How Racial Bias Influenced Stanford Swimmer's Rape Case

"A prison sentence would have a severe impact on him. I think he will not be a danger to others," Judge Aaron Persky said in his decision.

The ruling has spurred a recall effort against Persky that now has over 200,000 backers, and commenters on Twitter and bloggers have pointed to Turner's light sentencing, which he intends to appeal, as a sign of our criminal justice system's inherent bias in favor of the white and privileged. If Turner were black, they say, he would be under the jail. As witnessed by the "affluenza teen" who received a 720-day jail sentence for killing four people while driving drunk, this certainly seems true. However, statistical attempts at teasing out racial bias in courts have been surprisingly non-definitive.

In his 2012 study, "Do Judges Vary in Their Treatment of Race?," David Abrams, a professor at the University of Pennsylvania Law School, points this out. "The basic problem that plagues almost every study that tries to look at the role of race in criminal sentencing is that you can't randomly assign race, because that's what you do when you want to do an experiment," Abrams told me. "The way most studies go about it is to try to run a regression on control variables that they think might impact the [sentencing] outcome besides race. That can help, but it can't eliminate the basic problem." In other words, it's been impossible to tell how race affects sentencing because it's impossible to isolate it from the other factors that could be at play in the courtroom, like "how you dress, how you speak, and other things that might affect the sentence length that aren't necessarily about race but are probably correlated with it."

Nazgol Ghandnoosh, a research analyst with the Sentencing Project, also mentions that when comparing the race of defendants and their sentences, it's important to note the charges. "There's a whole variety of crimes that would fall under sexual assault and rape, so it's important to compare [Turner's case] to something similar," Ghandnoosh says. "Some of the other kinds of crime that I've seen people compare it to are just of a different nature." For example, ex-football player Brian Banks, who served five years in prison for a rape he did not commit, has been held up as a foil to Turner. Banks's case is appallingly tragic: He was tried as an adult when he was a minor and similarly had no prior record. But the difference in their sentencing outcomes comes down to their charges and not just bias. Banks was charged with forcible rape and kidnapping, while Turner wasn't even charged for rape. The prosecution originally tried to go after the 20-year-old for two counts of rape, but those charges were dropped, baffling everyone who has read the testimony from the victim—who was found behind a dumpster without underwear—and the police reports.

The differing charges brought against the two defendants—one white, one black—is an injustice itself, but it's one that's important to distinguish. Ghandnoosh explains that this illuminates how racial bias pervades every stage of the criminal justice system. "Prosecutors are likely to charge people more harshly if they're black than if they're white," she says. "There are racial disparities at each stage of the process. It snowballs as someone goes through the system." This starts with the likelihood of being arrested. Then there's the ability to post bail pre-trial, which research from the Sentencing Project shows leads to better trial outcomes, and to hire a defender. "Then there's the jury issue," she says. "In this case the jury did give a conviction, but there's evidence that black defendants are more likely to be convicted than white defendants."

There are numbers that are certainly indisputable. Abrams notes in his study that though 13.2 percent of the population is black, in 2008, 38 percent of sentenced inmates in the United States were black. That same year, black males were incarcerated at six-and-a-half times the rate of white males. So to figure out if biased judging actually played a role in this, Abrams came up with a novel workaround. "What we did is change the question a little bit. Rather than directly trying to answer how biased, or not biased, judges are, we looked at the variability across judges in their sentences between black and white defendants, which we call the racial gap," he says.

In doing this Abrams and his team of researchers took advantage of a variable that is random: the cases that judges are assigned. Since each judge will get a fairly even mix of types of crimes and racial defendants, "on average, every judge should be doing the same thing, unless there is bias involved." Using data from Cook County in Illinois, he looked at how individual judges rule on felony cases when the defendant is black and when the defendant is white. If there was no personal bias involved, then each judge would have the same racial gap, but that's not what Abrams found. The gap in incarceration rates between white and black defendants increases by 18 percentage points when moving from the 10th to 90th percentile judge in the racial gap distribution, meaning that some judges were racist as hell.

However, that's just for incarceration. Abrams was surprised that he was not able to find any statistically meaningful racial difference between blacks and whites in terms of sentence length, though he suspects it has more to do with his sample than the fact that racial bias in sentencing doesn't exist. "This might be because there's some sentencing guidelines in Illinois that constrains judges' leeway, but there's less that constrains whether they send someone to prison or not." When I asked him if he would expect to see a different result in a state without those restrictions, he said, "Yeah. If the underlying distribution bias in judges is the same, we would definitely expect to find more of an effect on sentencing."

Ghandnoosh says that while sentencing is perhaps the least statistically noticeable place to find racial bias in the courts, due to mandatory minimums on certain crimes, "judges are more likely to see a white defendant as a candidate for rehabilitation than they are a defendant of color." Clearly, racial bias is a real, and angering, part of our criminal justice system.

But there are also real interventions that can eliminate it if implemented, Ghandnoosh says optimistically. "One way is to bring everyone in the same, regardless of what resources they can bring to the table. In 2014, New Jersey passed legislation to reduce the reliance on bonds and instead use risk-assessment to determine pre-trial retention [in jail]," she says. "And when it comes to the bias, some jurisdictions have started to look very systematically about how the race of the defender impacts decisions and hold the courts accountable."

She also adds that it's perhaps time to think about our criminal justice system in entirely different terms. "When we think about equalizing outcomes, do we want to get everybody to the point where they're being sentenced too severely, or do we want to think about the just sentence? Making sure that everyone gets the worst outcome is not going to necessarily address the problems we have with over-incarceration." This is, of course, great in theory, when you think about the sheer numbers of people in jail serving mandatory minimums for non-violent crimes. At the same time, perhaps bratty frat bros who think rape boils down to the follies of campus drinking culture deserve to have a harsh wake-up call.