Assault on affirmative action.

Not long ago I rode on the train back from Detroit within earshot of a white, freelance electrician who works for an international corporate temp agency. In the course of chatting about his work with another white passenger, he complained that his Detroit job had been hampered by the rest of his crow. He knew from the time they showed up that they would be a problem, he said, because, "First of all, one of them was Puerto Rican." Nearly everyone else on the car was black, so he may strained to let the Puerto Rican stand in for other minorities. He then talked disparagingly about how the crew members comported themselves raucously in hotel restaurants and bars - saying nothing at all about the quality of their work.

Later in the conversation, the electrician recounted an incident from a different job, in which a co-worker and buddy locked a donkey in a hotel room as a practical joke. The result was considerable property damage, but he narrated the tale with good humor, as boys-will-be-boys hi-jinks.

This anecdote underscores a problem that permeates American society and has consequences for many public debates - ranging from affirmative action to capital punishment to welfare: the same or comparable actions are judged differently, depending on the racial identities of the actors.

Now that the Republicans and their rightwing Democratic allies have chosen to make affirmative action their racist "wedge" issue of the year, constructing a reasonable, honest, and progressive perspective on affirmative action is particularly important.

Fundamental to the argument against affirmative action is the view that it makes tradeoffs between merit and quotas.

This view rests on two key premises: 1) that it is possible to isolate in the practical world of human choices some pure or absolute notion of merit; and 2) that the criteria that determined access to jobs and schools before the advent of affirmative action depended on such a notion of pure merit.

Of course, there is no idea of merit that is innocent of prejudice. Despite all the blather about "reverse discrimination," the closer we get to real cases, the clearer it is that work, school, and our public spaces are still full of precisely the sorts of exclusionist behavior that civil-rights enforcement seeks to overcome.

Employers and supervisors base their decisions on considerations that include the extent to which they feel comfortable with a given candidate, their sense of how an individual would "fit" in the workplace situation, and on vague feelings and prior assumptions about candidates' general abilities.

As with the electrician, those assumptions are likely to be influenced by racial or gender stereotypes. Not surprisingly, several studies using discrimination "testers" (pairs of applicants matched on all obviously pertinent characteristics except race) recently have demonstrated that, when black and white candidates are identically qualified on paper, blacks are likely to be rejected in favor of whites for jobs, housing, even automobile purchases.

There is no objective measure that exists entirely apart from human prejudice and ideology.

Early psychometricians interpreted the finding that women performed better than men on some components of IQ tests as evidence of flawed design and adjusted the tests accordingly. Why? Because they presumed from the outset that men are smarter than women and that, therefore, a test result that showed the opposite had to have been structured poorly.

Similarly, Stephen Jay Gould's Mismeasure of Man documents the history of "race" researchers' revisions, often inversions, of their indicators of biological superiority upon finding that blacks seemed to score higher than whites on some tests. Pure, "scientific" reasoning decreed that the measures - say, relative lack of body hair - couldn't really indicate preeminence because blacks were clearly inferior.

Women are no longer (at least for the moment) presumed automatically to be less intelligent than men. In fact, intelligence testers have for some time crafted tests to correct for gross gender differences in results. But race differences don't set off the same alarms, because the idea of racial equality in intelligence hasn't yet attained the status of an automatic, incontrovertible assumption. This distinction is highlighted by the fact that the eugenicist book The Bell Curve within four months of publication sold 400,000 copies, and has been treated as respectable science by all-too-many pundits and academics.

The lesson here with respect to affirmative action is that even ostensibly neutral measures can be freighted with prejudice. That is why the U.S. Supreme Court in the 1971 Griggs v. Duke Power decision ruled that employers can use hiring and promotion standards that disproportionately exclude blacks only if they can demonstrate that those standards specifically measure the ability to get the job done. (Bell Curve authors Charles Murray and Richard Herrnstein attacked the Griggs decision early and often.)

It is unquestionably true that until the 1960s blacks directly and explicitly on racial grounds were denied access to public contracts and commercial credit. The effect was that whites enjoyed a competitive advantage that facilitated success in general and gave them a racial monopoly on public contracting in particular.

The same holds for many categories of employment. A white police officer, hired before affirmative action, as likely as not owed his employment and subsequent promotions in part to the existence of a racial and gender monopoly that limited the field against which he had to compete. A white carpenter who belonged to a union that didn't allow black or female members was in the same position.

The benefits accruing from this preferential access to desirable employment, moreover, extend through other aspects of life - from ability to accumulate capital in the form of home ownership and education to, beginning with the New Deal, eligibility for those forms of social welfare targeted to stably employed whites. Employment discrimination intensifies housing inequality, which reinforces unequal access to the political system and public policy, which reinforces unequal access to social resources, which intensifies income inequality, and so on and on.

That beneficiaries of these preferences may have worked hard for their attainments does not negate the equally significant fact that their efforts have been aided by the direct, indirect, and cumulative effects of the exclusion of others no less worthy than they.

