HPRgument Blog — June 23, 2010 4:43 pm

A better case for affirmative action

By Sam Barr

Liberals often try to defend affirmative action as fair compensation for historical injustice. To put their argument crudely and briefly, they say that whites got ahead unfairly for centuries, and now it’s time to help blacks get ahead.

Regardless of its philosophical merits or demerits, this argument is incredibly controversial. On its face, it allows an analogy to be drawn between historical discrimination against blacks and the proposed discrimination for blacks, and against whites, in the present. In other words, this argument is vulnerable to cries of “reverse discrimination” and “reverse racism,” and to calls for stopping discrimination altogether.

But the historical-injustice argument is not the only or the best argument for affirmative action. Liberals do better when they argue that affirmative action is justified as a remedy for current discrimination. They need to show how affirmative action is not the enemy of “merit,” but its protector.

For decades, conservatives have argued that affirmative action subverts merit by helping less qualified minorities and hurting more qualified whites, whether in college admissions, employment decisions, government contracts, etc. They have assumed the existence of impartial tests and procedures, e.g. the SAT or the written firefighting exam at issue in Ricci v. Destefano, whose results affirmative action ignores or shortchanges. When these tests show wide gaps in performance between the races, conservatives shrug their shoulders and say that’s too bad. After all, it’s not our fault blacks don’t do as well on these tests. Or, if it is, then the proper solution is better education (which, of course, means school vouchers) and better parenting. Whatever you do, though, you can’t just throw out the results of an impartial test of merit.

This argument is susceptible at its very roots. What if the tests aren’t impartial? What if they’re not predictive of future performance? What if they have a hidden bias that helps whites and hurts blacks?

That’s why the finding that the SAT “favors one ethnic group over another” is so potentially huge, if further studies verify it. As Daniel Luzer of the Washington Monthly summarizes:

[The study's] core findings suggest that disadvantaged black students do better on hard questions, which contain large words with unambiguous meanings. Students can learn these words in practicing for the SAT. Black students do worse, however, on easy questions, which are mostly made up of simple words. “Simpler words tended to have more meanings, and in some cases different meanings in white middle class neighborhoods than they had in underprivileged minority neighborhoods,” according to Matthews.

Now, this is just one study and it seems to be very controversial within the field of people who study such things. But it is one more reason for liberals not to accept the paradigm that makes them out to be the enemies of merit. Merit is a very difficult thing to measure, and conservatives have assumed for too long that merit is fully captured by historically contingent tests and procedures.

Maybe we’ll never have a definitive answer on whether the SAT is as fair as it could be. But it seems to me that, given what we know, you have to make one of two assumptions. Here’s how I put it last month, in response to the Stephanie Grace brouhaha:

In order to justify assessments of “merit” where blacks and whites perform differently from one another, you have to assume that they perform differently because they actually are different and immutably so: because blacks are dumber, or less cut-out to be firefighters, or what have you. In order to critique those assessments of “merit,” you have to assume that blacks and whites have basically equal capacities, and that differential outcomes on certain assessments are attributable to differences in cultural background, education, social class, etc.

So, this latest study on the SAT is just one datum in favor of the latter assumption. But the more important point is, look how revolting the former assumption is. Yet, what other assumption could underlie the right’s defense of the sanctity of test results, its insistence that merit is totally and exclusively measured by the likes of the SAT?

Affirmative action isn’t the privileging of “justice” or “compensation” over “merit.” It’s part of the search for merit, taking account of the fact that we live in a race-conscious society with a lot of historical baggage.

Related posts:

Obama Mad Libs
Biden and Gibbs Mad Libs
Social Studies Walkout: Video
Google Sells Out
  • Chris Danello

    Interesting points, Sam.

