You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

In Defense of Preference

Affirmative action is bad. Banning it is worse

The battle over affirmative action today is a contest between a clear principle on the one hand and a clear reality on the other. The principle is that ability, qualifications, and merit, independent of race, national origin, or sex should prevail when one applies for a job or promotion, or for entry into selective institutions for higher education, or when one bids for contracts. The reality is that strict adherence to this principle would result in few African Americans getting jobs, admissions, and contracts. What makes the debate so confused is that the facts that make a compelling case for affirmative action are often obscured by the defenders of affirmative action themselves. They have resisted acknowledging how serious the gaps are between African Americans and others, how deep the preferences reach, how systematic they have become. Considerably more than a mild bent in the direction of diversity now exists, but it exists because painful facts make it necessary if blacks are to participate in more than token numbers in some key institutions of our society. The opponents of affirmative action can also be faulted: they have not fully confronted the consequences that must follow from the implementation of the principle that measured ability, qualification, merit, applied without regard to color, should be our only guide.

I argued for that principle in a 1975 book titled, provocatively, Affirmative Discrimination. It seemed obvious that that was what all of us, black and white, were aiming to achieve through the revolutionary civil rights legislation of the 1960s. That book dealt with affirmative action in employment, and with two other kinds of governmentally or judicially imposed “affirmative action,” the equalization of the racial proportions in public schools and the integration of residential neighborhoods. I continued to argue and write regularly against governmentally required affirmative action, that is, racial preference, for the next two decades or more: it was against the spirit of the Constitution, the clear language of the civil rights acts, and the interests of all of us in the United States in achieving an integrated and just society.

It is not the unpopularity of this position in the world in which I live, liberal academia, that has led me to change my mind but, rather, developments that were unforeseen and unexpected in the wake of the successful civil rights movement. What was unforeseen and unexpected was that the gap between the educational performance of blacks and whites would persist and, in some respects, deepen despite the civil rights revolution and hugely expanded social and educational programs, that inner-city schools would continue to decline, and that the black family would unravel to a remarkable degree, contributing to social conditions for large numbers of black children far worse than those in the 1960s. In the presence of those conditions, an insistence on color-blindness means the effective exclusion today of African Americans from positions of influence, wealth, and power. It is not a prospect that any of us can contemplate with equanimity. We have to rethink affirmative action.

In a sense, it is a surprise that a fierce national debate over affirmative action has not only persisted but intensified during the Clinton years. After twelve years under two Republican presidents, Ronald Reagan and George Bush, who said they opposed affirmative action but did nothing to scale it back, the programs seemed secure. After all, affirmative action rests primarily on a presidential executive order dating back to the presidencies of Lyndon Johnson and Richard Nixon which requires “affirmative action” in employment practices from federal contractors—who include almost every large employer, university, and hospital. The legal basis for most of affirmative action could thus have been swept away, as so many noted at the time, with a “stroke of the pen” by the president. Yet two presidents who claimed to oppose affirmative action never wielded the pen.

Despite the popular majority that grumbles against affirmative action, there was (and is) no major elite constituency strongly opposed to it: neither business nor organized labor, religious leaders nor university presidents, local officials nor serious presidential candidates are to be found in opposition. Big business used to fear that affirmative action would undermine the principle of employment and promotion on the basis of qualifications. It has since become a supporter. Along with mayors and other local officials (and of course the civil rights movement), it played a key role in stopping the Reagan administration from moving against affirmative action. Most city administrations have also made their peace with affirmative action.

Two developments outside the arena of presidential politics galvanized both opponents and defenders of affirmative action. The Supreme Court changed glacially after successive Republican appointments--each of which, however, had been vetted by a Democratic Senate--and a number of circuit courts began to chip away at the edifice of affirmative action. But playing the largest role was the politically unsophisticated effort of two California professors to place on the California ballot a proposition that would insert in the California Constitution the simple and clear words, taken from the Civil Rights Act of 1964, which ban discrimination on the basis of race, national origin, or sex. The decision to launch a state constitutional proposition, Proposition 209, suddenly gave opponents the political instrument they needed to tap the majority sentiment that has always existed against preferences.

While supporters of affirmative action do not have public opinion on their side, they do have the still-powerful civil rights movement, the major elites in education, religion, philanthropy, government, and the mass media. And their position is bolstered by a key fact: how far behind African Americans are when judged by the tests and measures that have become the common coin of American meritocracy.

