Sandra Day O'Connor
Sandra Day O'Connor; drawing by David Levine

Table A

Table B

Table C


Thanks to Sandra Day O’Connor, affirmative action is still alive. Race can continue to be a factor in university admissions, although not as explicitly as it sometimes was in the past. But resistance to affirmative action remains strong, particularly within the Bush administration. The Court’s recent decision will very likely face serious challenges, some of them hardly mentioned in comments on the decision so far.

By now the principal findings of the Court are well known. In the decision concerning the University of Michigan Law School, Grutter v. Bollinger, race can be a “plus” in law school admissions, so long as every candidate receives “individualized” scrutiny. In Gratz v. Bollinger, Michigan’s method for admitting undergraduates was held to be unlawful since it automatically gave extra points to persons of certain racial or ethnic origins. The books under review are useful for understanding the challenges to the Court’s decision that are bound to arise.

Affirmative action largely concerns race. Michigan and other universities call Hispanics and American Indians “underrepresented minorities,” but the primary aim of preferential policies has been to increase the number of black students on campuses. In 1994, a white applicant to the University of Texas School of Law named Cheryl Hopwood filed a suit contending that she had been rejected by the law school because of her race. She found that black candidates with lower grades and scores than hers had been given places, one of which, she felt, should have been hers. The school replied that it had the right to encourage diversity among its students.

Two years later, however, a federal appeals court upheld Hopwood’s suit, saying that the school “may not use race as a factor in deciding which applicants to admit in order to achieve a diverse student body.” It based its decision chiefly on a new reading of both the Fourteenth Amendment’s “equal protection” clause and the Civil Rights Act of 1964, where the operative words are “no person shall, on the ground of race, color, or national origin, be subjected to discrimination under any activity receiving Federal financial assistance.”

The original aim of both the Fourteenth Amendment and the Civil Rights Act was to protect black Americans, who were being denied opportunities and benefits because of their race. Even so, the appeals court ruled that the law school had abrogated Cheryl Hopwood’s right to “equal protection” since she was discriminated against for being white. Under no circumstances, the court held, could race, white or black, be used to further academic ends. Lawyers for the white plaintiffs in both recently decided Michigan cases made this their leading argument. In the Grutter case, they persuaded only four of the Supreme Court justices, and as a result Hopwood was overturned. The decision had barely been released when the University of Texas announced it would reinstate the affirmative action programs it has been barred from using for the past seven years.

In 1996, California also set a precedent that will now be getting more attention. In a referendum, white voters were largely responsible for passing Proposition 209, now part of the state’s constitution, by which race cannot be considered in any activities of the state, including its universities’ admissions and faculty appointments. Two years later, the state of Washington’s voters passed a similar amendment. These actions mean that public colleges and universities in both states do not have the leeway to use affirmative action as was granted in the Michigan law school decision, which permits race to be considered in admissions only if a university wishes to employ it. But California and Washington’s voters have already said they do not want it used in schools that their taxes support.1 If those relatively liberal states could take such a step, it seems likely that at least a few others will now be inclined to do the same.


So the battle is far from over. Indeed, two weeks after Grutter was decided, plans were announced for a Michigan referendum that would negate the Court’s ruling for the state. The same day, Colorado’s governor called for state legislation having that effect. It will, moreover, take only one new Supreme Court justice to overturn or modify the Michigan law school decision. Conservative lawyers can be counted on to file suits against admissions offices they feel are ignoring the Court’s ruling against Michigan’s use of affirmative action for undergraduates. Thus Michigan could be called on to prove it gives “holistic” attention, as Justice O’Connor put it, to each one of its 24,141 undergraduate applicants,2 which it may have a hard time doing. In these and other respects, Peter Schuck’s Diversity in America suggests what form the debate may take.

While Schuck also discusses immigration, residential segregation, and religion, affirmative action is at the core of his book. He has written what is basically a lawyer’s brief opposing race-based programs. Indeed, Anthony Kennedy’s dissent in the Michigan law school case cited Schuck’s earlier work. Schuck, a professor at Yale Law School, writes that he has no quarrel when a state supports interests and identities that have “already been defined, authenticated, and valorized by civil society.” But he deplores what he sees as a “socially inflammatory” emphasis on race, which he regards as “the worst imaginable category around which to organize group competition and social relations more generally.” Even if “race” in common usage refers to a social reality, Schuck worries that a “relentlessly racialist rhetoric” has given it excessive emphasis.


