Showing posts with label Walter Dellinger. Show all posts
Showing posts with label Walter Dellinger. Show all posts

Thursday, July 02, 2009

New Haven's Firefighters: Race, White firefighters, and civil rights, Part the Nth

A really fantastic piece, or seemingly from my quick readthrough, by Richard Thompson Ford again, Stanford scholar and previously linked-to commentator on the "New Haven Firefighter Case" of Ricci v. DeStefano.

Ford analyzes how the SCOTUS' finding for the white firefighters (and one Hispanic firefighter) in this promotion case overturns years of civil rights precedent, while going with more recent precedents of treating any attempt at addressing civil rights as creating their own racial (or sexual, in the case of homosexuality) preferences, "reverse discrimination" or "special privileges."

There are two interesting points here (well, many more than that, but two I will make right now.) One is that recent discussions with various J-friends have, unhappily, split pretty clearly along racial lines (though not with any racial overtext, that is, whether or not people agreed or disagreed with the SCOTUS decision fell quite racially within my very small subsample of my friends). BUT, the point of contention seemed to (mainly) fall into "Ricci et al. deserved to win because they played by the rules," that is, the test happened, and taking back what they'd "earned" by going along with the test as written would not be fair, regardless of the errors, biases, whatever of the test. (The "compromise" position found between J-fave D and J was that they were deserving of some recompense for the efforts they put into preparing for the test, even if they were not owed a promotion.) Essentially, as I discussed with J-Mom, it seemed to be a focus on the individual, individualism, and the rights of the individual not to be "harmed" for the good of the whole, or good of another group at least. (This is not at all how the J-friends phrased it, but my interpretation of their objections to my points of view). J-friend Sean felt that institutional racism was real, but of a lessor magnitude than perhaps I may think it was, a lessor magnitude than discrimination from poverty (which is hard, if not impossible, to parse, but suffice it to say I think institutional racism is a profound and not terribly diminished problem, something D seems to agree with to some degree.) Indeed, Sean argued (if I am restating it correctly) that essentially positive preferences for any group are essentially the same, such that discrimination "for" blacks was as bad as discrimination against them (and you can replace "black" with white or Hispanic or what have you here).

Thompson argues that this places many, many programs to address racial programs at risk, as in the end, they all hinge on treating one group differently than another or end up displacing some people who would otherwise have received a job, promotion, or college acceptance. (He uses the example of the 10% programs, where, for example, Texas admits the top 10% of students in their high school class, a supposedly "race-neutral" way of achieving the end of diversity and representation of minorities. As has been pointed out by others, this only works because Texas schools are segregated, thus, you get diverse representation because the top 10% of the class at predominantly black or Latino schools is itself predominantly black, so minorities get represented essentially because they are not integrated (or have lower relative average scores where they are more integrated). As Thompson says:
The university now admits any student in the top 10 percent of his or her public high-school class, and because so many of the public schools in Texas are racially segregated, this guarantees a racially diverse student body. Opponents of race-conscious affirmative action have pointed to this policy as an example of a viable, race-neutral alternative. But no one denies that the motivation for dropping the traditional admissions criteria in favor of the 10 percent plan is to achieve a better racial mix. Extending the logic of Ricci, this looks like impermissible race discrimination against the students who would have been admitted under the old criteria, just as dropping the firefighter promotion exam was impermissible race discrimination against the white firefighters who would have been promoted.


I think this poses a challenge to those who agree with the Ricci decision, unless they hold the very narrow stance that it is only because the test already happened that it was unfair discrimination. As Thompson points out, how would the case be (legally) different if, before the test, the city chose to use one test that they knew from previous data would tend to favor a more diverse mix (i.e. less whites and more minorities) rather than a test that would favor more whites. The intent there is almost exactly the same as throwing out the results for, as some of my friends posed it, "having too many white people". So if Ricci is unfairly discriminating against whites, using a test that you know would admit more minorities also should be, to be logically consistent.