The reality of this enormous gap is clearest where the tests in use are the most objective, the most reliable, and the best validated, as in the case of the various tests used for admission to selective institutions of higher education, for entry into elite occupations such as law and medicine, or for civil service jobs. These tests have been developed over many years specifically for the purpose of eliminating biases in admissions and appointments. As defenders of affirmative action often point out, paper-and-pencil tests of information, reading comprehension, vocabulary, reasoning, and the like are not perfect indicators of individual ability. But they are the best measures we have for success in college and professional schools, which, after all, require just the skills the tests measure. And the tests can clearly differentiate the literate teacher from the illiterate one or the policeman who can make out a coherent arrest report from one who cannot.

To concentrate on the most hotly contested area of affirmative action--admission to selective institutions of higher education--and on the group in the center of the storm--African Americans: If the Scholastic Assessment Test were used for selection in a color-blind fashion, African Americans, who today make up about six percent of the student bodies in selective colleges and universities, would drop to less than two percent, according to a 1994 study by the editor of the Journal of Blacks in Higher Education.

Why is this so? According to studies summarized in Stephan and Abigail Thernstrom’s book, America in Black and White, the average combined SAT score for entering freshmen in the nation’s top 25 institutions is about 1300. White applicants generally need to score a minimum of 600 on the verbal portion of the test—a score obtained by eight percent of the test-takers in 1995—and at least 650 on the mathematics section--a score obtained by seven percent of the test-takers in 1995. In contrast, only 1.7 percent of black students scored over 600 on the verbal section in 1995, and only two percent scored over 650 on the math. This represents considerable progress over the last 15 years, but black students still lag distressingly far behind their white counterparts.

There is no way of getting around this reality. Perhaps the tests are irrelevant to success in college? That cannot be sustained. They have been improved and revised over decades and predict achievement in college better than any alternative. Some of the revisions have been carried out in a near-desperate effort to exclude items which would discriminate against blacks. Some institutions have decided they will not use the tests, not because they are invalid per se, but because they pose a barrier to the increased admission of black students. Nor would emphasizing other admissions criteria, such as high school grades, make a radical difference. In any case, there is considerable value to a uniform national standard, given the enormous differences among high schools.

Do qualifications at the time of admission matter? Isn’t the important thing what the institutions manage to do with those they admit? If they graduate, are they not qualified? Yes, but many do not graduate. Two or three times as many African American students as white students drop out before graduation. And the tests for admission to graduate schools show the same radical disparities between blacks and others. Are there not also preferences for athletes, children of alumni, students gifted in some particular respect? Yes, but except for athletes, the disparities in academic aptitude that result from such preferences are not nearly as substantial as those which must be elided in order to reach target figures for black students. Can we not substitute for the tests other factors—such as the poverty and other hardships students have overcome to reach the point of applying to college? This might keep up the number of African Americans, but not by much, if the studies are to be believed. A good number of white and Asian applicants would also benefit from such “class-based” affirmative action.

(I have focused on the effect of affirmative action—and its possible abolition—on African Americans. But, of course, there are other beneficiaries. Through bureaucratic mindlessness, Asian Americans and Hispanics were also given affirmative action. But Asian Americans scarcely need it. Major groups—not all—of Hispanic Americans trail behind whites but mostly for reasons we understand: problems with the English language and the effect on immigrant children of the poor educational and economic status of their parents. We expect these to improve in time as they always have with immigrants to the United States. And, when it comes to women, there is simply no issue today when it comes to qualifying in equal numbers for selective institutions of higher and professional education.)

How, then, should we respond to this undeniable reality? The opponents of affirmative action say, “Let standards prevail whatever the result.” So what if black students are reduced to two percent of our selective and elite student bodies? Those who gain entry will know that they are properly qualified for entry, that they have been selected without discrimination, and their classmates will know it too. The result will actually be improved race relations and a continuance of the improvements we have seen in black performance in recent decades. Fifteen years from now, perhaps three or four percent of students in the top schools will be black. Until then, blacks can go to less competitive institutions of higher education, perhaps gaining greater advantage from their education in so doing. And, meanwhile, let us improve elementary and high school education—as we have been trying to do for the last 15 years or more.