Thus he underlines that those who want to make campuses more varied “almost always mean ethno-racial diversity.” Selective schools, he says, have done little to attract students who are politically conservative or devoutly religious, and here he is on strong ground. (How many have enough undergraduates to form a “right to life” chapter, for example?) However, he then asserts that adding black students need not augment intellectual diversity, since some may have views and social advantages resembling those of white students. But here Schuck treats African origin simply as a census category rather than as an imposed condition that affects a person’s life. As Ronald Dworkin has pointed out in these pages, “The experience of a black person in American society is special, and cannot be duplicated by the experience of a white person of similar economic or social background.”3 Hearing oral arguments in a cross-burning case last year, Clarence Thomas suddenly made a comment about racial prejudice resembling Dworkin’s. But his animus toward affirmative action demonstrates that racial perspectives are by no means uniform.

Schuck’s argument is one commonly made on behalf of plaintiffs like Hopwood, Gratz, and Grutter, who claim that they were penalized because of preferences based on race. At issue, he says, are “people’s claims to be judged as individuals, not as members of ascribed groups.” Insofar as being black becomes an asset for admissions or employment, being white is turned into a liability. (Indeed, Barbara Grutter described herself as belonging to a “disfavored racial group.”) Implicit in Schuck’s book is a new constitutional concept, one that could be called “white rights.”

In the two Michigan cases, as with others at lower levels, the plaintiffs were marginal applicants, at or near the bottom of the “white” lists of candidates. (Needless to say, colleges will deny that racial rankings exist.) This may be one reason why the Bush administration filed briefs on their behalf in the Michigan cases, and why attacking affirmative action has become a strong conservative cause. White voters are Bush’s core constituency. In 2000 Bush received a majority of white votes, some 54 percent. Since the party’s greatest loyalty is to the better-off, the Republicans face the question of what to do for the less affluent, who are less likely to attend schools with strong college preparatory programs. This usually means that they end up with lower test scores. If they are rejected by a state university, they can be encouraged to believe that someone black was given “their” place. In fact, when whites are displaced, it is largely by Asians whose admission cannot be criticized because they tend to have better scores and grades. It is all too easy to transfer resentment onto blacks.

The Court’s rulings in the Michigan cases apply only to public institutions, which make up about 40 percent of all colleges, but account for more than three quarters of all enrollments. For all of Schuck’s objections to race-based preferences, he allows that private universities “should be able, as a legal and policy matter, to make many choices that the government may not properly make.”4 Here his argument seems unconvincing. Such universities as Stanford, Duke, and MIT are national institutions, with substantial influence on American society, and they are heavily dependent on public funds. Why shouldn’t they be subject to the Civil Rights Act?

Public universities, moreover, now collect as much from student tuition as they get in legislative grants. Schuck, for his part, doesn’t seem bothered by the fact that public law schools that cannot use affirmative action have one third to half as many black students as others that can. (See Table A) Nor does he say how he feels about the student body of Yale Law School, his employer, having an ongoing average of 9.3 percent blacks. It would have been interesting to hear his assessment of their academic performance, and his views on how their presence has affected the school’s atmosphere. Does he feel that some of them should have heeded his advice to black students generally, which is to drop “down a rung on the academic ladder” and not compete for places in the more selective universities?


It should be self-evident that people shouldn’t suffer discrimination because of their “race, color, or national origin” (Title VI’s categories) since these are conditions imposed on them at birth.5 Schuck feels strongly that to benefit from these circumstances is equally undeserved. In this, he echoes Clarence Thomas, who held in his dissent that such preferment “presumes that one needs special help because one cannot make it on one’s own.” My own suggestion is to add another birth condition to the list—call it “parentage”—to denote the benefits you get from being identified with and raised by your parents. Some of the opinions in Gratz v. Bollinger mentioned white candidates in Michigan who received extra points for being children of alumni, for having lived abroad, or for being from an “underrepresented” Michigan county.

Schuck has nothing to say about these kinds of advantages. Nor does he conclude that getting admitted thanks to parental influence or prestige will impose a “stigma” (Thomas’s term) on such students in the eyes of their classmates. (We don’t hear that George W. Bush slouched in shame around Yale’s quadrangles. In my own case I suspect I got into Amherst mainly because my father was a professor, and colleges like a leavening of faculty offspring.) The truth is that students seldom know who exactly benefited from one preference or another; and the specter of their being humiliated or snubbed because they benefited from affirmative action seems largely fantasy. In any case, giving extra points for belonging to a racial caste may be seen as offsetting the advantages of those who have parents able to ease their way in.