The second point, which you may have forgotten I even supposedly had at this point, is that Ford reiterates that Ricci is a change in the status quo. As J-Mom AND the article by Bazelon I've prattled about here for the past week pointed out, the previous standard for a "discriminatory" test was just that the results were discriminatory. Governments typically used an "80% rule":
Title VII requires employers not just to inspect their hearts and not find any discriminatory intent, but to consider the racial impact of things like tests. And the EEOC, in interpreting this requirement, has given clear guidance about what impact counts as suspect:

"A selection rate for any race, sex, or ethnic group which is less than four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact, while a greater than four-fifths rate will generally not be regarded by Federal enforcement agencies as evidence of adverse impact."

The rates at which blacks and Hispanics passed the New Haven tests were well below 80% of the rate at which whites passed. That means that those tests were presumptively in violation of the law. (from Obsidian Wings)


Thus, the Court set aside the typical way things were done, where (as I have pointed out to J-friends), a test that has that much racial disparity is presumptively discriminatory, which seems to me a fine first pass at such things in a world of institutional discrimination, which I see as a very large on-going problem, contra J-friend EssEee.

As to the earlier point over the "fairness" to the individual, two other interesting bits:

Obsidian Wings: I have read many professions of outrage about this decision, but most of them focus on whether it is a good thing or a bad thing that Frank Ricci didn't get his promotion, rather than what the law requires. This puzzled me.

And Walter Dellinger:
Given that no one had been promoted and no one had been denied promotion, it's very hard to see how the firefighters who brought suit were able to establish the very first element of a Title VII action: the existence of an "adverse employment action." In addition to satisfying the statute, it would have been far better for the process to judge New Haven actions after promotion decisions were actually made using whatever new standards the city chose to adopt. Completing the process would have shed light on the question of whether there were in fact equally good (or perhaps, better) criteria for determining promotions, and with far less racial disproportion... Which takes us to the fact that Justice Kennedy's opinion relies in part on a logically flawed, categorical error. He writes: "If an employer cannot rescore a test based on the candidates' race [citing the Title VII provision], then it follows a fortiori that it may not take the greater step of discarding the test altogether to achieve a more desirable racial distribution of promotion-eligible candidates..." This is wrong. There is a very powerful difference between setting aside the results of a test based on what you learn from general racial statistics about those who took the test, on the one hand, and adjusting individual test scores on the basis of race, on the other... Using race to identify a problem has never before been considered problematic. It is what necessarily happens before institutions adopt the most widely accepted race-neutral actions, like using admissions criteria for every applicant that have less of a racial impact (for example, accepting students in the top 10 percent of their high-school class, which, in states like Texas, would produce a racially diverse student body). Contrary to Kennedy's assertion, deciding not to use test results should be far less problematic than "rescoring based on race." ...Here, all New Haven did was set aside the results of a test. It seems to me that test would have been very hard to defend, given the other questionable employment rules that surrounded it. New Haven counts the multiple-choice test as 60 percent of what determines promotion. That places twice the weight on test-taking as the median for firefighter promotions around the country. How can that unusually great a weight be justified? Kennedy says only that it was insisted upon by the union. Yep. That was two decades ago, when the union was dominated by the white firefighters. That so great a weight should necessarily be given to a multiple-choice test used to pick leaders out of a group of qualified firefighters is hard to see.


Going back to Ford to wrap up, he asks:
And why stop there? Even recruitment efforts aimed at underrepresented minorities are designed to increase the representation of those groups in work forces and entering classes with a limited number of openings. If these outreach efforts are successful, some minorities will necessarily displace some whites who would otherwise have been hired or admitted. Are those efforts discriminatory, too?