    The big question that the article didn’t answer, though, remains somewhat crucial distinction between race versus income-based action. If minorities are disproportionally likely to attend worse public schools, have few books at home, lack parents with the time to read to them, and all the rest; of course they are disproportionally more likely to miss the easy words. But this is an argument for judgments based on backgrounds, not race. It’s right and proper to realize that a 2200 score from an inner city public school may be better than a 2400 from a suburban private one. Yet the trouble with affirmative action programs as traditionally construed is that an upper middle class minority child scoring that 2400 gets the thumb on a scale whereas a white or Asian scoring that 2200 enjoys no such benefit. I suspect you and I might diverge on the extent to which class should be a factor, but to contend that race supersedes it, as the authors seem to do, remains a ludicrous argument.

  • Sam Barr

    Chris,

    I don’t see where the authors of the study suggest that race “supersedes” class. More likely, like most academics, they believe that race and class are inextricably linked and are both important factors in shaping individuals’ life chances.

    At first glance, that might seem like a good reason to favor what we might call “class-based affirmative action.” Surely such a program would disproportionately help minorities, since minorities are disproportionately lower class.

    But several problems jump to mind. First, this study identified a cultural and linguistic divide between African-American and white students. You might think that upper or middle-class African-Americans would be culturally and linguistically “white,” and for some of them that’s probably the case. But I bet a lot of this cultural and linguistic divide stems from de facto segregation — different races living in different places learning and speaking differently. And it’s not the case, though many assume it is, that being higher-class (making more money) buys you a ticket out of segregated neighborhoods. When you, like many conservatives, argue that less impoverished blacks don’t also deserve a leg up, you assume that class trumps race. But it doesn’t.

    Okay, but now what about those few blacks who are wealthy, live in integrated communities, go to great schools, etc? I agree that the case for helping such individuals is a little weaker, but it doesn’t become negligible. There’s evidence, for instance, that even where incomes are roughly at parity, black families have much less overall wealth than white families, and wealth as well as income is correlated with test performance and other life outcomes. There’s also evidence of bias against “black names” in employment decisions, a form of discrimination that will not entirely disappear with higher class status.

    Finally, I don’t think it’s consistent to allow class-based affirmative action but not race-based affirmative action. Or rather, I should say, the reasons that conservatives often give for opposing the latter would apply just as well against the former.

    I’ve written about this before (see http://brfootnote.theclawmagazine.com/2009/07/31/conservatives-and-class-based-affirmative-action/) …

    “After all, class-based affirmative action would punish some and lift up others based on personal qualities beyond their control, just like regular old affirmative action. It would lead away from “merit”-based evaluations, like SAT scores and written firefighting exams, which conservatives have tended to exalt as fair, valid, and total measures of worth. It would rely on making assumptions about individuals based on the group one assigns them to — assumptions like: “Bill, whose family makes $30,000 a year, must have overcome more hardship and thereby demonstrated more ‘merit’ than Fred, whose family makes $60,000 a year.” Such an assumption seems just as crude and potentially invalid as one in which the students are separated by race instead of class. Finally, mightn’t class-based affirmative action “congratulate its practitioners on their virtue, condescend to its beneficiaries, and corrode the [class] attitudes of its victims,” to borrow from Douthat?

    Again, I know that all Douthat and Taylor have to say is “But I really do believe in class-based affirmative action,” and we’ll have to take them at their word. But this isn’t just about them: I am skeptical that the conservative populace, which for thirty years has been buying claims like those sketched above, is suddenly going to embrace a policy that, at its core, accepts all of their opponents’ premises and violates all of their stated principles…. I don’t see how you decide that treating people as members of categories, rather than as individuals, is suddenly a good thing when the categories are class-based rather than racial. Those are different kinds of categories, of course, but are they different in a morally relevant way?”

    -Sam

  • Pingback: ACE Forum: Class-Based Affirmative Action « The Harvard Political Review

  • Pingback: Affirmative Action and the SAT « Penn Political Review

  • Pingback: ACE Forum: Class-Based Affirmative Action, Round 2 | Penn Political Review

  • Pingback: Alliance of Collegiate Editors » Affirmative Action and the SATAlliance of Collegiate Editors

  • Pingback: » ACE Forum: Class-Based Affirmative ActionAlliance of Collegiate Editors

  • Pingback: » ACE Forum: Class-Based Affirmative Action, Round 2 (Penn PR)Alliance of Collegiate Editors

custom writing