Still, affirmative-action foes argue that specific efforts to correct for racial bias fly in the face of "fairness." The relative absence of minorities and women from attractive jobs, and the fact that they are clustered in unattractive positions should not, they say, be taken as evidence of discrimination. The logic of this view is that equal opportunity is guaranteed simply by proclaiming a policy of non-discrimination. Anything beyond that - like setting specific goals and standards to overcome discrimination - amounts to mandating quota systems, they say. And quotas reject "merit."

This complaint has a long, nefarious history. Supreme Court Justice Joseph Bradley's opinion in the 1883 decision overturning the 1875 Civil Rights Act proclaimed - eighteen years after the end of slavery and in the midst of the terrorist reimposition of white supremacy in the South - that it was time that the black American "takes the rank of a mere citizen, and ceases to be the special favorite of the laws."

At bottom, this perspective assumes that the real-world dynamics that concentrate advantages among whites or men don't mean a thing about equal opportunity. Systemic white or male privilege is presumed to be a reasonable, natural expression of merit. This at least verges on an assumption of inherent white, male superiority.

The acceptance of white, male social privilege as natural lay behind Reagan and Bush Administrations' lawsuits challenging affirmative-action plaintiffs to show an actual intent to discriminate - a requirement that is practically impossible to satisfy.

The Supreme Court majority in recent years has relied on an intent standard and has disregarded history and social context in a number of civil-rights rulings. The City of Mobile v. Bolden ruling in 1980 upheld the city's use of an at-large electoral system, even though no blacks had been able to win seats on the city council. The Court held that black underrepresentation could not be taken as sufficient evidence of discrimination. Alabama's - and Mobile's own - segregationist history, according to this interpretation, is irrelevant. The burden lay with those claiming exclusion to prove discriminatory intent.

The 1989 Richmond v. Croson ruling extended this logic to affirmative action in municipal contracting. In throwing out the city's minority set-aside program, the Court ruled that a pattern of extreme underrepresentation (blacks made up more than 50 percent of the population, and fewer than 1 percent of contractors) doesn't constitute evidence of entrenched patterns of exclusion sufficient to justify extraordinary measures to create equality of opportunity.

The 1992 Shaw v. Reno ruling made a comparable argument in throwing out the Congressional-reapportionment scheme that made it possible to elect the first black representatives in this century from North Carolina - a state that as recently as the late 1960s was home to more Ku Klux Klansmen than any other. And in its current session, the Court - in ruling for white firefighters claiming reverse discrimination as a result of a municipal hiring and promotion plan aimed lit approaching parity in black representation - treated as irrelevant even the particularly sordid white-supremacist history of Birmingham, Alabama.

To defend these illogical decisions, opponents of affirmative action have seized on a set of misleading arguments that revolve around the theme that affirmative action actually is self-defeating. One such claim is that affirmative action stigmatizes its beneficiaries as unqualified and therefore creates racist stereotypes, resentment, and thus discrimination. Another is that it instills in those beneficiaries debilitating doubts about both their competence their white, male colleagues' perceptions of their abilities. Yet another is that affirmative action actually restricts opportunity for minorities and women by slotting them into narrow quotas.

The notion that affirmative action creates racism is an old chestnut; it has been applied to every initiative undertake in support of racial equality from teenth Amendment to the Brown decision and the 1964 Civil Rights Act. It collapses before the simple fact that widespread racist practices prompted anti-discrimination efforts in the first place.

The claim that affirmative action is intended to compensate for past suffering - a kind of reparations - has become conventional and is even bandied about by ostensibly progressive pundits. This formulation gives rise to diatribes about how much whose immigrant grandparents suffered themselves and how many of them owned slaves and other such pointless blather.

But the purpose of affirmative action is to remove fetters on equality of opportunity by attacking existing patterns of inequality that originate from current discrimination and the cumulative results of past exclusion and discrimination.

Opponents now often claim that affirmative action is ineffective because it mainly benefits the well-off. But studies by Martin Carnoy (in his 1994 book, Faded Dreams) and Jonathan Leonard (in the Review of Economics and Statistics and the Journal of Human Resources) find a significant positive effect of anti-discrimination efforts for all skill levels among black Americans. And anecdotal information certainly supports those findings.

In the academic world, anti-discrimination policies have induced admissions committees to look more closely at pools of students who would earlier have been dismissed out of hand. And affirmative-action guidelines for faculty-hiring practices have been largely responsible for proliferation of open, nationally advertised searches and regularized procedures that break down nepotism and inbreeding.

The most visible recent "reverse-discrimination" controversy illustrates the weakness of the anti-affirmative-action position in academia. In a case currently pending, Cheryl J. Hopwood v. University of Texas Law School, it turns out that race was not a salient factor. The alleged victims were unlikely to have been admitted even if no blacks or Latinos had applied. Their academic records in fact were mediocre. The plain fact is that the charge of special preferences is a smoke-screen.

The options are not affirmative action or some pristine world in which "merit" is its own reward, but affirmative action or naked discrimination and exclusion.