Peter Bozzo has posted a very thorough reply to my reply to his column which argued that we should replace race-based affirmative action with class-based affirmative action. (Got that?)
Peter cites three studies, but only one of them seems to have analyzed the socioeconomic backgrounds of African-American college students, as opposed to the socioeconomic backgrounds of college students of all races. That study does seem to provide strong evidence that affirmative action primarily benefits middle- and upper-class African-Americans.
Okay, I said that I would support a switch from race-based to class-based affirmative action if that were true, and I will stick with that pledge.
However, as an excellent article that Peter co-wrote this fall explained, whether class-based affirmative action would actually help low-income African-Americans really depends on how you define “class.” If schools looked only at income in assessing an applicant’s “class,” African-American college enrollment would plummet, because there are more complex socioeconomic factors, like family wealth, neighborhood wealth, school-district poverty, etc., that also affect a student’s educational outcomes.
Peter probably thinks that schools should go ahead and look at these more complex socioeconomic factors, and I’d agree with that. My question for him is whether he thinks that would really be a “color-blind” policy.
Sure, it would be color-blind in formal terms, and judging from Peter’s interpretation of Brown v. Board, it seems he might be a bit of a formalist. But everyone would understand what the point of this policy would be: Schools wouldn’t get enough minority students if they looked only at income, so they decided not to do that. They’d look instead at close substitutes for the allegedly unacceptable racial classification.
So Peter would say, in response, that the socioeconomic classifications are relevant in a way that the racial classifications are not. And therefore it’s okay for schools to look at those socioeconomic classifications, but not at the racial classifications. But it is simply untrue that race is irrelevant when it comes to evaluating the educational opportunities that a student was likely to have been afforded. I mentioned in my first post just one way in which race makes an appearance: Research has shown that teachers and guidance counselors “track” minority students into less advanced classes. So, race and class are obviously both important factors, but the latter does not completely explain the educational disadvantages faced by African-Americans.
I find Peter’s SAT statistic (blacks and whites differ by only 50 points, but high- and low-income students differ by 400) entirely unconvincing in its attempt to show that class is more important than race in determining educational opportunity. (At least that seems to be what Peter wants to show.) The income gap between whites and blacks is big, unacceptably big, but it is a lot smaller than the income gap between the highest and lowest-income groups. Now, maybe if Peter showed that the difference between blacks and whites on the SAT was identical to the difference between the scores of whites of a certain income bracket and whites of a somewhat higher income bracket (the income gap between these two groups of whites being equal to the gap between median black and median white incomes), then I would be convinced.
Finally, I need to respond to Peter’s argument that, whatever racial discrimination may exist elsewhere in society, it is unacceptable to “reverse discriminate” against white applicants by using race-based affirmative action.
Now, this is precisely the sort of argument I had in mind when I said that Peter cannot consistently support class-based affirmative action. Simply replace “racial” with “socioeconomic,” and “white” with “rich,” and “race-based” with “class-based,” and you have an argument against class-based affirmative action. Again, Peter says in response, socioeconomic factors are actually relevant to determining whether someone has been disadvantaged. Except race is relevant too.
And even if Peter didn’t grant that, I wonder whether he would be willing to support race-based affirmative action if that were true? Given how strongly he seems to think it is inappropriate to compensate for discrimination in one area of society with discrimination in another, I have a hard time imagining how he could support race-based affirmative action under those circumstances. And therefore I have a hard time understanding why he supports class-based affirmative action under analogous circumstances.
There must be something about race as a classification, aside from its relevance or irrelevance in determining educational opportunity, that makes Peter uncomfortable. But there seems to be nothing about race that doesn’t likewise apply to class. At least nothing relevant. Yes, race is natural, and class is social. Why should that matter, though? (And race is social, anyway.)
If it seems like I’m reneging on my promise to support class-based affirmative action, I’m sorry for being unclear. My bottom line is this: I absolutely support class-based affirmative action using the complex socioeconomic metrics described above. But I do still object to the arguments Peter has made against race-based affirmative action. I think they rest on flawed sociological premises, like the one that says race is no longer a factor, separate from class, in affecting educational outcomes.
I also think they rest on flawed philosophical premises, like the idea that it is always wrong to fight discrimination with discrimination. On that last count, it seems to me that what matters is whether the end-results are such that we could have reasonably expected them to result from a discrimination-free society. I have no idea whether affirmative action programs produce such results, but I am pretty confident that a discrimination-free society would not produce the abysmally low minority college attendance rates that would have resulted from not using race-based affirmative action in the first place.