To make matters even more fraught, the United States Supreme Court, in a remarkable set of rulings at the end of June, declared for the first time in American history that the constitution is virtually colour-blind and that racial preferences are presumptively unconstitutional unless they are carefully designed as a remedy for past discrimination. And Governor Wilson's rivals for the Republican presidential nomination, including the Senate majority leader, Robert Dole, have pledged to introduce a bill in Congress to eliminate all federal race and gender preferences in a clean stroke. The political, constitutional and moral legitimacy of distributing public benefits on the basis of race and gender, in short, promises to be at the centre of the next presidential campaign; and it may dominate American politics for the rest of the decade.
Why did the debate over affirmative action (or reverse discrimination) become a national issue for Americans in 1995? No one has proposed a satisfactory explanation; but there is clearly a growing frustration, shared by many political liberals, that affirmative action has drifted far from its moorings in the civil rights' movement of the Sixties. In its original incarnation, affirmative action was founded on the ideal of equality of opportunity. As part of his effort to eliminate the effects of state-sponsored racial segregation (which the Supreme Court only declared unconstitutional in 1954), President John F Kennedy ordered all contractors doing business with the federal government to "take affirmative action to ensure that applicants are employed, and that employees are treated during employment, without regard to their race, creed, colour, or national origin". In other words, the opaque phrase "affirmative action" originally prohibited preferential treatment, rather than guaranteed it.
As affirmative action programmes expanded in the Sixties, they focused on broadening the pool of qualified candidates for jobs and university places, to encourage women and minority candidates to apply. But they were justified as temporary measures to move America closer to the ideal of a colour-blind society where all citizens could be judged, in Martin Luther King's phrase, not according to the colour of their skin, but the content of their character.
By the Seventies, however, the national commitment to equality of opportunity had given way to an exquisitely race-conscious focus on equality of results. Because of the lingering effects of segregation, blacks continued to score lower than whites on various standardised tests; and it soon became clear that simply broadening the applicant pool would not appreciably increase the numbers of blacks in universities and workplaces. Through a combination of court rulings and bureaucratic decisions, the federal government at first encouraged and eventually required public universities and employers to apply different standards to black and white applicants, to ensure that each race was represented in numbers roughly proportionate to its population in America as a whole. In a perverse inversion of the colour-blind ideal, failure to achieve proportionate representation was itself viewed as evidence of illegal discrimination.
Employers were forced to adopt increasingly elaborate racial classifications as the numbers of groups claiming they were entitled to special preferences multiplied. Hispanics, American Indians, women, the disabled, even Eskimos and Aleuts - all insisted that they had suffered historic discrimination which entitled them to special consideration. As a result, the unique historical injustice perpetrated against black Americans through slavery and then through segregation was increasingly obscured in the jumble of special-interest politics. In their eagerness to achieve proportionate racial representation, many employers abandoned the time-consuming effort of considering the individual merits of each applicant and, instead, adopted crude racial quotas. In 1978 the Supreme Court struck down an affirmative action programme, administered by one of the University of California's medical schools, which set aside 10 per cent of all places in its first- year class for "Blacks", "Chicanos", "Asians" and "American Indians". Race could be used as a goal or as a "plus factor", the court announced in a famous euphemism, but rigid racial quotas were illegal and unconstitutional.
In 1995 the euphemistic distinction between "goals" and "quotas" finally collapsed under the cold weight of empirical evidence. The University of California, for example, recently made public the dramatic gaps in test scores that continue to separate black and Hispanic applicants from white and Asian applicants. Last year Asian high-school students admitted to the University of California at Berkeley, the crown jewel of the California system, had a mean grade point average of 3.95 (out of a perfect 4.0). Whites had a 3.86 average; Hispanics, 3.65; and blacks had 3.43. The university estimated last week that if it removes race as a consideration in admissions' decisions, as Governor Wilson's new policy requires it to do, the numbers of black students will be cut in half (from about 4 per cent to 2 per cent of the entire campus). Hispanic enrolment will drop from about 13 per cent to 11 per cent; but the percentage of Asian students will increase dramatically, from about 30 per cent to 40 per cent of the class, or higher.
Few American politicians have been rash enough to venture a public explanation for why these distressing gaps in test scores have persisted, nearly 30 years after the first affirmative action programmes began. One explanation, privately popular among Democrats, is that blacks and Hispanics score lower on tests because they tend, on average, to come from poorer families. Earlier this year President Clinton toyed briefly with the possibility of replacing racial and ethnic preferences with class preferences. In other words, the son of a coal miner would have a better chance of getting into the University of California than the son of a venture capitalist. But this proposal was quickly abandoned when it became clear that the average college board admissions' test score for black students whose parents earn more than $70,000 a year was 20 points lower than the average score for white students whose parents earn less than $20,000. Some Republicans, for their part, have flirted with the uglier forms of 19th-century racialism and social Darwinism. Charles Murray's The Bell Curve became an American bestseller a few months ago by purporting to prove that Asians and whites were genetically predisposed to be cleverer than blacks.
