The Shape of the River
By William Bowen and Derek Bok

A Dream Deferred
By Shelby Steele.

"If color-blind admissions policies are put into effect," I was warned at a recent debate on the topic, "the number of black students at the most selective colleges and universities would plummet to around 2 percent. Should we as a society be willing to live with such an outcome?"

I hesitated, and in that moment of hesitation, my interlocutor saw his opportunity. "Well, should we?" he pressed.

The answer, it turns out, is yes. But it is an answer that supporters of the current system consider outrageous. They take for granted that the only possible response is "Of course not." So, for example, two pillars of the education establishment, former Princeton president William Bowen and former Harvard president Derek Bok, have just published a widely reviewed defense of affirmative action, The Shape of the River: Long-Term Consequences of Considering Race in College and University Admissions. They insist that some form of preferential recruitment is inevitable to avoid the unthinkable outcome of very few African Americans at top-ranked universities. "The adoption of a strict race-neutral standard would reduce black enrollment at . . . academically selective colleges and universities by between 50 and 70 percent," Bowen and Bok observe. "The most selective colleges would experience the largest drops in black enrollment."

These numbers are more or less correct. But what they actually illustrate is not the unacceptable future but the unconscionable present: the magnitude of racial preferences currently in effect. Affirmative action in practice does not mean -- as its supporters claim -- considering two equally qualified applicants and giving the minority candidate the nod. It has instead come to mean admitting Hispanic and African-American students with grade-point averages of 3.2 and SAT scores of 1100, while turning away white and Asian-American applicants with GPAs of 4.0 and SAT scores of 1300. Far from waging a war against discrimination, advocates such as Bowen and Bok find themselves waging a war against merit. And far from vindicating idealism and promoting social justice, they find themselves cynically subverting the principle of equal rights under the law to the detriment of society as a whole.

Before we can decide whether it is simply too embarrassing to permit elite institutions to enroll a very small percentage of blacks or other minorities, we must first ask the question of what produces the racial disparities that so unsettle us and that seem to require affirmative action to counteract. Consider the example of the National Basketball Association. It is no secret that the NBA does not "look like America": African Americans, who are 12 percent of the population, make up 79 percent of the players, while Jews and Asian Americans are conspicuously scarce.

Of course, one never hears demands that the NBA establish a preferential recruitment program for Jews or Asians. But before the notion is dismissed as simply silly, it is instructive to ask why. The answer is presumably that it is merit and not discrimination that produces the racial imbalance on the basketball court. If the coaches hire the best passers and shooters, we tend to think, it shouldn't matter if some ethnic groups dominate and others are hardly represented.

The lesson to be drawn from this example is that inequalities in racial outcomes that are produced by merit are far more defensible than inequalities produced by favoritism or discrimination. And when we turn from the NBA to America's elite colleges and universities, we discover a similar result: Ethnic inequalities are the result not of biased selection procedures but of unequal performance on the part of different groups.

Affirmative action has traditionally been defended as necessary to fight discrimination. But has anyone demonstrated that the blacks and Hispanics preferentially admitted to the best universities were in fact victims of discrimination? Has anyone uncovered at Berkeley or Princeton bigoted admissions officers seeking to exclude minorities? And is there any evidence that the white and Asian-American students refused admission were discriminating against anyone? The answer to these questions is no, no, and no. No one has even alleged unfairness of this sort.

There was, at one time, an attempt by advocates of affirmative action to argue for racial and cultural bias in the SAT and other standardized tests that most elite universities require their applicants to take. This argument, however, has collapsed in recent years, and even Bowen and Bok admit that it is no longer possible to claim that the SAT discriminates against blacks or other minorities. In The Shape of the River, they try to confuse the issue by insisting on the obvious point that standardized-test scores "do not predict who will be a civic leader or how satisfied individuals will be with their college experience or with life." But they are at last forced to the chagrined confession: "Almost all colleges have found that when they compare black and white undergraduates who enter with the same SAT scores, blacks earn lower grades than whites, not just in their first year but throughout their college careers. . . . Tests like the SAT do not suffer from prediction bias."

This is not to say that the test describes genetic or biological ability. It merely measures differences in academic preparation, and Bowen and Bok acknowledge that the low black enrollments at elite universities that affirmative-action policies seek to remedy are primarily produced by "continuing disparities in pre-collegiate academic achievements of black and white students." On those measures of merit that selective colleges use to decide who gets in, not all groups perform equally.

