Affirmative Action: Legislative History, Judicial Interpretations, Public Consensus
Affirmative action is often considered to be a public-policy issue on which Whites and Blacks are hopelessly divided (Delgado, 1996; Hacker, 1992; Kinder and Sanders, 1996; Thernstrom and Thernstrom, 1997). Racial division and polarization, however, do not tell the whole story. Once we move beyond the ambiguity surrounding the term “affirmative action”—and the confusion concerning existing affirmative-action programs—a good deal of agreement is revealed among Whites, Blacks, and members of other racial and ethnic groups concerning many affirmative action-related issues. “Agreement” includes a shared unease about programs involving overt racial preferences coupled with a willingness to support outreach programs as well as programs that benefit the disadvantaged, and certain other types of affirmative-action initiatives. Identifying and building on this agreement and consensus is a necessary first step in the development of any successful race-related public policy in a multiracial society, such as our own.
DEFINING THE CONCEPT
In 1995, President Clinton appointed investigators to review federal policy on affirmative action. An important and telling finding the investigators reported was that affirmative action had no clear and widely understood definition, and that this contributed to an atmosphere of confusion and miscommunication about affirmative action’s goals and modes of implementation (Edley, 1996:15–24; see also Smelser, 1999). Neverthe-
less, the investigators offered a useful definition of affirmative action as “any effort taken to expand opportunity for women or racial, ethnic and national origin minorities by using membership in those groups that have been subject to discrimination as a consideration [in decision making or allocation of resources]” (Edley, 1996:16–17). Swain (1996:1) offers a somewhat broader definition is offered that involves a “range of governmental and private initiatives that offer preferential treatment to members of designated racial or ethnic minority groups (or to other groups thought to be disadvantaged), usually as a means of compensating them for the effects of past and present discrimination.” Both of these definitions suggest a compensatory rationale for affirmative-action programs—i.e., members of groups previously disadvantaged are now to receive the just compensation that is their due in order to make it easier for them to get along in the world. Other useful definitions and characterizations of affirmative action would de-emphasize the retrospective, compensatory, and ameliorative nature of such programs and focus, instead, on the current value of such programs in enhancing diversity, particularly in educational institutions and in the workforce (Bergman, 1996; Bok and Bowen, 1998).
The actual programs that come under the general heading of affirmative action are a diverse lot; they include policies affecting college and university admissions, private-sector employment, government contracting, disbursement of scholarships and grants, legislative districting, and jury selection. Numerous affirmative-action programs have been enacted into law at local, state, and federal levels. In addition to programs that have been mandated by law, many private corporations and universities have developed affirmative-action programs voluntarily. Methods of implementing affirmative-action policies are similarly diverse and in the past have ranged from “hard quotas” to softer methods of outreach, recruitment, and scrupulous enforcement of antidiscrimination norms. Clearly, much of the ambiguity and conceptual fuzziness surrounding affirmative action as a theoretical idea occurs because there are differing forms of the policy and even more differing modes of implementation (Swain et al., 2001).
EVOLUTION OF AFFIRMATIVE-ACTION POLICY
Many scholars have traced the historical evolution of affirmative-action policy (Belz, 1991; Drake and Holsworth, 1996; Graham, 1990; Skrentny, 1996). Presented here is a very cursory overview of the history of the term in U.S. legislation, major court cases that helped shape public opinion, and analyses of recent public opinion surveys documenting current attitudes about affirmative action.
The actual term “affirmative action” emerged first in labor law in the
1935 National Labor Relations Act (Wagner Act), but it did not become firmly associated with Civil Rights legislation enforcement until 1961, the year President Kennedy issued Executive Order 10925. That order directed federal contractors to take “affirmative action” to ensure nondiscrimination in hiring, promotions, and all other areas of private employment. The concept of affirmative action, however, was not formally defined in the Executive Order, and it went largely unnoticed.
Kennedy’s executive order, nevertheless, represented considerable progress over the Civil Rights Acts passed in 1957 and 1960. Unlike the latter, Executive Order 10925 had a real effect in reducing discrimination in at least one area—government contracting. Kennedy’s order, however, failed to address many of the concerns of the National Association for the Advancement of Colored People (NAACP) and other civil rights organizations regarding discrimination in public accommodations, housing, government employment, and private-sector employment in firms that did not have contracts with the federal government. In the face of increased White-initiated violence in the South, and mounting pressure from civil rights leaders, Kennedy appealed to Congress to pass legislation mandating equal rights and equal access to all in all public accommodations and jobs in what was, at the time, hailed as “the civil rights bill of the century.” The Kennedy-proposed Civil Rights legislation, however, was too drastic for a Congress that was still, to a considerable extent, controlled by southern Democrats. It took the combination of the sympathy generated by Kennedy’s assassination in November 1963, the formidable legislative skills of President Lyndon B.Johnson, and the pleas and reassurances of Majority Whip Hubert Humphrey (D-Minnesota) to persuade Congress to pass a comprehensive Civil Rights bill.
Even before the concept of affirmative action was widely in use, opponents of the 1964 Civil Rights legislation argued that the lack of a definition of what constituted discrimination would lead to quotas in employment. Senator James Eastland (D-Mississippi) argued forcefully that the bill would discriminate against White people: “I know what will happen if there is a choice between hiring a White man or hiring a Negro, both having equal qualifications. I know who will get the job. It will not be the White man” (Swain, 1996:6). Largely because of such fears and predictions, the Civil Rights bill was amended explicitly to ban quotas in hiring.
When the bill eventually passed as the Civil Rights Act of 1964, it was comprehensive in its scope. In addition to its employment provisions, it banned discrimination in, for example, public accommodations, public conveyances, theaters, and restaurants; and it authorized the government to withhold federal funds from schools that had not desegregated in com-
pliance with the 1954 Brown decision. In all, it contained 11 sections or titles. Title VI and Title VII are most important to the evolving connection between Civil Rights enforcement and affirmative action. Title VI covers discrimination in federally assisted programs, and Title VII covers employment discrimination in all large and medium-sized private businesses. Congress created the Equal Employment Opportunity Commission (EEOC) to monitor Title VII violations, but the original legislation gave the EEOC no power to enforce its dictates. A year later the Labor Department established the Office of Federal Contract Compliance (OFCC, subsequently reorganized and renamed the Office of Federal Contract Compliance Programs, OFCCP), which was charged with regulating federal grants, loans, and contracts.
Far from promoting preferential treatment, Title VII, as originally written and interpreted, merely required employers and admissions officers to stop discriminating. If this was done in an acceptable manner, then businesses were considered to be in compliance with the law. Title VII, however, in combination with related executive orders and key administrative and court decisions, set the stage for stronger affirmative-action measures by requiring businesses doing business with the federal government to compile statistical data about the race and sex of their employees and job applicants in order to demonstrate their compliance with the law (Drake and Holsworth, 1996; Glazer, 1975).
Both Presidents Lyndon Johnson and Richard Nixon were instrumental in moving the country away from “soft” affirmative-action programs, that merely required employers and other private parties to make special efforts to recruit members of previously excluded groups, toward stronger policies mandating preferential treatment of women and minorities. Under these presidents, federal goals began to shift from equal opportunity as defined in the Civil Rights legislation to an emphasis on equal (or proportional) results.
On the specific issue of affirmative action for Blacks, Johnson set the stage for more aggressive, results-oriented policies. In a commencement address at Howard University, in June 1965, President Johnson introduced the powerful image of a shackled runner, which later influenced much of the debate concerning affirmative action:
You do not take a man who, for years, has been hobbled by chains, liberate him, bring him to the starting line of a race saying, “You are free to compete with all the others,” and still believe you have been fair. This is the next and more profound stage of the battle for civil rights. We seek not just freedom of opportunity, not just legal equity, but human ability; not just equality as a right and theory, but equality as a right and result (Johnson, 1965).