Yet we cannot be quite so cavalier about the impact on public opinion—black and white—of a radical reduction in the number of black students at the Harvards, the Berkeleys, and the Amhersts. These institutions have become, for better or worse, the gateways to prominence, privilege, wealth, and power in American society. To admit blacks under affirmative action no doubt undermines the American meritocracy, but to exclude blacks from them by abolishing affirmative action would undermine the legitimacy of American democracy.

My argument is rooted in history. African Americans—and the struggle for their full and fair inclusion in U.S. society—have been a part of American history from the beginning. Our Constitution took special—but grossly unfair--account of their status, our greatest war was fought over their status, and our most important constitutional amendments were adopted because of the need to right past wrongs done to them. And, amid the civil rights revolution of the 1960s, affirmative action was instituted to compensate for the damage done to black achievement and life chances by almost 400 years of slavery, followed by state-sanctioned discrimination and massive prejudice.

Yet, today, a vast gulf of difference persists between the educational and occupational status of blacks and whites, a gulf that encompasses statistical measures of wealth, residential segregation, and social relationships with other Americans. Thirty years ago, with the passage of the great civil rights laws, one could have reasonably expected—as I did—that all would be set right by now. But today, even after taking account of substantial progress and change, it is borne upon us how continuous, rooted, and substantial the differences between African Americans and other Americans remain.

The judgment of the elites who support affirmative action—the college presidents and trustees, the religious leaders, the corporate executives—and the judgment even of many of those who oppose it but hesitate to act against it—the Republican leaders in Congress, for example—is that the banning of preference would be bad for the country. I agree. Not that everyone’s motives are entirely admirable; many conservative congressmen, for example, are simply afraid of being portrayed as racists even if their opposition to affirmative action is based on a sincere desire to support meritocratic principle. The college presidents who support affirmative action, under the fashionable mantra of diversity, also undoubtedly fear the student demonstrations that would occur if they were to speak out against preferences.

But there are also good-faith motives in this stand, and there is something behind the argument for diversity. What kind of institutions of higher education would we have if blacks suddenly dropped from six or seven percent of enrollment to one or two percent? The presence of blacks, in classes in social studies and the humanities, immediately introduces another tone, another range of questions (often to the discomfort of black students who do not want this representational burden placed upon them). The tone may be one of embarrassment and hesitation and self-censorship among whites (students and faculty). But must we not all learn how to face these questions together with our fellow citizens? We should not be able to escape from this embarrassment by the reduction of black students to minuscule numbers.

The weakness in the “diversity” defense is that college presidents are not much worried about the diversity that white working-class kids, or students of Italian or Slavic background, have to offer. Still, there is a reputable reason for that apparent discrepancy. It is that the varied ethnic and racial groups in the United States do not, to the same extent as African Americans, pose a test of the fairness of American institutions. These other groups have not been subjected to the same degree of persecution or exclusion. Their status is not, as the social status of African Americans is, the most enduring reproach to the egalitarian ideals of American society. And these other groups have made progress historically, and make progress today, at a rate that incorporates them into American society quickly compared to blacks.

This is the principal flaw in the critique of affirmative action. The critics are defending a vitally important principle, indeed, the one that should be the governing principle of institutions of higher education: academic competence as the sole test for distinguishing among applicants and students. This principle, which was fought for so energetically during the 1940s and 1950s through laws banning discrimination in admission on the basis of race, national origin, or religion, should not be put aside lightly. But, at present, it would mean the near exclusion from our best educational institutions of a group that makes up twelve percent of the population. In time, I am convinced, this preference will not be needed. Our laws and customs and our primary and secondary educational systems will fully incorporate black Americans into American society, as other disadvantaged groups have been incorporated. The positive trends of recent decades will continue. But we are still, though less than in the past, “two nations,” and one of the nations cannot be excluded so thoroughly from institutions that confer access to the positions of greatest prestige and power.

On what basis can we justify violating the principle that measured criteria of merit should govern admission to selective institutions of higher education today? It is of some significance to begin with that we in the United States have always been looser in this respect than more examination-bound systems of higher education in, say, Western Europe: we have always left room for a large degree of freedom for institutions of higher education, public as well as private, to admit students based on nonacademic criteria. But I believe the main reasons we have to continue racial preferences for blacks are, first, because this country has a special obligation to blacks that has not been fully discharged, and second, because strict application of the principle of qualification would send a message of despair to many blacks, a message that the nation is indifferent to their difficulties and problems.