Most of us agree that the notion of “race” is a human creation, with no basis in genetics or biology. Still, what people take to be race continues to determine attitudes and prejudices. This isn’t simply a residue of the past that is too ingrained to be willed away. Maintaining the idea of race serves many political purposes. So even if racial classifications don’t have a genetic basis—Glenn Loury calls them “taken-for-granted social meanings”—references to the black and white races are part of the American language and have obvious effects on people’s lives.

I find it useful for purposes of analysis to start from the premise that America has only two races, consisting of persons of European and of African origins. As can be seen in Table B, people of the two races now make up 81.1 percent of the population, down from 94.2 percent a generation ago. Among those who call themselves white, most still feel able to point to a single ancestral country, like Italy or Poland or Greece. However, 26,508,073 now say their background is too mixed to be represented by any single country. And while 33,707,230 persons said they were simply black, it is revealing that only 784,765 others told the Census that they were also partly white—which in fact virtually all black Americans can truthfully say.

After inquiring whether respondents are black or white, the Census stops asking about colors, although the remaining 18.9 percent of Americans make up the fastest-growing group in the US. It asks people whether they are Latino or Hispanic, terms that are used interchangeably, and, in 2000, a total of 35,238,481 said they were. When asked to identify themselves as a member of a race, 47.8 percent wanted to be called white, and 42.5 percent indicated no racial affiliation. Indeed, to be Hispanic is to be a member not of a race but simply of a dem-ographic category including black Dominicans, white Argentineans, and persons of mixed racial backgrounds.

Those who are called Asian are also much too varied to be called a single race. Among those who identify themselves solely as Asian, according to the Census, are 2,432,585 persons of Chinese origin and 1,076,872 Koreans, as well as 153,533 Pakistanis and 20,145 Sri Lankans. Like Hispanics, this is also a category created by the Census for demographic convenience.

The 2,423,531 persons who call themselves American Indians prefer to be known by their tribes. The Census offers a long list from which to choose, so we are told that there are 281,069 Cherokees, 105,907 Chippewas, and 10,120 Comanches. The Census also found 1,022,092 Arab-Americans, as well as 151,006 Guyanese, and 71,816 Cape Verdeans. Virtually none of these individuals wanted to be designated a member of a race, any more than do the Indians, Hispanics, and Asians. Apparently they find it prudent to remain on the racial sidelines.


Among books that consider race in America, it’s almost as if that status applies only to blacks; whites are often characterized by their religion and by their various countries of origin. Richard Alba and Victor Nee do much to dispel this emphasis in their Remaking the American Mainstream, a humane and imaginative book which combines social analysis with historical understanding. They examine how different groups have increasingly come to share a common culture, a melding that now happens at a faster pace than it ever has in the past. Not the least reason is that even immigrants from the other side of the globe arrive here already familiar with American ways. (Note how quickly taxi drivers from Nigeria and Bangladesh grasp the folkways of Dallas and Kansas City.) The authors call this mainstream “Anglo-American,” which is itself instructive, since it is the phrase Alexis de Tocqueville used 170 years ago. And they feel it remains appropriate, even if only one in eight Americans admits to forebears who were English, Scottish, or Welsh.

This common culture has absorbed Chinese cuisine, Jewish humor, music derived from Africa, and a steady flow of other consumer goods and items of popular entertainment. But even more striking is the number of nationalities the common American culture has erased. For example, the largest immigrant group in the US came from Germany. Yet today it is hard to identify its impact on the larger society, or to what extent its culture endures in its descendants. Alba and Nee show how intermarriage, mobility, and new middle-class careers have created third or later generations in which ethnic distinctions are largely erased. If you walk through the suburbs of Minneapolis or Denver, you will largely see post-ethnic Americans who have a variety of surnames but are best described simply as white (and American).6

In the same vein, many non-Europeans are now designated as white. Indeed, in 2001, the Census quietly decided to classify people from North Africa and the Middle East as such, giving Egyptians and Lebanese the option of calling themselves white if they wanted to. And today, few object when Hispanics also say they are white, as almost half of them do. My own observations suggest that such people increasingly qualify as white if they speak fluent English and show a readiness to adapt to Anglo-American ways, which most non-Europeans are willing to do. That does not mean, for example, that English-speaking Hispanics are accepted as social equals in many parts of the US. But the proportion who marry whites of European origin—about one third—is growing. White America may see itself as needing new recruits, if only because the fertility of white Americans is at an all-time low. Last year, they provided only 57.3 percent of the nation’s births. So it is worth asking which other ethnic groups will soon be characterized as white.