Also in 1965, Johnson issued Executive Order 11246, which reaffirmed support for Kennedy’s 1961 order linking Civil Rights enforcement with “affirmative action” requirements.
In the aftermath of the bloody Newark and Detroit riots in the summer of 1967, Johnson issued Executive Order 11365, which established a National Advisory Commission on Civil Disorders; it came to be known as the Kerner Commission after its chairman, Otto Kerner, then Governor of Illinois. The Commission was composed largely of White and Black moderates and was directed to answer three basic questions about the riots: What happened? Why did it happen? What can be done to prevent it from happening again? The Commission’s report would attribute most of the problems of the inner-city ghettos to racism among Whites.
Paradoxically, it was the conservative President Richard Nixon who was initially responsible for the adoption of racial quotas, the strongest affirmative-action enforcement mechanism. Nixon’s endorsement of quotas came in the early 1970s when his administration approved the Philadelphia Plan, which involved the direct imposition of hiring goals in the construction trades, to be monitored through OFCC. The U.S. Department of Labor (USDL) subsequently issued Revised Order #4 that, in effect, amended Johnson’s Executive Order 11246 by extending the quota-like features of the Philadelphia Plan to all private contractors doing business with the federal government. Contractors were now required to establish target “goals and timetables” for the hiring of “underutilized” minority group members and women, and to show “good faith efforts” to meet these hiring goals and timetables. Although not a rigid quota system, the “goals and timetables” requirement was a results-oriented approach to employment policy that its critics would charge operated, in practice, little differently than a quota system. Labor Department staff supported the new “goals and timetables” approach as a way of ending old-style discrimination and securing positive results for previously excluded groups (Skrentny, 1996). Because Nixon had previously opposed legislative initiatives thought to be beneficial to Blacks and other minorities, there was some speculation as to why he would endorse the use of quotas. One theory is that he’d hoped to sow dissention and division among core Democratic groups (Kahlenberg, 1996:22).
Thus, affirmative action evolved from a vague concept buried in an executive order, to a set of legal regulations and practices. The shift from “weak” to “strong” methods of policy enforcement, it is important to recognize, was largely the result, not of legislative action, but of decisions made in the executive branch of the federal government and in federal regulatory agencies such as the USDL and the U.S. Department of Health, Education and Welfare. Partly as a result of the riots of the late 1960s, many policy makers had come to view stronger measures than race-
neutral nondiscrimination as necessary, at least as a transitional step, toward the goal of achieving a color-blind society, and as a necessary means of ensuring that Whites would not discriminate against Blacks (Belz, 1991; Graham, 1990).
THE ROLE OF THE COURTS IN EXPANDING, AND LATER RESTRICTING, AFFIRMATIVE ACTION
The U.S. Supreme Court rendered a crucial decision in the case of Griggs v. Duke Power Co. (401 U.S. 424, 1971), an early employment discrimination case, in which the Court ruled unanimously that under Title VII of the 1964 Civil Rights Act, any screening device that produced unequal consequences for different races—i.e., what in employment law came to be known as “disparate impact” in the sense of disproportionate group harm—would be held to constitute invidious employment discrimination unless the screening device were shown to be clearly job-related. Four years after Griggs, the Court reaffirmed its support for the disparate impact approach in Albemarle Paper Company v. Moody (422 U.S. 405, 1975), a case in which employers sought to protect themselves against discrimination charges by hiring enough minorities to counteract any statistical charges of racial imbalance.
Nevertheless, a year later in Washington v. Davis (426 U.S. 229, 1976), the Court refused to extend the theory of disparate impact developed in Title VII cases to discrimination cases brought under the equal protection provisions of the U.S. Constitution. The Court held that to establish a constitutional claim of unequal treatment (as opposed to a statutory claim of discrimination under Title VII), there had to be a showing of intent to discriminate, not simply a showing of disparate impact. The case in question involved unsuccessful applicants for positions of policemen in the District of Columbia, who charged that the police department’s use of a written test was discriminatory because Blacks had a disproportionately high failure rate. At the time the case was brought before the federal courts, Title VII, which would later be extended in scope, did not cover municipal employees.
In 1989, the Court, reflecting the more politically conservative influence of justices appointed by President Ronald Reagan, shifted the burden of proof in disparate impact cases from businesses to plaintiffs, making it more difficult to sustain an employment discrimination claim under Title VII. The switch occurred in the Court’s ruling in Ward’s Cove Packing Co. v. Atonio (490 U.S. 642, 1989), and Price Waterhouse v. Hopkins (490 U.S. 228, 1989). The Court in Ward’s Cove ruled that “a simple statistical com-
parison of racial percentages between skilled and unskilled jobs was insufficient to make a prima facie case” of employment discrimination (Hall, 1992:21). In Price Waterhouse the Court shifted the burden even further by requiring the plaintiff to prove that “employment practices substantially depended on illegitimate criteria” (Hall, 1992:351). However, the 1991 Civil Rights Act, which was passed by a Democratic congress and, with some reluctance, signed into law by President Bush, overturned Ward’s Cove by limiting the ability of employers to use “business necessity” as a defense against discrimination claims under Title VII. It also overruled the Court’s decision in Patterson v. McLean Credit Union (491 U.S. 164, 1989), where the Court had invalidated a Black woman’s attempt to seek relief from racial harassment under the 1866 Civil Rights Act.
Charges of “reverse discrimination” became common during the 1970s, as more and more corporations and private businesses, often under pressure from federal enforcement agencies, began more aggressive hiring of minorities and women. The question of whether Title VII of the Civil Rights Act also protected Whites against discrimination arose in McDonald v. Sante Fe Transportation Company (427 U.S. 273, 1976). The case involved a company where two White employees who had been charged with theft were fired, while a Black employee similarly charged had been retained. The Court ruled unanimously that Whites as well as Blacks are protected from racial discrimination under the antidiscrimination provisions of Title VII.
Despite this ruling, a number of subsequent court decisions would hold that Title VII permitted the preferential treatment of minorities and women in hiring and promotion decisions (but not in decisions affecting layoffs) if such treatment were part of an affirmative-action plan designed to increase the employment of previously excluded or underrepresented groups. Perhaps the most important of these decisions came in the case of United Steelworkers of America v. Weber (443 U.S. 193, 1979). The Weber case involved a White, blue-collar worker (Brian F.Weber) who was refused admission to an on-the-job training program at the Kaiser Aluminum Company plant at which he worked, although he had higher seniority than some of the minority workers who were accepted. In an attempt to increase minority representation in its workforce, Kaiser Aluminum had developed two seniority lists, one for Whites and one for Blacks, and it filled its vacancies by selecting persons from the top of each list. Weber filed suit, claiming that the 1964 Civil Rights Act specifically prohibited this use of racial quotas. Although winning at the district court level, Weber lost in the Supreme Court, which claimed that Title VII, though not requiring race-conscious affirmative-action preferences in employment, nevertheless permitted them if the purpose was to increase the employ-
ment of groups previously discriminated against. “It would be ironic indeed,” Justice Brennan wrote in the majority decision, “if a law triggered by a nation’s concern over centuries of racial injustice and intended to improve the lot of those who had ‘been excluded from the American dream for so long’ constituted the first legislative prohibition of all voluntary, private, race-conscious efforts to abolish traditional patterns of racial segregation and hierarchy” (Brennan at 443 U.S.: 193).
Quota-like employment practices were also upheld by the Court in Local 28 Sheet Metal Workers International Association v. Equal Employment Opportunity Commission (478 U.S. 421, 1986), where in a five-to-four decision a lower court ruling was allowed to stand that imposed a race-based quota requirement on a labor union. Similarly, in United States v. Paradise (480 U.S. 149, 1987), the Court, in another five-to-four decision, affirmed the constitutionality of a quota system involving the hiring of state police. Gender-based preferences would also be upheld under Title VII in the important case of Johnson v. Transportation Agency, Santa Clara County (480 U.S. 616, 1987). In these cases, Belz argues, the Court acknowledged that affirmative action is a prospective policy based on the idea of group rights that aims at achieving racial and gender balance, under the idea of proportional representation that is inherent in the disparate impact theory (Belz, 1991).
