997b. “Referendum in Houston Shows Complexity of Preferences Issue.” New York Times November 6, pp. A1, A26.
Washington Post, Kaiser Family Foundation, Harvard University. 1995. The Four Americas: Government and Social Policy Through the Eyes of America’s Multi-racial and Multi-ethnic Society. Washington D.C.: Washington Post.
Washington State Commission on African American Affairs. 1995a. Affirmative Action: Who’s Really Benefiting? Part 1: State Employment. Olympia WA: Washington State Commission on African American Affairs.
---- 1995b. Affirmative Action: Who’s Really Benefiting? Part 1I: Public Higher Education. Olympia WA: Washington State Commission on African American Affairs.
---- 1995c. Affirmative Action: Who’s Really Benefiting? Part 1II: Contracting in State Government. Olympia WA: Washington State Commission on African American Affairs.
Wheeler, Michael. 1994. Diversity Training. New York: Conference Board.
Wolfe, Alan. 1996. “Affirmative Action, Inc.” New Yorker. Nov. 25: 106-15.
Woodson, Robert. 1996. “Affirmative Action Is No Civil Right.” Harvard Journal of Law and Public Policy. 19: 773-78.
. For example, Edley (1996) devotes twelve of the 280 text pages of his book to “facts” about America’s racial situation (and of them, only three pages address the effectiveness of affirmative action policies). A special issue on affirmative action of the National Council for Research on Women (1996) devotes one of its twelve substantive pages to “the evidence.”
The main counterevidence to my claim is, of course, The Shape of the River by William Bowen and Derek Bok. I will discuss it below; for now, it is important to note only that it was published fairly recently (1998), that it was not written by research scholars as that term is usually understood, and that it stands out as a dramatic and unique exception to the circumstances I describe in the text.
. The past decade and a half have seen important judicial decisions on the extent of and reasons for legally acceptable affirmative action. These include City of Richmond v. J. A. Croson Co. (1989), Adarand Constructors, Inc. v. Peña (1995), and Hopwood v. University of Texas (1996). These decisions, however, did more to specify (mostly to narrow) the conditions under which affirmative action may be used and to define more precisely what the term itself may legitimately mean than to change the nature of the underlying debate over the need for affirmative action.
 Note the obvious but important point that there is no single entity or process called “affirmative action;” the term is used to mean everything from efforts to ensure a broad applicant pool to strict quotas. Both proponents and opponents play verbal games with the term.
 Another caveat: individual institutions and agencies sometimes analyze their own practices, and occasionally compare their own with similar institutions’ practices. But these analyses are seldom made public, and by definition are not conducted by neutral evaluators. Policy analysts have known for decades that it is virtually impossible for an organization to evaluate its own practices in ways that are uncontaminated by internal politics or by the goals of the evaluators, no matter how hard they try. So even if these analyses exist, they do not count much against the claim in the text.
 Many journalists have made some variant of this claim, but researchers have not investigated it carefully (Broder and Barnes 1995; Langfitt 1995; DelVecchio 1996; Pulley 1996; Kahlenberg 1996).
 See Hochschild (1998) for one effort to answer some of these questions.
 Half of women (compared with 57% of men) and 45% of Asian-Americans (compared with 60% of whites, 26% of blacks, and 30% of Latinos) voted for Proposition 209 (the California Civil Rights Initiative) in November 1996 (Ness 1996). White men and women supported the anti-affirmative action proposition more strongly than did men or women of color. In Washington, 65% of men, and 50% of women supported Initiative 200, which was modeled after Proposition 209. Almost 60% of whites, and 46% of nonwhites (almost all of whom are Hispanic or Asian American) similarly supported the initiative (Ladd 1998: 129). Onishi (1996) provides a good analysis of the Asian-American college students’ ambivalent views about affirmative action.
 Richmond is not alone. In San Francisco, those eligible for preference in bidding on city contracts include descendents from all nations in Asia, the Pacific islands, Africa, and Latin America, as well as Asian Indians, Arab Americans, Native Americans, women of all races, and locally owned businesses. In 1999, the city government considered adding Iranian Americans to the list.
. To give only one example of the many observations similar to my own: “affirmative action’s detractors have generally sidestepped time-consuming, substantive research to verify their suppositions about who actually participates in or benefits from such programs. Even scholarly opponents of affirmative action have resorted to emotionally-charged rhetoric without offering clear evidence of their claims of reverse discrimination” (Washington State Commission on African American Affairs 1995a: 10). I would simply add that advocates have done little more.
 The argument in this paper extends and develops my analysis in Hochschild 1995. That book eschewed discussion of specific policy or political issues, since it focussed on variations within and across broad ideologies, especially the ideology of the American dream. Disputes over the meaning of the American dream, however, are often played out in the political and policy arena; I argue below that affirmative action is one of the most important sites for such a dispute.
. Holzer and Neumark (1999) observe that “despite the intensity of the viewpoints held, the evidence to date on this issue [specifically, the underqualification of affirmative action beneficiaries] remains quite thin.” I agree with them on the intensity and the thinness; in my view, however, the evidence is thin because of rather than despite the intensity of the viewpoints.
 On the Bank War, see Meyers 1957; on McCarthyism, see Fried 1997, and Schrecker 1986. On the social construction of crises more generally, see Larana et al. 1994, and Gamson 1990.
. Two comprehensive recent surveys of empirical research on affirmative action in employment are Reskin 1998 and Holzer and Neumark 2000. The latter point to many “shortfalls between what we do know and what we would like to know,” but points to a “growing literature that… begins to ask and answer some of the right questions” (p. 484). They are speaking explicitly, however, only from the perspective of economists.