In this political and intellectual minefield both President Clinton and his Republican opponents are treading gingerly. All of the leading Republican presidential candidates have called on Congress to pass a law eliminating the race and gender preferences that the federal government administers. (There are, by the latest count, more than 100 of these programmes, ranging from bidding preferences for women and minorities who want to buy cable television stations to special assistance for banks owned by minorities and women.)
But even the Republicans, when pressed, fear the radical consequences of eliminating affirmative action with the stroke of a pen. Newt Gingrich, the loquacious Republican Speaker of the House of the Representatives, recently declared that his party has a moral obligation to propose alternatives to racial preferences before calling for their abolition. Otherwise, Gingrich said, the Republicans could be accused of cynically exploiting the race card in their eagerness to marginalise the Democrats as the party of minorities.
President Clinton, for his part, deserves at least grudging credit for remaining true to his instincts, even if his emotional defence of affirmative action proves to be a kind of political suicide. He indicated six months ago that he was ready to repeal some of the most controversial racial preferences, and to share the pain of "angry white men". But his speech last week was an unapologetic affirmation of the liberal orthodoxy. The only concession Clinton made to the growing chorus of critics was that "unqualified" people should not be hired over "qualified" ones. But, of course, the entire debate about affirmative action takes place in the agonisingly grey area where less qualified candidates are preferred to more qualified ones.
By committing himself so passionately to the old Democratic position on affirmative action, Clinton now seems a little like the leaders of the unreconstructed Labour Party during the Thatcher years, sentimentally committed to the rhetoric of socialism even as the political landscape shifted under their feet.
What makes Clinton's defence of affirmative action perversely admirable or politically insane, depending on your perspective, is that he could have avoided the consequences of strongly endorsing or attacking affirmative action entirely by passing the buck to the Supreme Court. The court's recent rulings make clear that at least five of the nine justices, a slim conservative majority, are willing to tolerate racial preferences only in the narrowest circumstances, as a remedy for proven examples of past discrimination. If the court means what it says, then much of affirmative action as it is currently practised will be struck down by judges over the next few years, even if Congress and the President do nothing at all.
After all, few racial preferences in 1995 can be linked to the present effects of past discrimination with any kind of empirical rigour. The proliferation of protected groups (which now include Native Hawaiians, Samoans, Northern Mariana Islanders, and so forth) and the rising numbers of recent immigrants have stretched the remedial theory to breaking point. It's simply not plausible to pretend that a federal highway contract awarded to an aristocratic immigrant from Spain in the 1990s can be linked to President Eisenhower's discrimination against Mexican Americans in the Fifties.
But Clinton did not ignore the Supreme Court entirely. On the same day that he delivered his rousing defence of affirmative action, he directed all agencies in the federal government to review their affirmative action programmes in light of the Supreme Court's recent decisions. Therefore Clinton may be left with the worst of all worlds: he has committed himself, body and soul, to defending racial preferences, and will have to suffer the political consequences in 1996. At the same time he may be forced by the courts and Congress to dismantle the very programmes he has gone out of his way to preserve.
Whether or not the presidential election turns on affirmative action, the political question ultimately pales before the moral question. Is America ready to become a colour-blind nation for the first time since the Civil War? The colour-blind principle at least has the virtue of political and moral clarity. There's something compelling about the bold proposition that racial discrimination of any kind should be immoral, illegal and unconstitutional, and that discrimination in favour of blacks is just as paternalistic and stigmatising as discrimination against them.
And yet few American conservatives, when pressed, are willing to accept the practical consequences of the colour-blind principle, which would mean in many cases a return to virtually all-white universities and workplaces. In a country where whites will be an ethnic minority in some states by 2000, widespread resegregation is not really a prospect anyone is willing to contemplate.
Unfortunately, President Clinton's decision to defend racial preferences as a moral imperative rather than a necessary evil appears even harder for the majority of American citizens to accept. And so, despite his obvious personal commitment to fighting racial discrimination (a commitment that none of his opponents has yet displayed), Clinton may not remain in office long enough to preside over the transition to the colour-blind era that is shimmering dimly on the horizon.
The writer is legal affairs editor of 'The New Republic'.