The Court’s support for affirmative action, however, was tenuous as shown by the many five-to-four decisions, and restrictions were placed on affirmative-action programs in a number of areas. In Firefighters Local Union No. 1794 v. Stotts (467 U.S. 561, 1984), for instance, the Court considered the validity of a district court order modifying the arrangements of a consent decree for hiring and promoting Black firefighters in Memphis. The decree attempted to protect newly hired Black workers from the “last hired, first fired” layoff policy by ruling that the use of the seniority system was illegitimate. The Supreme Court found the district court in violation of Title VII and modified the consent decree. A similar action occurred in Wygant v. Jackson Board of Education (476 U.S. 267, 1986), where the Court ruled that an affirmative-action plan that protected Black teachers while White teachers with more seniority were being laid off violated Title VII.
One form of affirmative-action preference that became popular among state and municipal governments in the mid-1970s was the minority contracting set-aside. Set-aside programs usually involve the reservation of a fixed proportion of public contracting dollars that by law must be spent on the purchase of goods and services provided by minority-owned busi-
nesses. Like preferences in hiring, set-asides have been enormously controversial and cries of “reverse discrimination” abound.
The Supreme Court first took up set-asides in the case of Fullilove v. Klutznick (448 U.S. 448, 1980), which challenged a provision of a federal law passed during the Carter administration. That provision required that 10 percent of federal funds allocated to state and local governments for public works projects be used to purchase goods and services from companies owned by members of six specified minority groups. The Court held in this case that the federal set-aside law did not violate the equal protection provisions of the federal Constitution on the grounds that the set-aside provision was a legitimate remedy for present competitive disadvantages resulting from past illegal discrimination.
Nine years later, however, in Richmond City v. J.A.Croson Co. (488 U.S. 469, 1989), the Court, again reflecting the influence of the Reagan-era appointees, held that racial classifications within state and local set-aside programs were inherently suspect and were to be subject to the most searching standard of constitutional review (“strict scrutiny”) under the equal protection provisions of the Fourteenth Amendment. By a six-to-three vote, the Court invalidated the Richmond City Council’s set-aside plan that had required contractors to subcontract at least 30 percent of the dollar value of contracts to minority-owned businesses.
The following year, however, in Metro Broadcasting Inc. v. Federal Communications Commission (110 S.Ct. 2997, 1990), the Court ruled constitutional a policy developed by the Federal Communications Commission that granted preferences in the purchase of broadcast licenses to minority-controlled firms. In a five-to-four decision, the Court held that, in this case, the race-based classification was “benign” in intent, and as such should not be subject to the “strict scrutiny” standard of constitutional review. Rather, an intermediate level of scrutiny was held to be appropriate. In this case, the state’s objective of enhancing broadcast diversity was important enough, the Court majority declared, to validate the state’s use of an otherwise suspect racial classification.
Five years later, however, in Adarand Contractors Inc. v. Pena (515 U.S. 200, 1995), the Court ruled that however benign in intent, affirmative-action programs that draw racial classifications, even those at the federal level, are subject to strict scrutiny. To critics, at least, it seemed that the Court had abruptly turned its back on settled law, particularly in regard to its earlier decisions affirming racial preferences at the federal level in Metro Broadcasting and Fullilove. Justice Clarence Thomas, a Bush appointee, the only Black on the Court, and a long-time foe of racial preferences of all kinds, issued his concurring opinion defending the color-blind principle of racial justice:
That these programs may have been motivated, in part, by good intentions, cannot provide refuge from the principle that under our Constitution, the government may not make distinctions on the basis of race. As far as the Constitution is concerned, it is irrelevant whether a government’s racial classifications are drawn by those who wish to oppress a race or by those who have a sincere desire to help those thought to be disadvantaged (515 U.S. at 240, Thomas concurring opinion).
Equally important—and equally controversial—were court cases dealing with affirmative action in higher education. Beginning in the late 1960s, many universities and professional schools began admitting minority students, particularly Blacks and Hispanics, with substantially lower grades and lower scores on standardized tests than White students. Many White students charged “reverse discrimination,” and some brought suit in federal court, claiming that affirmative action in higher education was a violation of Title VI of the 1964 Civil Rights Act, as well as of the equal protection provisions of the U.S. Constitution.
The first of these cases to come before the Court was DeFunis v. Odegaard (416 U.S. 312, 1974), involving a White applicant, Marco DeFunis, who had applied for admission to the law school of the University of Washington. DeFunis’s application for admission had originally been rejected, despite the fact that lower-scoring minority students had been admitted under a special admissions procedure. The procedure sorted applicants into two race-based applicant pools, with a much less rigorous standard applied to the minority group. Although the Supreme Court initially accepted the case and heard oral arguments, it finally decided to declare the case moot because DeFunis had been accepted to the University of Washington law school, by order of a lower court, and was nearing completion of his studies. Thus the decision in the case would have had no effect on him. DeFunis was an important case, however, because of a dissenting opinion against the decision to moot by Justice William O. Douglas. Although generally considered the Court’s leading liberal theorist, Douglas in his dissent offered a ringing denunciation of race-based admissions policies. Despite their benign intent, Douglas claimed, such policies were stigmatizing to minority groups by suggesting that minorities “cannot make it on their individual merit.” That, he said, “is a stamp of inferiority that a state is not permitted to place on any lawyer” (Dworkin, 1977:63–83).
Two years after the decision to moot in DeFunis, the Court again took up the issue of special preferences in higher education in the case of Regents of the University of California v. Bakke (438 U.S. 265, 1978). Alan P.
Bakke, a 32-year-old White mechanical engineer, had twice sought admission to the University of California at Davis Medical School. Both times he was rejected, despite the fact that his college grade point average and his score on the Medical College Admission Test were substantially higher than those of most of the minority students who were accepted. Because Davis filled 16 of its 100 entering slots through a special admissions program open only to minorities, Bakke claimed that he was a victim of reverse discrimination. Minority applicants, his lawyers pointed out, were permitted to compete for all of the 100 seats in the entering class, while Whites could only compete for the 84 “regular” seats specifically open to them. In a long-awaited decision, written by Justice Lewis Powell, the Court struck down the preferential admissions program at Davis as incompatible with the Fourteenth Amendment’s Equal Protection Clause, and ordered Bakke admitted. However, although ruling against Davis’s explicit racial-quota system, Powell declared, in his opinion, that admissions officers in institutions such as the Davis medical school could take race into account as one of many “plus” factors designed to enhance the diversity of a school’s student body.
The Bakke case was, in many ways, the most publicized of all the affirmative-action cases to come before the U.S. Supreme Court, and probably generated more heated commentary than any other case. It dealt with an issue that was even more contentious than affirmative action in employment, and one that drove the deepest wedge between traditionally liberal Jewish groups (most of which supported Bakke’s position) and Black groups (most of which supported the position of the University of California at Davis Medical School). Justice Powell’s decision was seen in many quarters as a Solomonic compromise, with something good in it for each side. To the disappointment of affirmative-action critics, however, the decision seemed to have little effect in changing the affirmative-action policies at most universities and professional schools, which continued to aggressively recruit and admit minority students who in some cases had lower grades and test scores than White students.
Affirmative action in higher education continued to suffer major setbacks during the 1990s. Borrowing from the activist strategy of civil rights groups, two public interest law firms, the Center for Individual Rights (CIR) and the Institute of Justice, recruited aggrieved White students to file reverse discrimination suits against institutions of higher learning. In 1995, the CIR litigated Hopwood v. Texas (78 F.3d 932 Fifth Circuit, 1995), the first high-profile, higher education, affirmative-action case since Bakke. Cheryl Hopwood, a White woman from a disadvantaged background, was denied admission to the University of Texas Law School, despite having higher test scores and grades than some of the Blacks and Mexicans admitted that year. Hopwood and three White males who were also
denied admission brought a racial discrimination suit against the law school. Although the plaintiffs lost in a lower court decision, they won their case when it reached the U.S. Court of Appeals for the Fifth Circuit, which ruled in Hopwood’s favor. Hopwood became legally binding in the states located in the Fifth Circuit region (Louisiana, Mississippi, and Texas), after the Supreme Court refused to review the case. The Hopwood decision has encouraged similar suits in other states.
Until 1997, the higher education cases involved disappointed applicants to professional schools. But a new twist was introduced in 1997 when two White undergraduates, Jennifer Gratz and Patrick Hamacher, filed racial discrimination suits against the University of Michigan and its administrators alleging discrimination in the university’s undergraduate program. At roughly the same time, White applicant Barbara Grutter sued the University of Michigan Law School, also charging reverse discrimination. In 1998, the cases of Gratz and Hamacher v. University of Michigan and Grutter v. Michigan were combined into a class-action suit that is still pending before Michigan’s U.S. District Court. Adverse decisions in the Michigan cases, in conjunction with the direct and indirect impact of California’s antiaffirmative-action initiative Proposition 209, could effectively eliminate race-conscious affirmative action in higher education. Already the effects of the ballot initiatives and court cases have been felt in California and Texas, where fewer minority students have applied to state universities, and fewer have been admitted to the flagship universities.
THE POLITICAL MILIEU OF AFFIRMATIVE ACTION
In the 1980s and 1990s, affirmative action became more politicized than ever as politicians began to view it as an issue that could win them votes. Antiaffirmative action sentiments were perhaps most visible in the 1990 North Carolina senatorial race between the Black Democrat Harvey Gantt and the Republican incumbent Jesse Helms (Applebome, 1990). During the last week of the campaign, while trailing Gantt in the polls, Helms aired a series of antiaffirmative-action television commercials, which were likely responsible for his victory. The commercial that got the most attention nationally featured the hands of a White male crumbling a job application, with the voice-over stating that although the White applicant needed the job, it was given to a lesser qualified minority because of a racial preference. The advertisement accused Harvey Gantt and Massachusetts Senator Ted Kennedy of favoring racial quotas. A second Helms advertisement showed a broadcasting station that Gantt had purchased through a federal set-aside program, then sold some weeks later to a White businessman for a huge profit. The voice-over explained how the
minority community was outraged by the sale, but that the deal made Gantt a millionaire. Postelection voter surveys attributed Gantt’s defeat to increased turnout among Whites spurred to vote by Helm’s anti-affirmative-action advertisements.
The frustration of Whites over affirmative-action programs was also given credit for former Klansman David Duke’s ability to win 55 percent of the White vote in Louisiana’s 1991 gubernatorial race. Three years earlier, Duke won 44 percent of the vote in his senatorial campaign. According to political analyst Gary Esolen, Duke’s message was always the same:
Why do reporters and politicians pick on me and say hateful things about me when anyone can see I’m a nice guy and reasonable fellow? Besides, even if I did say and do extreme things once or twice that’s all over with now.
The real reason they pick on me is that they are afraid of my message on the issues, which is that affirmative action has gone too far and become racism in the reverse, the Black underclass is dragging us down, we can’t afford welfare, and it’s time White people had some rights again (Esolen, 1992:137–138).
Polling data suggest that Duke’s antiaffirmative-action message resonated very well among Whites in Louisiana.
A development of even greater national significance occurred in California in 1995, when two academics, Glynn Custred and Thomas Wood, started the California Civil Rights Initiative (CCRI), a ballot proposal out-lawing government-sponsored racial preferences, which became Proposition 209. When it was passed in 1996, the law was immediately challenged by the American Civil Liberties Union, but was upheld by decisions handed down in 1997 by the U.S. Ninth Circuit Court of Appeals and the U.S. Supreme Court.
Control of the political agenda and of the language used in the initiative was an important factor in the passage of Proposition 209—i.e., “the state shall not discriminate against or grant preferential treatment to any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting” (California Constitution, Article I, Section 31). In 1998, a similarly worded initiative passed in the state of Washington. Although one could speculate that White Californians’ fears about the state’s increasingly racially diverse population—a 43 percent minority population in 1990—was a driving force behind the passage of Proposition 209, this would not explain the outcome in Washington where the minority population was only 13 percent; rather, the language of these two ballot initiatives seems to have been a key factor in their passage.
In Houston, Texas, supporters of the city’s various affirmative-action programs were able to word the initiative and frame the issue in a manner more favorable to their cause, so that voters voted on whether they wanted to dismantle all affirmative-action programs for women and other minorities, rather than whether they wanted to end all racial or gender discrimination. With the question so-framed, a clear majority of voters voted not to eliminate affirmative action. In the two states where affirmative-action opponents controlled the agenda, voters voted against state supported “discrimination” and “preferential treatment” rather than for affirmative action. The key difference between the measures that passed and the one that failed seems to have been wording.
Politicians, scholars, and the media largely control how affirmative action is presented to the public and debated. Hochschild (1999:619–620) argues that “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.” Political actors, she argues, “find affirmative action an immensely valuable issue over which to debate, and therefore have little desire to figure out just how it operates” (Hochschild, 1999:629). These political actors often speak and act as if they know what the American people think about affirmative action. Research shows, however, that the actual positions that Americans hold about affirmative action are often more complex and less polarized than one would discern from the rhetoric of public debate.
MEASURING PUBLIC SUPPORT FOR AFFIRMATIVE ACTION
It is generally believed that White Americans oppose, whereas Blacks enthusiastically embrace, affirmative-action programs. Racial polarization is seen as characterizing the dominant views of both groups, and there are survey data to support this view. After years of focusing on polarization between Whites and Blacks, however, survey researchers have begun to realize that public opinion on the issue is highly sensitive to question wording and question context. Attitudes about affirmative action often depend on how survey questions are framed, the answer choices that respondents are given, and the context of the questions. It is how we define and measure support for affirmative-action policy that determines, to a large extent, what we eventually find (Adkins, 1996; Gamson and Modigliani, 1987; Kinder and Sanders, 1990, 1996; Schuman et al., 1997; Sigelman and Welch, 1991:144; Steeh and Krysan, 1996; Stoker, 1997; Swain et al., 2000). Respondents’ answers to direct questions about their support of, or opposition to, affirmative action tell little about the types of public policies that a given individual will endorse. In fact,
Norman (1995) observed that respondents who say they oppose affirmative action may actually support more types of affirmative-action programs than a person who identifies himself as an affirmative-action supporter may.
Greater awareness of the sensitivity of affirmative-action questions to framing and contextual problems has led some researchers to conclude that validity could be greatly improved by abandoning use of the term “affirmative action” and, instead, describing the content of specific policies (Adkins, 1996; Fine 1992; Steeh and Krysan, 1996). Because of the problem of context and question framing, much of the survey data on public attitudes in this area is difficult to interpret.
Stoker (1997), in particular, has argued that survey questions that generalize across, or ignore, the context in which affirmative-action programs are implemented greatly misrepresent public opinion on the issue. The context of a question can include the reason why the program was adopted, the location of the question in the survey, and what preceded it. In describing a series of affirmative-action scenarios for people to respond to, Stoker provided either no context to justify the implementation of racial quotas or two rationales—underrepresentation of minorities and proven discrimination by a given company:
Question 1: No Context
Do you think that large companies should be required to give a certain number of jobs to Blacks or should the government stay out of this?
Question 2: Underrepresentation Context
There are some large companies where Blacks are underrepresented…. Do you think that these large companies should be required to give a certain number of jobs to Blacks, or should the government stay out of this?
