How UW research convinced our state’s highest court to toss out the death penalty

Professor Katherine Beckett uncovered a pattern of racial bias. The Washington Supreme Court couldn't ignore it.

Seven winters ago, Heather Evans, a graduate student in sociology, sat at her kitchen table stunned by what she saw on her computer screen. She had been running a statistical model of death sentences in Washington state, and the results were profound and troubling—a black defendant was at least four times as likely to be sentenced to death than a prisoner of any other race. She wondered if she had made a mistake.

“Honestly, I did not think we were going to find a race effect,” she says of the project she had been working on with her professor, Katherine Beckett. So Evans started switching out variables like mitigating factors, the race of the victim, the locations of the murders and the heinousness of the crimes. But little changed. Black defendants were still much more likely to receive a death sentence.

Evans called Beckett with the news. “Then she started asking about changing the variables and re-running the numbers,” says Evans, ’05, ’16. “I kept telling her, ‘Yes, I did that.’ And ‘I did that, too.’”

Evans and Beckett, a UW professor in the departments of Sociology and Law, Societies & Justice, undertook the study at the request of a legal team representing an African American man on death row in Washington. The team asked Beckett for a statistical analysis of Washington’s death penalty cases to see if race was a factor. She agreed, believing the project wouldn’t take more than 20 hours of her time. “But we quickly figured out that this would be much more complicated,” she says. She hadn’t realized it would take years.

Every state’s statute on the death penalty is different, so the UW researchers had to first decide where their study should focus and then figure out what data was available. Using trial report forms prepared by judges for all of the state’s capital cases since 1991, they decided to home in on two decision-making points—one where prosecutors determine whether the defendant should face the death penalty and one where the juries choose the death sentence.

“Our first report had two surprises,” Beckett says. When the researchers initially looked at the prosecutors’ decision-making, they found that the race of neither the defendant nor the victim had any influence on whether to pursue the death penalty. Media attention, however, did appear to be a factor when prosecutors were more likely to seek death, Beckett notes. The higher-profile the case, the more likely capital punishment would be sought.

The second surprise came with the jury decision. “That’s when I got the phone call from Heather,” Beckett says, “because the size of the effect [that a black defendant was at least four times as likely to get death] was so big.” The sociologists ran the models again and again, considering different data like the defendant’s prior convictions, the victim’s race, the number of victims and whether the victim was a female, child or stranger.

Ultimately, Beckett and Evans reviewed all potential death-penalty cases in Washington between 1981 and 2014—a total of 332. They relied on the judges’ post-trial reports as well as other case materials and legal records. “I was the data geek of the team,” says Evans, who completed her Ph.D. in 2016 and now lectures in sociology at the UW. “As we kept going back to run the numbers with different variables, at first I was concerned that it was going to delegitimize our results. In fact, it did the opposite.”

That race was a factor here in Washington didn’t completely surprise the researchers. “If you just do back-of-the-envelope calculations, you can see with the naked eye that a large proportion of black defendants were getting the death penalty more than others,” says Beckett. Among cases in which death notices were filed and special sentencing proceedings occurred, juries imposed death in 38.8 percent of the cases involving non-black defendants, but 64.3 percent of the cases involving black defendants. Many other studies from around the country show the significance of race in disparities in the American criminal justice system.

But the 45-page analysis that would eventually be known as the “Beckett Report” would not only deepen the understanding of the institutional racism in the judicial system, it would profoundly—and permanently—change the law in Washington.

***

In December 2012, Lila Silverstein, an appellate public attorney with six years in practice, was invited to join the biggest case of her career.

Working alongside criminal defense lawyer Neil Fox, she would represent Allen Eugene Gregory, a Tacoma man on death row since his conviction for the 1996 rape and murder of his neighbor, Geneine Harshfield. Gregory was 24 at the time of the crime. His criminal record was not imposing; it held just one drug possession conviction and a conviction for the theft of a skateboard when he was 13. But because of the brutality of the 1996 crime, Gregory was charged with aggravated murder and the prosecutors sought the death penalty, which the jury imposed.

