Daniel Larison Is Making Sense

I’m hard-pressed to find anything I disagree with in his latest post on Sonia Sotomayor. The key paragraph:

[T]he appeals court ruled against Ricci because it recognized that New Haven had tried to avoid a lawsuit that would have been possible and likely successful because of current law. In other words, the city tried to avoid falling afoul of the law, and the court did not penalize it for doing so. What is to blame in all of this is the law, rather than the judges who seem to have done what they were supposed to do. Indeed, what some people seem to have wanted to see Sotomayor do is to punish New Haven for trying to stay within the limits of the law, and for failing to do so she is declared to be an enemy of the rule of law. I submit that this doesn’t make a lot of sense.

If I recall correctly, the “disparate impact” standard under which New Haven feared being sued was originally created by the Court in decisions of the early 1970s, as a test to determine whether illegal discrimination had occurred under the Title VII of the 1964 Civil Rights Act. But this standard was written into statutory law – and, indeed, toughened – by the 1991 Civil Rights Act. The court in Ricci had no need to revisit the correctness of the standard, but if some future court chose to do so, it would not be revisiting precedent but seeking to invalidate an act of Congress – moreover, one clearly and plainly intended to enforce an amendment to the Constitution, as that amendment provides for. Such an act by the court might be right or wrong, but it’s practically the definition of judicial activism.

And, if I recall correctly, Jeffrey Rosen was extremely critical of Justice O’Connor’s jurisprudence regarding redistricting because it put the states in a judicially-created bind. If they ignored the need for minority representation, they were potentially in violation of the Voting Rights Act. But if they designed districts that were transparently created to lump African-American voters together regardless of geography or other factors, so as to ensure explicit representation for that segment of the population, O’Connor would reject that as also a violation. Race could – indeed, had to be – a factor, but couldn’t be the exclusive or, indeed, the predominant factor. Which meant, in practice, that a redistricting that was legal was one that passed Justice O’Connor’s “smell test,” which sounds not much like the rule of law, much more like the rule of (wo)man. If the Court overturns and finds for the plaintiff in Ricci, and does so without explicitly declaring chunks of the Civil Rights Acts of 1991 and 1964 to be unconstitutional (which I find highly unlikely), wouldn’t it be putting employers into an analogous bind? If they don’t pay attention to the potentially discriminatory impact of their hiring practices, they’ll face litigation – but if they pay too much attention, they’ll also face litigation. And they’ll only be able to tell the difference between too little and too much by consulting Justice Kennedy. Right? It sounds an awful lot like Rosen wants the Court to do with affirmative action precisely what he said they ought not have done when it came to redistricting.

I want to be clear about two things.

First, I think the disparate impact standard is an honest attempt to address a real problem: the desire to ferret out racial and other forms of discrimination that do not boldly announce themselves to the world as such. I find it very hard to take claims that such discrimination does not exist; the better debate would be about the relative priority of rooting it out versus other policy goals, including other policy goals related to social justice.

Second, because of the persistent measured gaps in average general intelligence between different groups (that they are persistent is a simple fact; whether they are inherent is still a much more speculative question), the disparate impact standard makes it extremely difficult to hire based explicitly on relative measured intelligence. To the extent that we believe that selecting on that basis is important (and we do seem to believe that when we choose who gets to serve in the armed forces, who gets into what college, indeed, pretty much every time we have to sort large numbers of people anonymously for positions that do not require specific skills), the disparate impact standard creates a real problem.

But it’s a policy problem. Weighing conflicting claims to justice and coming up with workable rules for adjudicating them that can be the basis of social consensus is what the political process is supposed to do. The courts are then supposed to apply those rules in specific cases in a way that provides reasonable predictability and is understood to be fair and just. If, as a society, we’re not happy with the way the rules work now, we should talk about how to change them and find a new set of rules that “work” better. To do that, we need to be able to say more, not less, so policing our language is counter-productive. By the same token, to do that we need to be able to hear more, not less – which means that liberal claims about Sotomayor’s background being a net-positive actually do have some teeth to them.