But what he does not adequately address is the intrinsic unfairness of the direction in which he urges us to move. Klitgaard persuasively provides a framework for choosing elites, yet he does not account for the possibility that such a framework would systematically and unfairly perpetuate such at, elite. He tells us that admissions officers cannot avoid questions of just desert in drawing up policies, but points us in directions that would further skew society's distribution of fruits and rewards.

KLITGAARD'S SECTION on affirmative action focuses on this troubling dilemma. Here he restates the most controversial findings of the Klitgaard Report, and while he is careful to qualify his conclusions, he minces no words. For unknown reasons, he states, at the right tail of academic qualifications, "there are surprisingly large differences in the performance of various ethnic groups." In 1983, for instance, 570 Blacks had combined SAT scores above 1200, compared to 60,400 whites--numbers which are also disproportionate to the numbers of the groups taking the test. More troubling, according to his research, standardized test scores tend to overpredict the academic performance of Blacks at the right tail.

To his credit, Klitgaard does not take these figures as an excuse for universities to end their affirmative action programs. His purpose, again, is not normative, but rather to set out such a policy's costs and benefits. He is extremely sensitive to the correct benefits of affirmative action--in terms of educational diversity and creating incentives for disadvantaged groups. His findings suggest the general factors which universities must consider in setting up their affirmative action policies. The problem is that he understates the extent to which the educational system itself undercuts the opportunities of Blacks and students in lower socio-economic classes.

The proposition that standardized tests are not biased against Blacks is only true in the narrow sense that they do not underpredict their later academic success. But Klitgaard does not examine the other possible biases of these test although there is a growing body of evidence demonstrating them. Journalist David Owen's recent devastating attack on the validity of the SAT, for example, shows that contrary to popular assumptions, students can be coached to do better on standardized tests--perhaps even to the tune of 100 points. Not withstanding the considerable doubt such findings cast on the notion of the SAT as objective measure of "aptitude," it is clear that the benefits of coaching are only available to a select few who can afford it.

Generally speaking, Klitgaard's book is dogged by the possibility that he is merely setting up a framework for the perpetuation of the current elite structure in this country. Two stark facts stare from his analysis. One is that test scores and grades are the only indicator that can satisfactorily predict academic success in college. The other is that, this said, there is little that can help us predict success in later life. But isn't this latter kind of success exactly the kind of success we are most interested in fostering? Because this type cannot be predicted adequately, Klitgaard seems to say, let's ignore it. Yet this seems self-defeating, as we may then be forced to reward people for the wrong reasons according to some unfair standard of academic "merit" whereby some members of society have unequal advantages.

The danger exists, under Klitgaard's suggestions, that we may be tyrannized by excessive devotion to a flawed method of selection in order to select an elite class based on the wrong principles Klitgaard aptly quotes psychologis David McClelland on this point: the testing movement is in grave danger of perpetuating a mythological meritocracy in which none of the measures of merit bears a significant demonstrable validity with respect to any measures outside the charmed circle.

of academic performance. One wishes the author had further developed this point, for it is a crucial one. It is indeed difficult to imagine few more pressing social problems than how we select our future elite--both effectively and fairly. Klitgaard superbly addresses the practical problems of effective choice. But if the net result of this effort is an affirmation of a flawed status quo, then, one wonders, why bother?