What counts as discrimination? In our post-Ferguson era of heightened awareness of disparate racial treatment by police, no domestic question is more pressing. The U.S. Court of Appeals for the 5th Circuit has just addressed it in the context of a Texas voter identification law-- and gave a measured, socially astute answer.
The court held that there wasn’t sufficient evidence to say that the law requiring all voters to show an acceptable ID was intentionally racist. But it said the effect of the law nonetheless was racially discriminatory -- because it made it disproportionately harder for blacks and Latinos to vote than for whites. There’s a lesson here for the rest of us: In America today, there’s still racism that needs to be rectified, but it isn’t always intentional. Sometimes focusing on the effect rather than the intent can make anti-discrimination efforts more socially effective.
The appeals court began by acknowledging that the stated public purpose of the law, preventing voter fraud, was legitimate. Then it went on to consider whether the Texas Legislature that passed the law had additional, impermissible and discriminatory motives.
The court introduced its analysis this way: “We recognize the charged nature of accusations of racism, particularly against a legislative body, but we also recognize the sad truth that racism continues to exist in our modern American society despite years of laws designed to eradicate it.”
This was more than mere rhetoric, or rather it was rhetoric that actually matters. The court was hinting that it intended to find a middle way, which would seek to fix racism without the “charged” elements of political backlash that occur when racism is described as intentional.
To reach a conclusion of discriminatory intent, the district court judge who first heard the case relied on the deep history of racism and racist laws in Texas. Logically speaking, this reliance makes sense. If you’re trying to do the hard work of determining whether a facially neutral law, like the voter ID law, covers hidden racism, it should be relevant that the predecessors of those who passed it were profoundly racist.
But good history doesn’t always make for good politics. The court of appeals went out of its way to say that “we respect and appreciate the district court’s efforts to address this difficult inquiry.” Yet it concluded that the history of racism in Texas shouldn’t be used to ascertain the motives of the current Texas Legislature.
In so doing, the appeals court cited the Supreme Court’s opinion in the landmark 2013 Voting Rights Act case, Shelby County v. Holder. There the high court said that “things have changed” since the Voting Rights Act was signed 50 years ago, and struck down Section 4 of the law (not at issue in the voter ID case) because Congress had taken account of those historic changes. The appeals court was taking only partial guidance from the Supreme Court. It was echoing the Supreme Court’s insistence on historic change -- but as the prelude to a different result.
That result came in the court’s analysis of the separate question of whether the Texas voter ID law had a racially discriminatory effect. The centerpiece of its conclusion was a regression analysis performed by an expert in the case, my colleague Stephen Ansolabehere of Harvard University’s department of government. Ansolabehere’s study showed that Latinos were 195 percent more likely than whites not to possess ID that would be allowed under the law, and that blacks were 305 percent more likely than whites not to have the relevant ID.
The appeals court was wise to rely heavily on the study -- not because statistical studies are perfect or perfectly reliable, but because they have the socially valuable effect of showing discrimination without pointing an accusatory finger at any one actor. The court was taking advantage of the powerful numbers to reach a conclusion that would invalidate the Texas law without telling the Texas Legislature that its members are racist.
Why is that a good thing? From the standpoint of discrimination’s victims, it may seem sorely inadequate to say that discrimination exists but that no one person or entity can be blamed for it. And it may seem perverse to suggest that it’s better to get rid of racial discrimination by pretending it’s floating abstractly in the air than by specifically identifying the culprits and holding them accountable.
But when it comes to producing systematic racial justice in place of systematic racial injustice, pragmatic results weigh more heavily than the moral value of blame. The civil-rights movement is at the stage of getting people to renounce racism. It’s getting institutions to recognize their unacknowledged, systematic biases. Most, maybe almost all, of the white police officers who’ve been involved in killing or otherwise demeaning blacks in the conflicts that have gripped our attention this past year were free of conscious racism: That is, if asked about their attitudes toward black people, they almost certainly would have said in good conscience that had no discriminatory beliefs.
Yet the actions of those police were discriminatory nevertheless -- because of unconscious discrimination that inheres in police practices, customs and institutional arrangements.
The way to fix this systematic, unconscious racism is by identifying its effects -- and then targeting their causes. For this task, statistics matter, and blame isn’t particularly useful. The 5th Circuit was trying to show us the way.