Readers of an earlier version of this chapter pointed out that “much of the debate over affirmative action has taken place in the legal arena.... [Therefore] the lawyers and their experts build a record to support or oppose an affirmative action policy.... Moreover,... many public agencies and other organizations that receive federal assistance have conducted searching reviews of their affirmative action policies and extensive fact-finding” (personal communication from Marianne Engelman Lado and Luke Harris). They are, of course, correct, and this huge subterranean literature would be a gold mine for researchers seeking to make comparisons and draw generalizations about how and why affirmative action functions. But part of my point is the fact that researchers have seldom dug into this gold mine. And in any case each analysis in this literature is by definition biased in favor of or against affirmative action.
. From here on, I will focus mostly on race-based rather than gender-based affirmative action, simply because I know more about the former than the latter. As I understand it, the evidence is roughly similar with regard to outcomes for white women as it is for black women and men. The symbolic politics of affirmative action, however, probably differ significantly between race and gender, and it would be illuminating to develop their differences in more detail.
. For example, “between 1970 and 1990, the number of black electricians more than tripled (from 14,145 to 43,276) and the number of black police officers increased almost as rapidly (from 23,796 to 63,855)” (Karabel 1993: 159).
. However, in Washington State “whites [including men as well as women] are the primary beneficiaries of the state’s affirmative action program affecting hiring,...of special admissions programs at public institutions of higher learning,... [and] of programs designed to assist minority- and women-owned firms seeking to do business with the state” (Washington State Commission on African American Affairs, 1995a: 1; see also ibid. 1995b and 1995c). We do not know how extensively or when these counterintuitive results hold.
. However, Holzer (1998: 225) points out that “the magnitudes of the changes appear to be somewhat larger among white-collar than blue-collar occupations.” See also Holzer and Neumark 1999.
. Echoing my complaint above, Holzer (1998: 243) points out that “we still have too little evidence on whether these programs are efficient, or whether they generate any net new employment for less educated workers.” Firms subject to affirmative action regulations experience an increase of about 5% in labor costs compared with firms not so subject, according to the only set of scholars who have studied the issue (Griffin et al.1996).
. Average undergraduate tuition charges in private four-year colleges rose from under $4000 in 1980 to almost $13,000 in 1996-97. Tuition charges in public four-year colleges rose during the same period from about $1000 to $3000 (“Tuition Trends” 1997). For more detail, see Clotfelter 1996).
 Would-be students who are not admitted to elite universities have stronger grounds than almost anyone else for opposing affirmative action. But their claim of harm or injustice is weak for two reasons. First, the overwhelming majority of applicants would not be admitted to Harvard or Princeton even if every beneficiary of affirmative action were denied a slot in the freshman class. (About 12% of applicants to Princeton or Harvard are admitted.) Second, no one has a right to admission to an elite private (or public) university or to a given job; one may be deeply disappointed by rejection, but one cannot claim unjust treatment.
 Over four in ten Americans (41% of whites and Asians, 62% of blacks, and 57% of Latinos) perceive that “white men are generally covered under federal affirmative action” (Morin 1995). Legally they are correct. But I think it fair to assume that in the context of public opinion surveys, this result indicates ignorance of how affirmative action normally operates rather than subtle knowledge of the law.
 Skerry (1997) gives a somewhat similar analysis.
 As President Clinton put it rather plaintively in his conversation with a group of racial conservatives, “I’d like to… hear from you… on the question of, ‘Do you believe that race… is still a problem in some ways?’ And if so, instead of our getting into a big fight about affirmative action….” His plea was largely ignored; most responses to his and Vice-President Gore’s questions came back to a statement of opposition to preferences (“Excerpts from Round Table…”).
 Comment by Christopher Edley, in Holmes 1997: 1. Edley goes on to observe, “You can look at those figures [on how many are affected by federal contracts and elite college admissions] and ask, why are black folks making such a big deal out of affirmative action? That’s the wrong question. The real question is why are white folks making such a big deal out of it?”
 In 1942, only 32% of Americans agreed that white and black students should attend the same schools, and only 46% opposed “separate sections for Negroes in streetcars and buses.” As late as 1963, fewer than half of Americans agreed that they would vote for a black presidential candidate of their political party even if he were qualified. Most importantly for this paper, in 1944 and again in 1946, fewer than half of white Americans agreed that “Negroes should have as good a chance as white people to get any kind of job.” (The alternative response was “white people should have the first chance at any kind of job” (Schuman et al. 1997: 104 [all responses are percentages of those giving a substantive answer]).
 This is the spelling of “America” sometimes used by strong critics from the left. I do not know its provenance; it may be intended to evoke as association of the United States with the Ku Klux Klan.
 The 1990 survey was of residents of New York City; the 1996 survey was a national random sample.
 If I were made race czar, I would work to strengthen at least the “soft” forms of affirmative action. In the long run, however, it would be more valuable for the state of California, for example, to engage in structural reforms of inner city schools rather than merely allowing U.C. Berkeley to pick out the few black or Latino students who have somehow triumphed over the wasteland that has demoralized all of their fellow students. The university has taken tentative steps in that direction, in response to the brutal abolition of affirmative action through Proposition 209 (Tien 1999; Ponessa 1997).
 Polls prior to the election showed that two-thirds of Houston’s voters would have supported a proposition to “not discriminate against or grant preferential treatment to” any person on group based on race, sex, or ethnicity (Verhovek 1997a). The two propositions would have had identical effects; all the difference lay in the wording or, in my terms, in which variant of the American dream is invoked by the fluid concept of affirmative action.