Many, including leading black advocates of eliminating preference, say no: the message would be, “Work harder and you can do it.” Well, now that affirmative action is becoming a thing of the past in the public colleges and universities of California and Texas, we will have a chance to find out. Yet I wonder whether the message of affirmative action to black students today really ever has been, “Don’t work hard; it doesn’t matter for you because you’re black; you will make it into college anyway.” Colleges are indeed looking for black students, but they are also looking for some minimal degree of academic effort and accomplishment, and it is a rare ambitious African American student seeking college entry who relaxes because he believes his grades won’t matter at all.

One of the chief arguments against racial preference in college and professional school admissions is that more blacks will drop out, the quality of blacks who complete the courses of instruction will be inferior, and they will make poorer lawyers, doctors, or businessmen. Dropping out is common in American higher education and does not necessarily mean that one’s attendance was a total loss. Still, the average lower degree of academic performance has, and will continue to have, effects even for the successful: fewer graduating black doctors will go into research; more will go into practice and administration. More blacks in business corporations will be in personnel. Fewer graduating black lawyers will go into corporate law firms; more will work for government.

And more will become judges, because of another and less disputed form of affirmative action, politics. Few protest at the high number of black magistrates in cities with large black populations—we do not appoint judges by examination. Nor do we find it odd or objectionable that Democratic presidents will appoint more black lawyers as judges, or that even a Republican president will be sure to appoint one black Supreme Court justice. What is at work here is the principle of participation. It is a more legitimate principle in politics and government than it is for admission to selective institutions of higher education. But these are also gateways to power, and the principle of participation cannot be flatly ruled out for them.

Whatever the case one may make in general for affirmative action, many difficult issues remain: What kind, to what extent, how long, imposed by whom, by what decision-making process? It is important to bear in mind that affirmative action in higher education admissions is, for the most part, a policy that has been chosen (albeit sometimes under political pressure) by the institutions themselves. There are racial goals and targets for employment and promotion for all government contractors, including colleges and universities, set by government fiat, but targets on student admissions are not imposed by government, except for a few traditionally black or white institutions in the South.

Let us preserve this institutional autonomy. Just as I would resist governmentally imposed requirements that these institutions meet quotas of black admissions, so would I also oppose a judicial or legislative ban on the use of race in making decisions on admission. Ballot measures like Proposition 209 are more understandable given the abuses so common in systems of racial preference. But it is revealing that so many other states appear to have had second thoughts and that the California vote is therefore not likely to be repeated. (A recent report in The Chronicle of Higher Education was headlined “legislatures show little enthusiasm for measures to end racial preferences”; in this respect, the states are not unlike Congress.)

We should retain the freedom of institutions of higher and professional education to make these determinations for themselves. As we know, they would almost all make room for a larger percentage of black students than would otherwise qualify. This is what these institutions do today. They defend what they do with the argument that diversity is a good thing. I think what they really mean is that a large segment of the American population, significant not only demographically but historically and politically and morally, cannot be so thoroughly excluded. I agree with them.

I have discussed affirmative action only in the context of academic admissions policy. Other areas raise other questions, other problems. And, even in this one area of college and university admissions, affirmative action is not a simple and clear and uncomplicated solution. It can be implemented wisely or foolishly, and it is often done foolishly, as when college presidents make promises to protesting students that they cannot fulfill, or when institutions reach too far below their minimal standards with deleterious results for the academic success of the students they admit, for their grading practices, and for the legitimacy of the degrees they offer. No matter how affirmative action in admissions is dealt with, other issues remain or will emerge. More black students, for example, mean demands for more black faculty and administrators and for more black-oriented courses. Preference is no final answer (just as the elimination of preference is no final answer). It is rather what is necessary to respond to the reality that, for some years to come, yes, we are still two nations, and both nations must participate in the society to some reasonable degree.

Fortunately, those two nations, by and large, want to become more united. The United States is not Canada or Bosnia, Lebanon or Malaysia. But, for the foreseeable future, the strict use of certain generally reasonable tests as a benchmark criterion for admissions would mean the de facto exclusion of one of the two nations from a key institutional system of the society, higher education. Higher education’s governing principle is qualification--merit. Should it make room for another and quite different principle, equal participation? The latter should never become dominant. Racial proportional representation would be a disaster. But basically the answer is yes--the principle of equal participation can and should be given some role. This decision has costs. But the alternative is too grim to contemplate.

This article appeared in the April 6, 1998 issue of the magazine.