Stanley Fish is professor of English and law at Duke University. This article appeared in The New York Times December 26, 1996. Copyright © 1996 The New York Times Company. Reproduced by permission.

Suppose you were arguing for something but were told that you wouuld have to make your case without the facts that supported it. This is the situation proponents of affirmative action face when they find themselves defending their position in terms of principle rather than policy.

A policy is a response to actual historical circumstances; it is directed at achieving a measurable result -- like an increase in the representation of minorities in business and education. A principle scorns actual historical circumstances and moves quickly to a level of genralization and abstraction so high that the facts of history can no longer be seen.

Affirmative action is an attempt to deal with a real-world problem. If that problem is recharacterized in the language of principle -- if you stop asking, "What's wrong and how can we fix it?" and ask instead, "Is it fair?" -- the real world fades away and is replaced by the arid world of philosophical puzzles.

The recipe for making real-world problems disappear behind a smokescreen of philosophizing was given to us years ago by the legal scholar Herbert Wechsler in his enormously influential Harvard Law Review article "Toward Neutral Principles." Wechsleer was trying to justify the Supreme Court's decision in Brown v. Board of Education, which declared segregated schools unbcinstitutional. What troubled Wechsler was that the Justices, in reaching their decision, seemed moved by the practical desire to secure a result they favored (integrated schools) rather than by some general principle whose application would yield that result independently.

Unable to find any such principle spelled out in the Court's arguments, Wechsler was driven to provide one himself: the "right of freedom of association." But in attempting to make this case he soon realized that the principle of freedom of association turned out not to justify Brown but to make it even more of a puzzle. "If the freedom of association is denied" by segregation, integration forces an association upon those to whom it is . . . repugnant," he wrote. And "given a choice between denying the association to those . . . who wish it and imposing it on those who would avoid it," he was unable to find a principle that would justify either the one or the other.

Here in as naked a form as one might like (or not like) is the logic of neutral principle. When Wechsler characterizes the choice as being between the rights of those who wish to associate and the rights of those who wish not to, these two wishes have lost all contact with the issue that made their opposition meaningful -- whether the schoolhouse door should be open or shut. Once the historical specificity of that issue is lost, there no longer seems to be any moral difference between the two sides, although the difference was perfectly clear before Wechsler began his tortured analysis.

In other words, the puzzle of Brown is only a puzzle if you forget everything that made the case urgent in the first place -- the long history of racism and its effects. You have substituted philosophical urgencies for social urgencies. This is what the demand for principle does, and what opponents of affirmative action intend it to do. After all, isn't it convenient to be able to deny a remedy for longstanding injustices by invoking the higher name of principle?

It is a very bad game, but it is alive and well in the phrase "reverse racism," which does in an instant what Wechsler neded an entire essay to do. The phrase makes the actions of college admissions officers who give preference to minority candidates equivalent to the hate crimes of the Ku Klux Klan. It does so by claiming that each is motivated by race-consciousness, an argument that makes sense only if the very thought of race, no matter the content or context, is considered the sin. Like the freedom of association in Wechsler's argument, race-consciousness invoked as an abstraction rides roughshod over history while laying claim to the noblest of motives.

That is in effect what Justice Clarence Thomas did in his concurring opinion in Adarand v. Pena, in which the Court struck down the policy of giving incentives to Federal contractors who hired minority subcontractors. "It is irrelevant," he wrote, "whether a government's racial classifications are drawn by those who wish to oppress a race or by those who have a sincere desire to help those thought to be disadvantaged. In each instance, it is racial discrimination, plain and simple."

But both the plainness and the simplicity are apparent only if the complex facts of history have been suppressed or declared out of bounds. In his dissent, Justice John Paul Stevens returned to history to make the truly plain and simple point: "There is no moral or Constitutional equivalence between a policy that is designed to perpetuate a caste system and one that seeks to eradicate racial subordination."

The important word in Justice Stevens's statement is "moral," for it shows that the choice here is not between the principled and the non-principled. It is between neutral principles, which refuse to acknowledge the dilemmas we face as a society, and moral principles, which begin with an awareness of those dilemmas and demand that we address them.

Those who favor affirmative action are moved by moral principles -- principles that recognize the reality and persistence of historical inequities. And yet those who favor affirmative action are often maneuvered into using a vocabulary designed to remove from sight the very realities on which their case depends.

Of course, you could also try to work within that vocabulary and fight over its terms, arguing that "fairness," equality," and "color-blindness" really belong on your side. But even if you got good at the game, you would be playing on your opponent's field and thus buying into his position, and why would you want to do that?

It would be far wiser to refuse the lure of "fairness," "merit," and "equality," now code words for ignoring the effects of the long history of racial oppression. Let's be done with code words and concentrate on the problems we face and on possible ways of solving them. Those who support affirmative action should give up searching for theoretical consistency -- a goal at once impossible and unworthy -- and instead seek strategies with the hope of relieving the pain of people who live in the world and not in the never-never land of theory..

Let us stop asking, "Is it fair or is it reverse racism?" and start asking, "Does it work and are there beter ways of doing what needs to be done?" Merely asking these questions does not guarantee that affirmative action will be embraced, but it does guarantee that the shell game of the search for neutral principle will no longer stand between us and doing the right thing.