Question 3: Proven Discrimination Context
There are some large companies with employment policies that dis criminate against Blacks…. Do you think these large companies should be required to give a certain number of jobs to Blacks, or should the government stay out of this?
Stoker found considerable support for compensatory measures based on the proven discrimination scenario; she also notes that proven discrimination is the only instance for which the Supreme Court in recent years has endorsed the use of quotas.
White support for racial quotas was weakest on Question 2, where quotas are being used to remedy underrepresentation. Organizational leaders, college presidents, corporate executives, and state and local gov-
ernments often point to the underrepresentation of minorities as their justification for implementing or expanding affirmative-action measures. But while diversifying the workplace is a value that many corporations, public bodies, and academic institutions share, many White Americans do not believe that this is a sufficient justification for racial quotas. Stoker’s data suggest that societal elites must develop a more compelling reason to justify preferential treatment than simply the underrepresentation of minorities if they hope to win greater public support.
Sniderman and Piazza (1993) conclude that dislike of affirmative-action policies is causing some Whites to dislike Blacks. They found that when asked the question below before being asked questions about Black stereotypes, White respondents were more likely to stereotype Blacks negatively than a similar group who encountered the stereotype questions first.
In a nearby state, an effort is being made to increase dramatically the number of Blacks working in state government. This means that a large number of jobs will be reserved for Blacks, even if their scores on merit exams are lower than those of Whites who are turned down for the jobs. Do you favor or oppose this policy?
The context and the wording of Sniderman and Piazza’s question also show, however, that respondents were asked about two forms of affirmative action that are now illegal: race norming, in which the test scores of Blacks are compared only to the scores of other Blacks, and quotas as a remedy for underrepresentation. White respondents were reacting negatively to clearly unfair public policies and practices that the U.S. Supreme Court declared unconstitutional years ago and that Congress banned with its passage of the 1991 Civil Rights Act. Many White Americans who are incensed by such policies are unaware that they are illegal and, in fact, probably think that quotas are an essential part of affirmative-action programs. Researchers should be aware that widely reported survey responses to questions about aspects of affirmative-action policy that may actually have changed can reinforce false notions about affirmative action and add to the public’s confusion and misconceptions about the policy.
Survey questions that elicit the greatest degree of racial polarization are sensitive to framing and contextual issues that go beyond question wording. For instance, responses to the questions below are no doubt influenced by the inclusion of the words “past discrimination” and the differential meanings that these words hold for different Americans depending on their backgrounds, ages, and current events at the time. Racial polarization was particularly evident in responses to questions asked in the late 1970s, in the wake of two 1978 Supreme Court decisions: the
Bakke decision and the decision not to review a lower court ruling supporting American Telephone and Telegraph Company’s affirmative-action program against charges of reverse discrimination (Johnson, 1983).
Some people say that because of past discrimination against Blacks, preference in hiring and promotion should be given to Blacks. Others say that preferential hiring and promotion of Blacks is wrong because it gives Blacks advantages they haven’t earned. What about your opinion—are you for or against preferential hiring and promotion of Blacks?
Some people say that because of past discrimination, it is sometimes necessary for colleges and universities to reserve openings for Blacks and other students. Others oppose quotas because they give to Blacks advantages that they haven’t earned. What about your opinion—are you for or against quotas to admit Black students?
After years of discrimination, is it only fair to set up special programs to make sure that women and minorities are given every chance to have equal opportunities in employment and education?
To Question 1, 85 percent of Whites and 32 percent of Blacks opposed preferential treatment. To Question 2, 70 percent of Whites and 20 percent of Blacks opposed quotas. All three questions make reference to past discrimination; but the respondent is left to infer independently whether the question is about slavery or blatant pre-Civil Rights discrimination. Blacks’ support is greatest for Question 3; 91 percent agree that after years of discrimination, it’s only fair to set up special programs for women and minorities (Johnson, 1983:302). Racial polarization is much less evident, however, on a Gallup poll question (Gallup Index, 1977) asked a year earlier.
Some people say that to make up for past discrimination, women and members of minority groups should be given preferential treatment in getting jobs and places in college. Others say that ability, as determined by test scores, should be the main consideration. Which point comes closest to how you feel on this matter?
Eighty-three percent of White respondents and 64 percent of Black respondents said that ability, as determined by test scores, should be the main consideration in employment and college admissions decisions. In a different poll, an interesting pattern appeared in the responses of Black leaders and the general public. In response to a question asking whether preferential treatment or ability should be used in obtaining jobs and
college placement, 77 percent of Black leaders endorsed preferential treatment, while only 23 percent of the Black public did (Lichter, 1985).
Examination of a range of survey questions shows that Whites and Blacks are not as polarized as is commonly believed, and in some policy areas they seem to be moving more toward consensus. Blacks are by no means enthusiastic supporters of racial preferences, and in recent years have given affirmative action a less than ringing endorsement. They have strongly endorsed self-help initiatives. As many as 70 percent of Blacks, for instance, agreed with the majority of Whites that Blacks should pull themselves up as other groups have done (Sniderman and Carmines, 1997).
The Irish, the Italians, and the Jews and many other minorities overcame prejudice and worked their way up. Blacks should do the same without any special favors.
Similarly, 48 percent of Blacks agreed with 53 percent of Whites on a 1997 Joint Center for Political and Economic Studies (JCPS) question that stated, “Blacks who can’t get ahead in the U.S. are mostly responsible for their own condition.” Fifty-nine percent of Black Republicans and 57 percent of Blacks making more than $60,000 a year agreed with the majority of Whites on this question (Bositis 1997a, 1997b).
The lack of Blacks’ enthusiasm for affirmative action reveals itself on a question asked in 1984 by the National Black Election Study (NBES). Only 43 percent of Blacks agreed with the statement:
Because of past discrimination, minorities should be given special consideration when decisions are made about hiring applicants.
On a related NBES question stating that “job applicants should be judged solely on the basis of test scores and other individual qualities,” 54 percent of Blacks agreed. The NBES revealed no groundswell of Black support for affirmative-action policies.
When asked in 1988 whether “Blacks and other minorities should receive preference in hiring to makeup for past inequalities,” only 48 percent of Blacks agreed; 44 percent disagreed. White and Hispanic disagreement, however, was greater: 85 percent of Whites disagreed (only 10 percent supported preferences), along with 64 percent of Hispanics (31 percent supported preferences). On another question, a majority of the three groups opposed preferential treatment in college admissions. A 1991 Newsweek poll showed 72 percent of Whites opposed to preferential treatment in hiring (only 19 percent favored the idea), while Blacks were again almost evenly split, 48 percent against to 42 percent in favor. Similar
patterns of ambivalence among Blacks have been seen in other recent polls (Schneider, 1991).
In a 1997 survey commissioned by JCPS, the following question was asked of a random sample of the U.S. population:
We should make every possible effort to improve the position of Blacks even if it means giving them preferential treatment.
Of those responding to the survey, almost a majority of Blacks (49 percent to 45 percent) and an overwhelming percentage of Whites (83 percent to 15 percent) opposed preferential treatment of Blacks as a means of improving the group’s societal position. A further demographic breakdown showed that a majority of Black baby boomers, Black men, college-educated Blacks, and Blacks earning more than $15,000 per year opposed preferential treatment (Bositis, 1997a).
All respondents, including Blacks, seem to oppose giving a lesser-qualified minority a job over a better-qualified White applicant (Norman, 1995). We asked the following split ballot, random assignment question on the 1996 Princeton Survey:
Suppose that a company that has few (female/minority/Black) employees were choosing between two people who applied for a job. If both people were equally qualified for the job and one was (a woman/a minority person/a Black person), and the other (a man/was not a minority person/White) do you think the company should hire the (woman/minority person/Black person), hire the (man/other person/White person), or should they find some other way to choose?
When “minorities” were mentioned, 82 percent of Whites and 71 percent of Blacks said the company should find some other way to choose. Only 22 percent of Blacks and 12 percent of Whites said that an underrepresented minority should be hired. On a follow-up question, 66 percent of the respondents who had favored hiring a qualified minority applicant over a nonminority applicant said that they would not feel the same way if the minority person “was slightly less qualified” than the other person. Consistent with these findings, Norman (1995:49) reports that there is consensus opposition among all Americans to giving a minority applicant a job instead of a better-qualified White, even if the workplace has few minorities; and only 13 percent of women and 22 percent of Blacks approved of such a policy.
The majority of survey respondents agree on certain types of policies designed to ensure equal opportunity in education and job training, provided those policies are open to all who are qualified (Lipset and
Schneider, 1978; Lipset, 1992; December 6–7, 1997, New York Times-CBS Poll; Sniderman and Piazza, 1993; Sniderman and Carmines, 1997). Whites are generally not supportive of programs designed exclusively for Blacks and other racial minorities, though they are somewhat more amenable to programs that help women. Self-interest could obviously be a factor here, though White women are not among affirmative action’s strongest supporters.
Although White racism may be a factor in some White opposition to preferential treatment of racial minorities, there are principled reasons that might account for some of their behaviors (Sniderman and Piazza, 1993; Sniderman and Carmines, 1997). Perhaps equal opportunity programs and job-training programs garner more support because they are racially inclusive and available to all who are disadvantaged. Similarly, group preferences in college and university admissions—e.g., for athletes, musicians, and offspring of alumni—garner significant White support because they are not restrictive.
Knowledge of Americans’ unease with racial preference programs has led some scholars to call for race-neutral public policies, based on the assumption that a consensus exists for such programs, or can be built where it is absent (Kahlenberg, 1996; Wilson, 1980, 1987, 1996). Race-neutral programs are attractive because they are not as vulnerable as racial preference programs to judicial attacks and can survive the test of “strict scrutiny” under the Fourteenth Amendment’s Equal Protection Clause. However, it has been pointed out that race-neutral programs will not necessarily lead to the kind of diversified workforce and university campuses that many people desire (Bok and Bowen, 1998; Appiah and Gutmann, 1996; Glazer, 1995). To remove race differentials purported to be inherent in standardized tests, social scientists such as William Julius Wilson (1999) argue in favor of flexible merit-based criteria in college admissions that would de-emphasize standardized test scores and take into consideration some combination of race and class. In addition, Wilson is in favor of comprehensive policies that would embrace all races. He also favors a shift in national focus by referring to “affirmative opportunity” programs rather than the ambiguous term “affirmative action.” Race-neutral programs are part of a set of unresolved issues that must be debated at greater length by those concerned with racial justice and racial harmony in America.
CONSENSUS ON RACE, CLASS, AND COLLEGE ADMISSIONS
Racial consensus among ordinary Americans—but not among elites— can be found in the highly contentious area of college admissions. Survey data indicate that most Americans reject the use of race as a primary
criterion in college and university admissions. Their rejection of racial preferences, however, does not indicate that they believe that admission to colleges and universities should be automatically awarded to the highest-scoring students. Americans seem to want a system flexible enough to allow consideration of the circumstances that a person has had to overcome to achieve, whatever scores are presented to the admissions committee. Under some conditions, Americans will support the admission of a less-prepared student who has had to overcome severe hardship, even if that means that a member of their own racial group loses out.
Using a vignette embedded in a national survey, Swain et al. (2000) randomly assigned the races and genders of two hypothetical students so that 16 possible combinations of race and gender were presented to equal numbers of respondents randomly assigned to answer the question below:1
Please suppose that a state university is deciding between two high school seniors who have applied for admission. I will read you a brief description of these two students. Then I will ask you to decide, if the college has space for only one more student, which of these do you think they should admit?” The interviewer then explains that “the first student attends a local public high school where [he or she] has maintained a B average. [He or she] is a [Black or White] student from a low-income family and has held a job throughout high school to help support [his or her] family. [He or she] scored slightly below average on [his or her] college admission tests. The second student attends a well-respected private school, where [he or she] has been an A student. [He or she] comes from a prominent [White or Black] family and has spent two summers studying abroad. [He or she] scored well on [his or her] college admission tests.” The interviewer next asks, “Based on what I have told you about these two students, which one do you think the college should admit?” After respondents have given their answer, they are asked “Regardless of who you think should be admitted, which student do you think the college would probably admit?
Swain, Rodgers, and Silverman found that a majority of respondents favored admission of the B student whenever the two students were from
the same racial group. The strongest support for the B student occurred when both students were White. If both students were Black a majority of respondents favored the B student, but by a smaller margin. In a mixed-race condition, support for the B student dropped; indeed, a majority of White respondents then favored a Black A student over a White B student, with this preference being strongest among highly educated Whites. Blacks favored the B student, even in the mixed-race situation in which the A student was Black, and the B student was White; and their support increased with education. Whites with a high school education or less preferred the B student by a margin of 21 to 16 (57 percent). On the other hand, 81 percent of White college graduates favored the Black A student over the White B student.
Despite their own preferences, the majority of respondents (around 90 percent in the same-race conditions and 80 percent in the mixed-race conditions) believed that the institution would select the A student. In the mixed-race condition, however, about 20 percent of respondents of each race believed that a B student of the opposite race would be admitted over an A student of their racial group. The belief that the A student would be admitted over the B student was even stronger among the B student’s supporters—i.e., the majority of respondents did not expect the institution to operate in a manner that they considered just. Whites and Blacks agreed on this matter.
The college admissions vignette suggests that people want university officials to exercise some flexibility in defining merit criteria, and they want a system that creates opportunities for the disadvantaged. The majority do not believe that diversity enhancement is a compelling reason for racial preferences in college admissions, or that being Black is equivalent to being disadvantaged. A December 1997 New York Times-CBS poll of a random sample of the U.S. population found consensus on the items below:
Suppose a White student and a Black student are equally qualified, but a college can admit only one of them. Do you think the college should admit the Black student in order to achieve more racial balance in the college, or do you think racial balance should not be a factor?”
Of those expressing a view, 77 percent of Whites and 72 percent of Blacks said that the race of the student should not be a factor. Clearly, these people felt that the institution should find some other way to choose. For them, perhaps, flipping a coin would be better than choosing students based on race.
A second question, about students with similar backgrounds and unequal qualifications, met with a similar response:
Suppose there is a White student who has an A average and a Black student who has a B average, and the students are equal in every other way, but a college can admit only one of them. Do you think the college should admit the Black student in order to achieve more racial balance, or do you think that racial balance should not be a factor?
Among those expressing an opinion, a decisive majority of both races (more than 75 percent of Blacks and more than 90 percent of Whites) said that the A student should be admitted over the B student.
For a related question, opinions changed when evidence was brought to bear that showed achievement despite unequal starting points:
In general, in hiring, promoting, and college admissions, do you think that it is a good idea or a poor idea to select a person from a poor family over a person from a middle-class or rich family if the person from the poor family and the person from the rich family are equally qualified?
Fifty-three percent of Whites and 65 percent of Blacks said that it was a good idea. Clearly, Whites and Blacks can agree on helping the disadvantaged in many important situations; this has obvious implications for public policies.
Fifty-nine percent of White respondents in the NYT-CBS Poll favored special educational programs to help minorities compete more effectively for college admissions, 64 percent favored job-training programs for minorities in industries where they were underrepresented, and 65 percent endorsed laws to protect minorities against racial discrimination. However, Whites disagreed with Blacks about the continued need for affirmative-action programs. When the term, “affirmative action” was used in the question, a majority of Whites wanted affirmative action ended (13 percent) or phased out over the next few years (45 percent). The majority of Blacks (80 percent, compared to 35 percent of Whites) felt that affirmative-action programs and racial preferences are needed to ensure diversity and to make up for past discrimination. Still, 67 percent of Blacks agreed with 81 percent of Whites that as a result of affirmative-action policies, less qualified individuals are sometimes hired and admitted to colleges.
