Sunday, May 06, 2018

Racial Bias in the US Justice System

Commenter Earl Wajenberg sent an excellent link to an older Slate Star Codex post that reviewed a great deal of the literature on whether the police and the courts are biased against black people, Race and Justice: Much More Than You Wanted to Know. As the AVI post he comments under, Chesterton, Paradox, Life Lessons From Sports has gone off the front page, from whence most of you are unlikely to ever go back and check it, I thought it best to bring it forward for its own post. Thank you, Earl.  Long article, but very much worth it.

Dr. Alexander is bending over backward to be fair and objective, as he always does. My own shorthand prior to reading his post was that while there is considerable difference in black vs white experiences in the court and sentencing parts of the criminal justice system, and there is more hassling by police over minor issues, there is no demonstrable police racial bias in dealing with violent crime. (And as it is those interactions that the National Anthem protests are ostensibly about, the case of the protestors would be weakened.)  I will have to step back from that.  The numbers point in a somewhat different direction: there is no clear overall bias in traffic stops, searches, and arrests, though there may be local bias. There does seem to be bias in sentencing, as low as 10% or as high as 20%, after all confounding factors are accounted for. That is less than popularly claimed, but still pretty significant. That is a very rough summary.  Alexander explains it better.

Interestingly, the difference in how the public is treated seems more exactly correlated to the neighborhood.  Police treat people in bad neighborhoods worse, whatever their race. One can see why the police would do this - part of that is expectation and hyperalertness, part a conscious decision to come down harder and nip things in the bud in bad neighborhoods.  Yet once can also see why this would be particularly disadvantageous to black people, who are more likely to live in bad neighborhoods.

After running through the data, Scott suggests that there may be some subtler explanations, based on a more panoramic view of US society and what it considers to be a crime and justice.  He doesn't suggest much under this category.  I have very little patience with such closings, which smack of "There may be a larger story which shows much of this is wrong, but I can't think of much."  Perhaps he knows that he needs to soften some of his conclusions in today's world or people will fry him. I have a few other quibbles, but the overall post is far better than anything I could do, so I'll let him make his own arguments.

I did like one of his lines in discussing the complexity of one part of the discussion: "Never trust the media to give you any number more complicated than today’s date."

No comments: