One of the things people don't always get about "criminal justice reform" is that what actually needs to be reformed, and how that reform looks, differs from state to state and county to county. Pretty much every state in America has contributed to the country's mass incarceration problem, but they've gotten there by many different policy routes. And while there are some defining themes — like racial disparities between black and white offenders — there are plenty of different points in the process where those disparities might come out.
If you want to learn what it looks like on the ground when an official wants to fix criminal justice in his jurisdiction, it's worth reading this New Yorker article from May in which Jeffrey Toobin profiles Milwaukee County prosecutor John Chisholm. In particular, the article gets at one of the underappreciated paradoxes of the criminal justice system: that it devalues both African-American offenders and African-American victims of crime.
This is what researchers from the Vera Institute of Justice found was happening in Milwaukee County shortly after Chisholm took over:
Prosecutors in Milwaukee declined to prosecute forty-one per cent of whites arrested for possession of drug paraphernalia, compared with twenty-seven per cent of blacks; in cases involving prostitution, black female defendants were likelier to be charged than white defendants; in cases that involved resisting or obstructing an officer, most of the defendants charged were black (seventy-seven per cent), male (seventy-nine per cent), and already in custody (eighty per cent of blacks versus sixty-six per cent of whites).
The racial disparities persisted across the board — with one big exception:
"When I first saw the data, I thought, Here is some good news," Chisholm told me. "It said that we charge white offenders for property crimes at a higher rate than we do black offenders for those kinds of cases. So I thought, Good, here is a disparity the other way. That must balance things out. But a deputy of mine pointed out that what the data really meant was that we devalue property crimes in the center city. We don’t charge a car theft, because we think it’s just some junker car that’s broken down anyway. It meant that we were devaluing our African-American victims of property crimes—so that was another thing to address."
It's an excellent illustration of a phenomenon we've discussed at Vox: that African-American communities are both overpoliced and underpoliced. But it's also a challenge to anyone who wants to reform criminal justice to reduce racial disparities. If police and prosecutors are told to be less aggressive in going after low-level offenders, how can they make sure they don't end up devaluing African-American victims? If they're told to take property crime in black communities as seriously as property crime in white communities, how can they avoid overreacting by rounding up and hauling in too many young men of color for things white men would get away with?
This isn't to say it's impossible to find a balance between overpolicing and underpolicing. But law enforcement experts are only just beginning to recognize how powerful implicit racial bias is, and it's going to take a while to figure out how to counteract it.
It's also a recognition of the simple reality that black communities are victimized both by crime and by incarceration. Later in Toobin's article, the Milwaukee police chief puts it succinctly: "Everything we do is going to have a disparate impact on communities of color."