Some generally interesting things to chew on in this thread. Here are a couple thoughts that I don't think have been adequately made thus far.
First, nothing said so far should be taken as a criticism of the LSAT itself. Many have argued that standardized tests are inherently culturally or racially biased, such as the famous case of the question that relied on an understanding of the word "regatta." While these biases may be present in the LSAT, this has nothing to do with the problem of "stereotype threat." In fact, the LSAT has proven to be the single most reliable predictor of performance in law school, a point which should not be glossed over. The widespread use of standardized tests in the past century was an important progressive accomplishment, playing a key role in transforming our nation's top universities from coddling grounds for East Coast elites into national meritocratic educational institutions. We shouldn't lose sight of that fact, and we should be very cautious of throwing out the baby with the bath water when we criticize standardized testing.
Second, the application of this point to the AA argument as made by red. It is critical to realize that the problem of stereotype threat exists because of deep-rooted cultural attitudes and prejudices, which appear to exert a powerful effect on all people at some level. To put it bluntly, people tend to perform up to or down to the expectations of those around them. Thus, if one believes in the existence of the stereotype threat, one must realize that no test, however perfectly designed to predict law school performance, would ever solve the problem of score differential between racial groups so long as these stereotypes existed. Indeed, the more widely respected the test gets as a predictor of performance, the more pronounced the problem would be. The problem is not in the test, but in social attitudes.
Thus, there is a compelling argument to use Affirmative Action based on simple fairness and accurate prediction of true performance in law school. However, the bitter irony is that AA itself does help to entrench some of the very same racial stereotypes that cause the performance differential in the first place. Make no mistake: AA does send the message that different racial groups can not compete on a level playing ground. While our society puts out this destructive message in many ways, no other way commands the institutional support of affirmative action. This board has ample examples of the rancor and bitterness thrown around by those who believe that some people are getting into schools that they are unqualified for. Such rancor would be impossible if there were no AA.
Thus, we have a conundrum. AA is an important corrective to a biased admissions process, and yet AA contributes to the biases that it is intended to correct. Indeed, the gaps in scores on the SAT have not gotten smaller in the past 15 years, a fact that frustrates the hopes of many, including the Supreme Court, that AA will one day be unnecessary. Solving this problem is a very difficult square to circle. Clearly, we need to do a better job of providing equal access to education from a very young age, but this would not eliminate the stereotype threat. One wonders if perhaps better education about issues like the stereotype threat would be helpful in reducing it. I do believe that affirmative action is an important part of any just college admissions program in a prejudiced country, but I do not think it is in any way a permanent solution. I fear that it may ultimately prove an obstacle to some of the deeper reforms that need to be done.