For the civil-rights leadership, these results have come as a nasty surprise. The movement led by Martin Luther King Jr. originally placed itself on the side of merit in opposition to racial nepotism. If laws and public policies were allowed to judge solely on the basis of individual merit, King repeatedly promised, we would see social rewards in America widely dispersed among groups.

In the generation since King's death, it is this premise -- that equality of rights for individuals would invariably produce equality of results for groups -- that has proved false. The dismaying truth is that even merit sometimes produces ethnic inequality. Consequently it is hardly surprising that some who manned the barricades alongside King now insist that merit is the new guise in which the old racism manifests itself. It is now fashionable for advocates of affirmative action to place the term "merit" in quotation marks or to speak sarcastically of "so-called merit." Their main objection is that merit selection is not producing the outcomes they desire, and their enthusiasm for affirmative action can be attributed to their rediscovery of the blessings of nepotism.

Meanwhile, behind the scenes, there has been underway a fascinating debate about why merit produces such ethnic inequality. Two views have dominated the debate. The first is the "bell-curve" position, put forward most publicly in recent years by Charles Murray and Richard Herrnstein, which implies that there may be natural or biological differences between groups that would account for their unequal performance on indices of merit. The second is the traditionally liberal position, which insists that when group differences in academic achievement and economic performance exist, they have been artificially created by social deprivation and racism.

These two views have functioned like a see-saw: When one goes up, the other goes down. In the early part of this century, most people took for granted that there were natural differences between the races and that these accounted for why some groups were advanced and others relatively backward. This view was fiercely attacked in the middle of this century by liberals who argued that it was unreasonable and unconscionable to contend that natural deficiencies were the cause of blacks' doing poorly when blacks were subjected to so much legal and systematic discrimination, especially in the South.

The liberal view was entirely plausible, which is why the biological explanation was largely discarded. But the liberal view has begun to collapse in recent years, precisely as it proved unable to explain the world that resulted from its triumph. Consider a single statistic: Data from the college board show that, year after year, whites and Asian Americans who come from families earning less than $15,000 a year score higher on both the verbal and math sections of the SAT than African Americans from families earning more than $60,000 a year.

This stunning statistic, whose accuracy is unquestioned by anyone in this debate, is sufficient by itself to destroy the argument of those who have repeated for years that the SAT is a mere calibration of socioeconomic privilege. But it is equally devastating to the liberal attribution of black disadvantage to racial discrimination. Even if discrimination were wide-spread, how could it operate in such a way as to make poor whites and Asians perform better on math tests than upper-middle-class blacks?

On this question, most advocates of affirmative action do not know how to react. Some simply refuse to discuss the implications of the evidence. Others, like Nathan Glazer, seem to adopt a private conviction of the veracity of the bell-curve explanation. A few years ago, in a review of Murray and Herrnstein in the New Republic, Glazer seemed to accept the existence of intrinsic differences in intelligence between the races -- while objecting to any mention of the fact in public.

In more recent articles, Glazer has reversed his longtime criticism of affirmative action and said he is now willing to bend admissions standards to avoid the distressing outcome of very few blacks in the best universities. Glazer's second thoughts about affirmative action point to something often missed in such debates, for if the bell-curve thesis is correct, then it in fact constitutes the strongest possible argument in favor of affirmative action.

If there are natural differences in ability between ethnic groups that cannot easily be eradicated, then it makes sense for those of us who do not want America to be a racial caste society to support preferential programs that would prevent the consolidation of enduring group hierarchies. Forced, by the collapse of the liberal view, to accept natural inequality, Glazer unsurprisingly now treats blacks as a handicapped population that cannot be expected to compete against other groups.

But there is, in fact, a third possible view of racial inequality -- a view advanced by Thomas Sowell and me and others who find profoundly condescending and degrading the notion that blacks require a "special Olympics" of their own. Basically, we contend that there are cultural or behavioral differences between groups. These differences can be observed in everyday life, measured by the techniques of social science, and directly correlated with academic achievement and economic performance. Even The Black-White Test Score Gap, a recent study by two noted liberal scholars, Christopher Jencks and Meredith Phillips, proves upon careful reading to implicitly endorse this cultural view. Jencks and Phillips make all the appropriate genuflections to racial pieties, but they are courageously seeking to make the cultural argument more palatable to liberals.

A few years ago, a Stanford sociologist named Sanford Dornbusch was puzzled at claims that Asian Americans do especially well in math because of some presumed genetic advantage in visual and spatial ability. Dornbusch did a comparative study of white, black, Hispanic, and Asian-American students in San Francisco and concluded that there was a far more obvious reason for the superior performance of Asian Americans: They study harder. Asian Americans simply spend a lot more time doing homework than their peers.

Of course, this sort of finding leaves unanswered the question of why they study harder. The causes are no doubt complex, but one important factor seems to be family structure. It is obvious that a two-parent family has more time and resources to invest in disciplining children and supervising their study than does a single-parent family. For Asian Americans, the illegitimacy rate in this country is approximately 2 percent. For African Americans, it's nearly 70 percent.

Such a huge difference cannot easily be corrected. Indeed, in a free society, public policy is limited in its ability to transform behavior in the private sphere. Still, while not reverting to the discredited liberal position, the cultural view of racial inequality is at least more hopeful than the bell-curve acceptance of ineradicable difference: We cannot change our genes, but we can change our behavior.

One thing is clear: Racism is no longer the main problem facing blacks or any other group in America today. Even if racism were to disappear overnight, this would do nothing to improve black test scores, increase black entrepreneurship, strengthen black families, or reduce black-on-black crime. These problems have taken on a cultural existence of their own and need to be confronted in their own terms.

The difficult task is rebuilding the cultural capital of the black community, and the role of black scholars, black teachers, black parents, and black entrepreneurs is crucial. The rest of us cannot be leaders, but we can be cheerleaders. Rather than try to rig the numbers to make everyone feel better, we are better off focusing our collective attention on developing the skills of young African Americans at an early age so that they can compete effectively with others in later life.

So why doesn't this obvious solution win broad support? In his new book, A Dream Deferred: The Second Betrayal of Black Freedom in America, Shelby Steele argues that affirmative action is popular with black and white elites because it serves the purposes of both groups. White elites get to feel morally superior, thus recovering the ethical high ground lost by the sins of the past, and black elites enjoy unearned privileges that they understandably convince themselves they fully deserve. (In The Shape of the River, Bowen and Bok devote several chapters to proving the obvious point that blacks who go to Ivy League schools derive financial benefits in later life as a result and are generally satisfied with attending Yale instead of a community college.)

Steele's book bristles with the psychological insights that are his distinctive contribution to the race debate. White liberals, Steele argues (and he might as well be speaking directly of Bowen and Bok), are quite willing to assume general blame for a racist society causing black failures -- so long as it's the careers of other people, all the qualified Asian-American and white students rejected from Harvard and Princeton, that are sacrificed in order to confer benefits on blacks and win for liberals recognition as the white saviors of the black race.

What Steele is doing -- and it has drawn considerable criticism from reviewers -- is something that advocates of affirmative action have always done: questioning the motives of the other side. For years, conservatives have treated liberals as well meaning in their goals though mistaken in their means. And during that same period, liberals have treated conservatives as greedy, uncaring racists. By asking advocates of preferences what's in it for them, Steele unmasks the self-interest that frequently hides behind the banners of equality, diversity, and social uplift.

Steele's main objective is to show that neither the black nor the white elites have an interest in asking fundamental questions: Isn't color-blindness the only principle that is consistent with the fundamental principles of American society? Isn't equality of rights under the law the only workable basis for a multiracial society? Is the black community well served in the long term by a public policy that treats them as an inferior people incapable of competing with others?

Advocates of racial preferences "offer whites moral absolution for their sins and blacks concrete benefits that are hard to turn down," Steele observed to me a few weeks before the recent electoral victory of a referendum abolishing affirmative action in Washington state. "I think we are going to lose because our side has only one thing to offer, and that is moral principle." I ruefully agreed that the scales were tipped in precisely that way. But the astonishing triumph of the referendum in Washington by a comfortable majority -- like the triumph of a similar measure two years ago in California -- shows that we should not underestimate the power of moral principle in American politics.

When the issue is posed in the basic vocabulary of right and wrong -- a lexicon that is utterly incomprehensible to Bowen and Bok -- the tortured rationalizations of affirmative-action advocates collapse and the common-sense moral instinct of the American people tends to prevail. There is no cause for conservatives to lose their nerve. The election in 2000 could be the moment when color-blindness is at last the issue on the ballot in many states and at the center of the Republican party's agenda.