The purpose of affirmative action is to break down the wall of occupational segregation that excluded racial minorities and women from entire occupational sectors throughout American history. Whatever else one might want to say about affirmative action, it has achieved its policy objective: substantial desegregation of the American workplace, for women and minorities alike. This is true not only in the professions and in corporate management, but also in major blue-collar industries and in the public sector, where nearly one out of every three black workers is employed. If logic and principle had prevailed, we would now be exploring ways to expand affirmative action to industries and job sectors that have been immune to change.

The problem is that “for more than two decades, affirmative action has been under sustained assault,” as Sturm and Guinier write in their opening sentence. Though they do not say so explicitly, they seem resigned to the fact that the Supreme Court, which has already eviscerated affirmative action through a series of decisions, is now poised to deliver the coup de grace. Against this background, Sturm and Guinier declare that “it is time to shift the terrain of debate.” The entire thrust of their argument is to explore alternatives to affirmative action that will broaden access of minorities and women to jobs and universities.

At first blush, this strategy may appear to be a sensible concession to political reality. However, two troubling questions arise. First, are Sturm and Guinier capitulating to the anti-affirmative action backlash and prematurely throwing in the towel for the sake of an illusory consensus? Second, would their proposed reforms of the selection process, even if enacted, provide the access to jobs and opportunities that are today secured by affirmative action?

The logic of Sturm and Guinier’s brief can be stated as follows:

1. Affirmative action is assailed by critics as violating cherished principles of “merit.”

2. On closer examination, the “testocracy” that is used to assess merit is neither fair nor functional.

3. Therefore–alas, here the syllogism runs into trouble. Sturm and Guinier could have concluded that the case against affirmative action is specious and therefore affirmative action should be upheld. As the saying goes, “if it ain’t broke, don’t fix it.”

Instead Sturm and Guinier make a case for overhauling the selection process that evaluates candidates for jobs and college admissions. To be sure, there are compelling arguments for abandoning standardized tests that favor privileged groups who, aside from the advantages that derive from better schooling, have the resources to pay for expensive prep courses. Sturm and Guinier also make a compelling case that it would be fairer and more productive to judge applicants on the basis of performance criteria, rather than scores on “paper-and-pencil” tests. The problem, though, is that they implicitly advocate these reforms as a surrogate for affirmative action policy. They may tell themselves that they are driven by realpolitik, but they end up acquiescing to the reversal of hard-won gains and falling back on reforms that are unlikely to be enacted in the foreseeable future. Their ideological enemies will revel in this retreat to a second line of defense by two law professors who are identified with the cause of affirmative action. Nor will Sturm and Guinier get the concessions they are bargaining for. Is this not the lesson of Bill Clinton’s ill-fated proposal to “end welfare as we know it”?

What evidence is there that overhauling the selection criteria would open up avenues for women and minorities? In most large-scale organizations–corporations and universities alike–employees are routinely evaluated by superiors on an array of performance criteria. Is so-and-so a “team player”? Does she do her job well? Does he have good communication skills? Does she make the tough decisions? Does he demonstrate leadership? Such judgments are easily tainted by personal prejudices, especially when the people doing the evaluations are white and male and the people being evaluated belong to stigmatized groups. Indeed, studies have consistently found that performance appraisal ratings of women and people of color are prone to bias.

A second problem with the “performance-based model” advocated by Sturm and Guinier is that the benefits would be diffused to many groups, and could easily miss African Americans, who were the original targets of affirmative action policy. Besides, how do people demonstrate “performance” when they cannot get their foot in the door?

Finally, Sturm and Guinier place emphasis on jobs “where customers, clients, and perspectives may themselves be identified by race and gender.” Granted, corporations need black managers to interface with the black consumer market, and police departments need women to deal with domestic violence. But women and minorities deserve equal access to all jobs in the workforce. Though Sturm and Guinier endorse this principle, their proposal runs the risk of typing jobs by gender and race, thereby validating the patterns of internal segregation that exist within many job structures.

The lesson of history is that the only mechanism that has ever worked to counteract occupational segregation is affirmative action, and that even good-faith efforts were ineffectual until they were backed up with specific goals and timetables. Too much is at stake to retreat to a second line of defense, especially one so fraught with difficulty. The crusade against affirmative action may well be on the verge of achieving its nefarious objective. But this is all the more reason to remain steadfast in defense of a policy that has not only advanced the cause of justice for women and minorities, but in doing so, has enhanced American democracy. Instead of venturing into the realm of personnel relations and testing, it would have been far better had Sturm and Guinier used their talents as legal scholars to plead the case for affirmative action. As we know it.