Six years later, Gregory’s death sentence was reversed by the state Supreme Court for reasons of prosecutorial and judicial misconduct. Then, in 2012, the case was tried before a second jury. That jury reimposed the death penalty. So Gregory was onto his second appeal—one in which he and his attorneys could bring up legal issues from the more recent trial as well as raise new constitutional claims.

This was just the sort of work Silverstein had in mind when she enrolled at the UW Law School in 2004. She knew she wanted to have a career in civil rights or criminal defense. “Those were two ways you could protect marginalized communities and support individual rights,” she says. She joined the law school’s Innocence Project Northwest Clinic, which works to free innocent prisoners and remedy causes of wrongful convictions. Silverstein also took classes in constitutional law and appellate practice, knowing those would provide expertise she’d need. “I feel like UW Law really prepared me for a career in public service,” she says. “It made a difference right out of the gate.”

After graduating, Silverstein, ’06, was hired by the Washington Appellate Project, a nonprofit dedicated to providing legal counsel to indigent clients. Since then, she has argued more than 26 cases before the Washington Supreme Court. One involved race discrimination in jury selection while another addressed a client’s First Amendment right to criticize police.

For Gregory’s defense, Silverstein took the lead in commissioning a statistical study to look at whether race had a role in death penalty sentencing. Right away, she knew she wanted Beckett to take on the project. She had seen another of the UW sociologist’s studies, one that focused on racial disparity in the enforcement of drug laws in Seattle. “I knew she was respected in both the social science and legal communities,” Silverstein says.

The Gregory case came on the heels of another, failed, death penalty appeal from a black defendant. Though the appeal had failed, Silverstein pointed out a dissenting opinion where a judge noted that there seemed to be racial disparity in the application of the death penalty. “It was really a call to action in future cases,” says Silverstein. It was time to look at the numbers and see if race had a role. “We knew we had to take that step.”

***

On February 25, 2016, Silverstein stepped to the front of a crowded courtroom at the Temple of Justice in Olympia and argued that the numbers in the Beckett Report showed that the death penalty in Washington is imposed with racial bias. The justices peppered her with questions, but Silverstein stuck to her message. “In every single model analyzing jury decisions [the UW team] tested, race always mattered and it was always statistically significant,” she said. “Had Allen Gregory been white and everything else were exactly the same, his likelihood of receiving the death penalty would have been notably lower. …The system as a whole is unconstitutional because it is infected with arbitrariness and racial bias.”

Later that day a Pierce County prosecutor stood before the justices and questioned the value and integrity of the Beckett Report. She argued to uphold Gregory’s death sentence. Kathleen Proctor, a deputy prosecuting attorney for Pierce County, said that her office hadn’t had time to review the Beckett Report and that, since it was not brought up at Gregory’s criminal trial, the Supreme Court shouldn’t consider it at all. “We are not talking about some sort of independent scholarly research thing that was done,” Proctor said. “It was hoping to show this bias.”

But the judges didn’t agree. “We’ve got a piece of evidence here … statistics or evidence that is based on trial reports that are public records … a numbers-crunching exercise to arrive at certain conclusions,” said Judge Charles Wiggins. Several weeks after the oral arguments, the judges issued a ruling instructing the state to conduct its own review of the Beckett Report. Beckett, a widely published scholar who has been at the UW since 2000 and a full professor since 2009, wasn’t too worried. “We’re used to having our work evaluated,” she says. “Peer review is standard practice.”

But it turned out that this evaluation was far from a typical peer review. The prosecutors found their expert at UC Irvine. He was an assistant professor of psychology who had joint appointments in the departments of Psychology and Social Behavior as well as Criminology, Law and Society. The prosecution’s expert wrote that he couldn’t reproduce Beckett’s findings. Not only that, but he wrote that when he used the data that the UW team had used, he found that black defendants were not more likely to be sentenced to death. He even suggested that Beckett and Evans had manipulated their results.