The responses to the Swain, Rodgers, and Silverman college admissions vignette suggests that many Americans are committed to principles that allow for a substantially broader definition of merit than that used by the leading protagonists in the affirmative-action debate. That broader definition of merit includes consideration of the obstacles and hurdles that a given person has had to overcome to achieve the scores presented to the admissions committee. Although a majority of Americans seem to
agree that a hardworking underdog deserves a break, they clearly oppose the use of race even as a tiebreaker between two similarly advantaged persons. This agreement between the races might have been overlooked had the researchers simply asked respondents their views of affirmative action. Given such agreement, college and university admissions officials may have more latitude than many now assume. Americans seem to be asking them to take into consideration more factors than academic preparation alone. At the same time, neither Whites nor Blacks believe that racial preferences should play a significant role in college admissions decisions. This agreement might also have been overlooked if the authors had used code words in the vignette or asked about preferential treatment of the B student; most likely, a majority of Whites and a plurality of Blacks would have said no.
Consistent with other studies, the above data suggest that Americans care about helping those disadvantaged who try to help themselves. Americans seem to want to live in a country where hardworking, disciplined, and self-reliant people can rise above their initial station in life. Despite evidence that some middle-class Blacks have become cynical about the ideal of the American Dream (Hochschild, 1995), the more highly educated Black respondents in the Swain, Rodgers, and Silverman study seemed eager to help make the dream a reality for the disadvantaged, and supported the B student regardless of the B student’s race.
Americans can agree on the types of policies that they want for their children. In a national poll geared exclusively to children’s issues, Bositis (1997b) found Whites, Blacks, and Hispanics in agreement on several important issues. Clear majorities of each group supported universal healthcare for children, opposed “gangsta rap,” and even endorsed then-House Speaker Newt Gingrich’s controversial proposal for the government to build orphanages for children from dangerous or unhealthy homes. The poll, however, found a significant generational gap between old and young for both Whites and Blacks; younger people were more supportive of the proposals than older people. Likewise, younger people endorsed school vouchers and governmental spending on education to a greater degree than older people. Interestingly, Blacks and Hispanics were more supportive than Whites of government vouchers for parents to send their children to “the public, private, or parochial school of choice.”
Although survey data reveal considerable agreement between Whites and Blacks, and seem to support a convergence hypothesis, race-of-interviewer effects might arguably account for some of the changes reported in the attitudes of Blacks. Since 1964, for example, the percentage of Blacks interviewing Blacks in the NBES has dropped significantly. Anderson et al. (1988a, 1988b) have shown that some questions are susceptible to race-of-interviewer effects, especially questions on government assistance to
minorities. Race-of-interviewer effects do not, however, account for all of it; agreement and convergence has been seen even in polls where the race of interviewer was matched to the respondent (Bositis, 1997a, 1997b; 1996 Princeton National Survey). The preponderance of data indicates a nontrivial amount of agreement among Black and White respondents on a host of important issues that go beyond those mentioned in this paper. This agreement can and should be used to help shape public policies in ways that ensure maximum support among all Americans.
WHY RACIAL DISAGREEMENT OCCURS
When Whites and Blacks disagree about affirmative-action issues, some of their differences can be explained by the differing conceptual schemas that they use, as well as divergent framing effects that cause groups to see the world through different lenses. In psychology, a schema is a cognitive structure of prior learning and knowledge that draws on past experience to guide the processing of new information and the retrieval of stored memories (Conover and Feldman, 1984:96). Such schemas can be content-specific and based on categories such as race, sex, and social class. Cognitive schemas of this type color how a person views the world. For a White person, the schema used to interpret affirmative action might include direct experience with reverse discrimination, as well as prejudicial observations about minorities based on stereotypes. Awareness of racial discrimination against Blacks may not be on the radar screen of the average White American. A Black person’s schema, on the other hand, may be constructed around individual and group experiences with overt discrimination.
Blacks in focus groups commissioned in 1995 never defined affirmative action in terms of preferential treatment; and when they discussed quotas, they used the term differently than Whites and members of most other groups. For them, quotas were restrictions used to limit the number of minorities in a given setting. A Black male in Edison, New Jersey, commented that employment quotas make sure that “only a certain number of Blacks can succeed, or benefit from affirmative action, when maybe there’s a whole cluster of people that are qualified. Maybe affirmative action says…you have to have three…[Blacks]. What about the rest of those people?” To them, the possibility remained strong that qualified Blacks will be overlooked once a certain threshold has been reached by a given institution (Swain et al., 2001).
Discussions of discrimination dominated the affirmative-action discourse in these focus groups, with several Blacks viewing affirmative action as the only, albeit flawed, way to combat the continuing problem of racial bias. Whites, not seeing the extent of the continuing discrimination
against Blacks, or not really understanding the links that Blacks make between affirmative action and protection against bias, see affirmative action as reverse discrimination (Gamson and Modigliani, 1987; Gamson, 1991). Different cognitive schemas and the internalization of different ways of framing the problem to be addressed may thus help to explain continued racial polarization in this area.
Perhaps an even greater consensus among Americans could be forged if Whites were made aware of the nature and extent of racial discrimination that Blacks experience. Highly publicized racist incidents and blatantly discriminatory practices show that overt racial discrimination has not been eliminated from the American scene and that it crops up in the unlikeliest places. Most Whites do not observe overt discrimination as a matter of course, unless they have been sensitized to something made highly visible, such as the frequency with which Black motorists are interrogated by the police, something they have witnessed or had their attention drawn to. Blacks and other racially identifiable minorities are exposed to and see more racial discrimination than White Americans do. It is the differential experiences with discrimination that cause some disagreement among Whites and Blacks about the need for certain types of public policies. Thus, their vastly different perceptions make it difficult for the two groups to communicate.
One thing that might help close the gap in perception is studies that can identify and expose hidden racism and discrimination in housing, employment, police actions, and college admissions. Such studies can heighten public awareness of the pervasiveness of discrimination and remove one of the sources of disagreement between Whites and Blacks— their assessments of the amount of discrimination minorities face in everyday life. The implementation of more audit studies, in which similarly matched Whites and minorities test the fairness of the system by applying for housing, jobs, apartments, etc., is one possible solution. If more Whites are sensitized to the discrimination that Blacks and other minorities confront, their sense of fairness may lead them to support increased enforcement of antidiscrimination laws. Given existing congressional attitudes toward funding government audits, it may be necessary for private foundations to step in to fund these studies. More evidence of documented discrimination collected from government- and privately financed audit studies could be used to pressure Congress to adequately fund the agencies charged with monitoring discrimination. Moreover, greater punitive damages against companies and institutions that engage in discriminatory actions would be helpful as a deterrent. Of course, this will not totally alleviate the problem, but increased public awareness and pressure could lead to increased funding for agencies like EEOC. Increased aware-
ness and pressure could also lead to changes in individual behavior, which might be cumulative over time.
SUMMARY AND CONCLUSION
Affirmative action is a complicated national issue, and there are no easy answers. The issue has historically been, and continues to be, plagued by ambiguity surrounding the concept and by the ways in which the various policies have been implemented. Confusion also stems from factors including the self-serving interpretations of politicians and inaccurate media portrayals. Government officials who rely on public-opinion polls to guide them in their decisions about affirmative-action programs need to exercise caution in how they interpret survey results. It is grossly inaccurate to conclude that racial polarization characterizes the opinions of most Americans on affirmative-action and related issues. How Americans feel about affirmative-action programs depends on a number of factors, including who the beneficiaries of the program are and whether the program was adopted to address proven discrimination. Neither Whites nor Blacks are enthusiastic supporters of racial preference programs; however, both groups favor creating opportunities and helping disadvantaged people. Agreement and consensus among Americans on affirmative action-related issues are detected most readily when respondents to national surveys are presented with clear-cut questions and concrete examples that allow them to decide among choices that they understand. Government officials and public-policy makers should consider the identification and expansion of areas of consensus and agreement among racial groups as part of their official mandates. Racial harmony is a goal worth striving for. Our society is becoming increasingly racially diverse. To achieve any measurable success in race relations, compromise, consensus building, and open discussion are essential as we enter the twenty-first century.
The Mellon Foundation, the Smith-Richardson Foundation, the Sloan Foundation, and the National Science Foundation SBR-9357951 supported the 1996 Princeton Survey and 1995 focus groups. I am also indebted to the Center for Advanced Study in the Behavioral Sciences and many of its 1997–1998 visiting scholars. In addition, I wish to thank Jennifer Hochschild and Neil Smelser for commenting on an earlier draft of this paper.
Adkins, R. 1996 Affirmative action and public opinion polls, in Race Versus Class: The New Affirmative Action Debate, C.Swain, ed. Baltimore: University Press of America.
Anderson, B., B.Silver, and P.Abramson 1988a The effects of race of interviewer on measures of electoral participation by Blacks in SRC national election studies. Public Opinion Quarterly 52:53–83.
1988b The effects of race of interviewer on race-related attitudes of Black respondents in SRC/CPS national election studies. Public Opinion Quarterly 52:289–324.
Appiah, A., and A.Gutmann 1996 Color Consciousness: The Political Morality of Race. Princeton: Princeton University Press.
Applebome, P. 1990 Helms kindled anger in the campaign, and may have set the tone for others. New York Times (November 8).
Belz, H. 1991 Equality Transformed: A Quarter Century of Affirmative Action. New Brunswick, N.J.: Transition Books.
Bergman, B. 1996 In Defense of Affirmative Action. New York: Basic Books.
Bok, D., and W.Bowen 1998 The Shape of the River. Princeton: Princeton University Press.
Bositis, D. 1997a Joint Center for Political and Economic Studies 1997 National Opinion Poll on Race Relations (June). Washington, D.C.: Joint Center for Political and Economic Studies.
1997b Joint Center for Political and Economic Studies 1997 National Poll on Children’s Issues (June). Washington, D.C.: Joint Center for Political and Economic Studies.
Conover, P., and S.Feldman 1984 How people organize the political world: A schematic model. American Journal of Political Science 25:617–645.
Delgado, R. 1996 The Coming Race War? And Other Apocalyptic Tales of America after Affirmative Action and Welfare. New York: New York University Press.
Drake, A., and R.Holsworth 1996 Affirmative Action and the Stalled Quest for Black Progress. Urbana, Ill.: University of Illinois Press.
Dworkin, R. 1977 DeFunis v. Sweatt. In Equality and Preferential Treatment, M.Cohen, T.Nagel, and T.Scanlon, eds. Princeton: Princeton University Press.
Edley, C., Jr. 1996 Not All Black and White: Affirmative Action and American Values. New York: Hill and Wang.
Esolen, G. 1992 David Duke’s use of television, in The Emergence of David Duke and the Politics of Race, D.Rose, ed. Chapel Hill: University of North Carolina Press.
Fine, T. 1992 The impact of issue framing on public opinion toward affirmative-action programs. Social Science Journal 29(3):323–334.
Gallup Index 1977 Report no. 143 (June). Princeton, N.J.: Gallup Organization.
Gamson, W. 1991 Talking Politics. New York: Cambridge University Press.
Gamson W., and A.Modigliani 1987 The changing culture of affirmative action. Research in Political Sociology 3:137– 177.
Glazer, N. 1975 Affirmative Discrimination. New York: Basic Books.
1995 Race, not class. The Wall Street Journal (April 5):A12.
Graham, H. 1990 The Civil Rights Era: Origins and Development of National Policy. New York: Oxford University Press .
Hacker, A. 1992 Two Nations: Separate, Hostile, Unequal. New York: Ballantine Books.
Hall, K., ed. 1992 The Oxford Companion to the Supreme Court of the United States. New York: Oxford University Press.
Hochschild, J. 1995 Facing up to the American Dream. Princeton: Princeton University Press.
1999 Affirmative action as culture warfare. In The Cultural Territories of Race: White and Black Boundaries, M.Lamont, ed. Chicago: University of Chicago Press.
Johnson, C. 1983 Black support for affirmative-action programs. Phylon XLIV(4).
Johnson, L. 1965 Howard University commencement address. Washington, D.C., June 4.
Kahlenberg, R. 1996 The Remedy: Class, Race, and Affirmative Action. New York: Basic Books.
Kinder, D., and L.Sanders 1990 Mimicking political debate with survey questions: The case of affirmative action for Blacks. Social Cognition 8(Spring):73–103.
1996 Divided by Color. Chicago: University of Chicago Press.
Lichter, L. 1985 Who speaks for Black America? Public Opinion Quarterly (August/September):43.
Lipset, S. 1992 Affirmative action and the American creed. The Wilson Quarterly 16(Winter):52– 62.
Lipset, S., and W.Schneider 1978 The Bakke case: How would it be decided in the court of public opinion. Public Opinion (March/April):38–44.
Norman, J. 1995 America’s verdict on affirmative action is decidedly mixed. Public Perspective 6(June/July):49–52.
Schneider, W. 1991 In job quota debate, advantage GOP. National Journal (June 8):1374.
Schuman, H., C.Steeh, L.Bobo, and M.Krysan 1997 Racial Attitudes in America. Cambridge, Mass.: Harvard University Press.
Sigelman, L., and S.Welch 1991 Black American Views of Inequality: The Dream Deferred. New York: Cambridge Press.
Skrentny, J. 1996 The Ironies of Affirmative Action: Politics, Culture, and Justice in America. Chicago: University of Chicago Press.
Smelser, N. 1999 Problematics of affirmative action: A view from California. In Dilemma and Promise: Perspectives on Racial Diversity in Higher Education, E.Lowe, Jr., ed. Princeton: Princeton University Press.
Sniderman, P., and E.Carmines 1997 Reaching Beyond Race. Cambridge, Mass.: Harvard University Press.
Sniderman, P., and T.Piazza 1993 The Scar of Race. Cambridge, Mass.: Harvard University Press.
Steeh, C., and M.Krysan 1996 Poll trends: Affirmative action and the public, 1971–1995. Public Opinion Quarterly 60:128–158.
Stoker, L. 1997 Understanding Whites’ resistance to affirmative action: The role of principled commitments and racial prejudice. In Perception and Prejudice: Race and Politics in the United States, J.Hurwitz and M.Peffley, eds. Chicago: University of Chicago Press.
Swain, C., ed. 1996 Race Versus Class: The New Affirmative Action Debate. Baltimore: University Press of America.
Swain, C., K.Greene, and C.Wotipka 2001 forth-coming Understanding racial polarization on affirmative action issues: The view from focus groups. In Color Lines: Affirmative Action, Immigration, and Civil Rights Op tions for America, J.Skrentny, ed. Chicago: University of Chicago Press.
Swain, C., R.Rodgers, and B.Silverman 2000 Life after Bakke where Whites and Blacks agree: Public support for fairness in educational opportunities. Harvard BlackLetter Law Journal 16(Spring).
Thernstrom, S., and A.Thernstrom 1997 America in Black and White. New York: Simon and Schuster.
Wilson, W. 1980 The Declining Significance of Race, 2nd ed. Chicago: University of Chicago Press.
1987 The Truly Disadvantaged. Chicago: University of Chicago Press.
1996 When Work Disappears: The World of the New Urban Poor. New York: Knopf.
1999 The Bridge Over the Racial Divide: Rising Inequality and Coalition Politics. Berkeley, Calif.: University of California Press.