Skip
Current Issue
This Month's Print Issue

Follow Fast Company

We’ll come to you.

6 minute read

World Changing Ideas

To Reduce Bias, What If Prosecutors Didn't Know The Race Of The People They Were Prosecuting?

Before they see a courtroom, prosecutors make enormous decisions about what crimes to charge and how high to set bail. Why do they need to know the race of defendants to do that?

To Reduce Bias, What If Prosecutors Didn't Know The Race Of The People They Were Prosecuting?

Photo: Siberia via Shutterstock

In the United States, there are two systems of justice—one for white people, and one for everyone else. When compared to African- and Hispanic-Americans committing the same crimes, white Americans are less likely to be stopped by the police or arrested in either the adult or juvenile detention systems.

Study after study confirms that bias permeates nearly every level of criminal justice—from arrests to convictions to sentencing—and that includes one group who holds a disproportionate amount of power when it comes to determining the specifics of criminal proceedings. Prosecutors have an absolutely huge amount of discretion in criminal cases, including power to determine whether to charge a defendant with a crime, the length of prison terms, and the specifics of plea deals. That last one is especially crucial considering that the vast majority of defendants—more than 90%—never go to trial, meaning that it’s often up to a single prosecutor to decide how justice plays out.

"Depending on who you get—literally which prosecutor you get—you could serve two years in jail or no time at all over a small crime," says Shima Baughman, a professor at the University of Utah College of Law. "I don’t think people really understand how this system works."

Flickr user Tim Evanson

Just like in every other area of the criminal justice system, bias rears its ugly head in plea bargaining deals and other prosecutorial decisions. White Americans, for example, are less likely to spend time in jail awaiting trial. By contrast, black Americans are more likely to have their probation revoked and receive plea deals that include prison time, while whites receive, on average, prison sentences that are about 10% shorter than those of racial minorities facing the same offenses.

Measuring how biased prosecutor offices are is tough since offices aren't required to collect or report that kind of data. A paper published in the journal Behavioral Science and Policy proposes an alternative solution—when possible, take race out of the equation entirely. Co-authored by Baughman and University of Arizona law professor Christopher Robertson, the paper proposes using automated electronic systems, paralegals, or other intermediaries to remove references to the race of the defendant prior to the prosecutor reviewing the case file. In a phone interview, Baughman added that the race of the victim(s) could also be removed as well as information on which neighborhood the crime occurred in. While prosecutors may figure out a defendant’s race at some point in the legal process, blinding during the early stages could reduce bias in both decisions about what crimes to charge them with and what kind of plea bargain deals they're offered, says Sunita Sah, a behavioral ethics researcher at Cornell University and the paper’s lead author.

"Just a slight reduction in bias could have huge effects, not just on people in prison, but also an effect on families and society in general," Sah says, adding that reducing bias could also improve the perceived legitimacy of the legal system. Sah’s paper is just a proposal and did not measure current bias levels or evaluate the impact blinding prosecutors might have. "We need to pilot it and really see how this would work in the field," she says.

Blinding is already used in some areas of the criminal justice system, like eyewitness identifications. To prevent arresting officers from providing nonverbal cues that may guide an eyewitness to a certain suspect, lineups are frequently conducted by a law enforcement agent who has no knowledge of the case. In 2014, the National Academy of Sciences recommended that agencies institute double-blind lineups wherein neither the lineup administrator nor the witness knows which photo is the real suspect, or if the suspect is among the array of photos at all. Many agencies have adopted the practice, though it’s not legally required. To reduce bias in federal capital cases (which are the minority as most capital cases are overseen by states), the Department of Justice requires the U.S. Attorney’s Office to remove information about the defendant’s race and ethnicity before submitting the case to the review committee that determines whether to seek the death penalty.

Flickr user Tom Woodward

Even with some precedent in place, Sah says that implementing racial blinding will be tough, and race will need to be considered in certain circumstances, like hate crime cases.

"There will for sure be challenges in implementing this. The first one is overcoming people’s reactions to whether they need it or not," Sah says. "Other people will acknowledge that unconscious bias could have an effect and [blinding] sounds good in theory, but practically it’s going to be difficult to implement."

Ezekiel Edwards falls into that latter group. Director of the ACLU’s Criminal Law Reform Project—an initiative dedicated to reducing racial disparity in the justice system and challenging unconstitutional policies and practices—Edwards says that blinding prosecutors may work in some instances when charging a defendant with a crime but it would be very difficult to implement at other spots in the legal process.

"If you are a prosecutor gathering evidence and conducting eyewitness ID procedures or statements or trying to take confessions, you will know the race of the person," Edwards says. "At numerous other points along the line where decisions are made about what bail to ask for, whether or not to indict somebody, whether or not to add charges, how to strategize about what to do at trial, I think it’d be almost impossible to do this."

Edwards agrees that further steps need to be taken to reduce bias, but believes that changing prosecutors themselves, through implicit bias training and improving diversity, are more practical options. The latter has become a major talking point in justice reform circles. While juries are supposed to represent a diverse segment of the population, prosecutors are overwhelmingly white. Research published last year by the Women Donors Network’s Reflective Democracy Campaign found that 95% of elected prosecutors are white and 14 states don’t have any elected prosecutors of color. That doesn’t cover the entire profession (a very small handful of states fill district attorney slots by appointment), but it does cover a large chunk.

The United States is the only country in the world where prosecutors are elected, and the reasons for their racial uniformity are myriad and complicated. Socioeconomic barriers to entering the profession, the lack of diversity in top law schools, voter bias, and the fact that incumbent prosecutors rarely turn over (85% run unopposed) all factor in. The lack of diversity doesn’t just impact individual defendants, as researchers from the Stanford Criminal Justice Center noted in a study of California’s prosecutor offices. Those holding prosecutor positions often use those jobs as springboards to other elected offices, judicial appointments, or higher positions in the legal field. In other words, if prosecutor offices don’t have people of color, there’s a higher likelihood that positions with even more political or judiciary power won’t either.

On top of improving diversity, Edwards says that the legal system should also make conscious efforts to increase cultural competency and regular meaningful interactions between prosecutors and the communities they serve. Step one, he says, is to encourage prosecutor offices to undergo assessments designed to evaluate possible bias in their practices.

"I frankly think there are probably prosecutor offices just like police departments that don’t really want to know what the data’s going to reveal, and they don’t really want the public to know," Edwards says. "If they think about it long enough, they’re probably worried that it’s not going to look great."

loading