A good clue, Alba and Nee say, is the “ready acceptance of intermarriage between whites and Asian Americans.” They note that 61 percent of men of Chinese, Korean, and Japanese ancestry marry non-Asians, almost entirely whites, while for women the figure is 66 percent. Moreover, Asians can move almost anywhere they like. In Los Angeles, the typical Chinese-American family lives in a Census tract where only 14 percent of their neighbors are also Chinese. This absorption leads the authors to see “a break with the conventional equation of the mainstream with white America.” I would interpret the evidence differently. The term “yellow” has been allowed to disappear. As a result, those who used to be described as such will in time be perceived as white. No longer labeled “Oriental,” many take Anglo-American first names, and most speak flawless English. When they run into prejudice, it may be because they do so well, whether in school or in business.

Still, Alba and Nee write, while many “immigrant minorities approach residential parity with whites,…that is not true for socioeconomically successful African Americans.” Even middle-class black professionals “are more likely to reside in areas where their own group is in the majority and white neighbors are rare.”7 In all, Remaking the American Mainstream confirms that America’s basic division remains sharply drawn, between black and white. In the eyes of the larger society, children born of a white and a black parent are seen as, and called, black. Regardless of proposals for a “mixed race” category, there are few attempts to use it, since even a fractional black origin is considered enough to be the overriding component of a person’s identity.8


If Peter Schuck wants us all to be “judged as individuals,” it remains to ask how that judgment should be made. If the influence of parentage is to be ignored, as it in fact is not, the standard usually invoked is “merit,” which in educational admissions usually means a superior record in school and on tests. (Ivy League schools claim that they expect athletes to meet the same standards as their more sedentary classmates.) Thus the majority in the University of Michigan’s undergraduate college case and the dissenters in the Michigan Law School case objected that whites with higher scores were turned down, while blacks who ranked lower were admitted. The measures they cite are the Scholastic Assessment Test (SAT) and the Law School Admission Test (LSAT). Each yields a three-digit number, denoting where one stands among all those who took the tests.

Test results have become the chief criterion of merit for obvious reasons. Owing to grade inflation, transcripts are untrustworthy. A single test also sorts out students from more and less demanding schools. Letters of recommendation are generally effusive, and essays have often been polished with help from parents. Hence the conclusion that the most objective measures of merit are provided by multiple-choice, machine-graded, standardized tests. From this follows the view that those who can’t succeed at such tests should not be admitted on the basis of other criteria.

The problem is that, thus far, relatively few black students have excelled in these tests. Sandra Day O’Connor concluded the Court’s Grutter decision with the hope that “25 years from now, the use of racial preferences will no longer be necessary.” Clarence Thomas apparently agreed, since he opened his dissent by affirming that “blacks can achieve in every avenue of American life.” Yet, later in his opinion, after reviewing LSAT scores over a span of years, he had to admit that he saw no “evidence that the gap in credentials between black and white students is shrinking.” Each year, he pointed out, about 11 percent of whites have scores in the test’s highest category, while only 1 percent of blacks do. Nor was he optimistic that this gulf would narrow in the decades ahead.

A perennial question concerns the respective roles of class and race in achieving success in education and in employment. When it comes to economic advantage, SAT figures show a close connection between students’ scores and their family’s income. Nor is this surprising, since having money opens the way to better schools with superior preparatory programs. But the racial imbalance is even more pronounced than white economic advantages would suggest. As Table C shows, within identical income brackets, black students have discernibly lower scores than their white counterparts. Indeed, those whose parents make $80,000 to $100,000 have lower average scores than whites in the $20,000 to $30,000 bracket. It seems fair to assume that black families in the higher range send their children to reasonably good schools.9 So apparently even economic progress may not lead to parity in academic performance, at least as measured by standardized tests.

There are several reasons for this result, but at the center is the racial isolation described by Richard Alba and Victor Nee. Black Americans at all economic levels spend more of their lives among themselves than members of most other minority groups do. Even those with middle-class jobs tend to live in neighborhoods where they are surrounded by members of their own race. The few who are in integrated schools still tend to see more of their black friends and their families, both of whom may reinforce cultural and intellectual styles that put them at a disadvantage in competition with whites.

It is true that professional parents may succeed by adapting to white ways at their places of work. Yet even their children may absorb modes of perception and expression that are barriers to mastering the multiple-choice format and the approach to knowledge on which it is based. So long as sharp racial separations persist, and examinations like the ones now used continue to be crucial for admissions, some kinds of preferential programs will be required if our nation’s most visible race is to have more than a token presence on the nation’s campuses.

This Issue

August 14, 2003