Among other things, I see this case as reflecting a problem of where many want us to be, especially in white America, and where we actually are (or at least how we perceive it in black America, to speak in broad generalities). If you think racism is a significant and real, persistent still-present problem, interventions of a certain size, from affirmative action to throwing out the New Haven test (which depending on how you look at it, was unfair to those who would have been promoted based on how the test was set up before hand by taking away what they'd earned, or would have been permissible because it threw out results based on the heretofore legally correct presumption that such skewed results were de facto racist) seem reasonable. If you think, know, or wish that racism has receded to the point where a lot less aggressive, race-blind, and minor intervention is all that's needed, it seems understandable to take it from Ricci's side.

Argh. Ok, I have a headache now (unrelated to the post, I think). Até...

Thursday, June 28, 2007

SCOTUS Plays "Hide-and-Go-Fuck-Yourself"

Sigh. So the Supremes have handed down their decision on the de/segregation case, in Kentucky and in Oregon, finding in essence that race can't be used in methods to address racism. This is rather like the Ari Fleischerian formulation of hoping your own hopes aren't dashed. Good commentary/analysis here and here, which big picture says Justice Kennedy's opinion, being the narrowest, (and, frankly, the closest to a "swing") is the "controlling" -- i.e. precedent-setting -- opinion. His opinion has the excellent formulation the race can be used when addressing past discrimination, but really, that's kind of icky and breaks too many eggs to make the integration omelet. He goes on to say it's ok to use race as a last resort, and if you show there aren't other means, such as zoning school districts etc. to make them racially diverse. (Great pre-commentary from Walter Dellinger on the legacy of Brown v. Topeka and the folly of banning race use in addressing racial discrimination here.) In other words -- you can use proxies for race because then we can continue to pretend that race doesn't matter. Fucking A.

For reference: The two cases were in school districts where they wanted to encourage a breaking up of the pre-existing housing segregation, which of course has been in no small part a result of discriminatory practices dating back to the GI Bill and white flight, and before than even, Jim Crow and forced and legal segregation. Their goal was to make sure every school had some portion of white & minority students, and not a ridiculous skew that didn't reflect overall city demographics one way or the other. Actually, Dellinger (a Duke University Law Professor) summed it up this way:
The idea that the principle of Brown condemns the valiant efforts of, say, the Louisville community to maintain schools attended by both black and white students seems profoundly wrong to me. The Louisville school system (I keep using Louisville, because I know that case better) takes account of the race of students to keep each school integrated. They don't try to replicate the one-third-black percentage of the district as a whole in each school, but they do take race into account where that figure would otherwise fall below 15 percent or above 50 percent. Good people, black and white, in Louisville have refused to give up on the public schools. They know that sharp imbalances in the race of a school population leads to "white flight' from the schools and that using race to keep schools integrated is essential to the viability of public schools.

Of course, they could have a system in which each school almost always reflects the racial proportions of the district's overall population simply by assigning all students by lottery. Each school would usually wind up about one-third black. No system would do that because assignment by lottery would impose enormous costs on families in transportation and deprive them of the great advantages of neighborhood schools. But, as I argued in my last posting, using neighborhood as the sole mechanism for school assignment means that the schools will replicate the housing segregation that defines Louisville as it does many of our metropolitan areas. The carefully planned Louisville system combines neighborhood schools with parental choice and some use of race to ensure an integrated experience and a viable public school system.
So. Yes.

Dellinger & Slate's Dahlia Lithwick have been conducting a dialogue on the SCOTUS' recent decisions (linked to throughout my post). They've added a conservative member to bring a new view, whose response is titled: Don't Despair -- There's a Better Way to Achieve Integration. I'm deathly afraid of reading this. Though, there's some slight possibility he's going to suggest higher education for all, universal improvement in housing and human services, cleaning up the pollution that disproportionately affects minorities and poor whites, and general redistributional justice. And perhaps he'll also bring me a dollar with compound interest for my wisdom teeth from 5 years back or so, and take us all to a magical gumdrop land full of chocolate rivers and strawberry fields.