Beckett was traveling with her family in Italy when she heard about the state expert’s review. “The tenor of his report was unlike anything I’ve ever seen in a peer review process,” Beckett says. “It took us aback.”

Beckett cut her trip short and flew home. She and Evans had just a few weeks to rebut the prosecution’s report. “We spent a lot of time at Heather’s kitchen table with three computers running,” Beckett says. “We were just trying to be so careful and so thorough.”

First, it took the UW pair some time to figure out how the other expert reached his results. But here’s where the “genius of Heather” served them, says Beckett. Evans started imagining what mistakes she might make with less experience and was able to reproduce his results.

Reviewing the state expert’s report, Beckett and Evans found two key concerns. First, because he failed to conduct necessary data transformations, he dropped 22 cases from his analysis. Beckett and Evans suspected these cases were ones in which the defendants had no prior convictions, no mitigating circumstances or both.

The prosecution’s expert also made an error in measuring for race. He only looked at cases involving white and black defendants and white and black victims. That meant he didn’t consider the cases in which the defendant or victim were neither white nor black. And he didn’t count the cases with multiple victims of multiple races.

The UW team had just 45 days to write its response. Even after updating the report to include data from the death penalty trial reports that had become available since they first ran the study, the results continued to show that black defendants were four times as likely to receive the death penalty. A court-appointed commissioner reviewed all of the reports and responses. Then she followed up with her own questions for Beckett and Evans. Then there were briefs filed by leading scholars from around the country supporting Beckett’s findings and arguing that the death penalty in Washington failed to comport with the state constitution. The process went on for many more months.

Ultimately, last Oct. 11, the Supreme Court revisited the constitutionality of the death penalty—and determined that it violated the state constitution’s prohibition on cruel punishment because of its disproportionate use for black defendants.

In the end, Silverstein and Fox were not only able to help their client, they, with the Beckett Report in hand, were able to convince the court to end the death penalty in Washington. Gregory and the seven other men on Washington’s death row had their sentences commuted to life without parole. The majority opinion pointed to the Beckett Report, crediting it for convincing them that “the association between race and the death penalty is not attributed to random chance.”

Washington became the 20th state to abolish the death penalty, and only the third—after Massachusetts and Connecticut—to do so based on racial disparities. In the first days after the ruling, the Chronicle of Higher Education wrote about Beckett’s role in changing state law and detailed some of the extended debate and attacks aimed at undermining her work. A piece in The Atlantic focused on the Washington Supreme Court’s finding that data should matter in judicial decision-making and, in ending the death penalty, the state court made a bold decision that the U.S. Supreme Court has avoided. The author described the ruling as a “kind of judicial declaration of independence.”

Despite the arduous experience, Beckett and Evans continue to take on projects where data analysis can be used in support of change. “I went to a community college and a state university,” says Evans. “This is a way I can stay connected and give something back to my community.” One of their current studies looks at the role race plays in long and life prison sentences in Washington state.

From Beckett’s standpoint, her work furthers the UW tradition of providing solid data from social science to inform public policy and practice. She points to faculty predecessors like Robert Crutchfield and George Bridges, ’72, whose work led to changing bail practices in King County. “I have examples in my department of people whose work had a real focus and impact,” she says.

While Silverstein continues to serve indigent defendants charged with serious crimes, she doesn’t expect to see another case like this one soon. “This was a really stressful, brutal project,” she says. “But it was worth the effort. Nothing exactly like this has happened in other states … there aren’t many state Supreme Court cases that you can point to like this one. It’s a great case, and it should inspire other attorneys and other courts around the country to take similar action.”

***

The UW has a tradition of its research affecting public policy. A few other examples:

Vaccination
The School of Pharmacy made Washington the first state to train pharmacists to provide flu shots and other vaccinations. Other states have followed suit.

Suicide prevention
The School of Social Work’s Forefront Suicide Prevention Center led Washington to become the first state in the nation to require suicide-prevention training for behavioral and health professionals.

Minimum wage
The Seattle Minimum Wage Study, produced out of the Evans School of Public Policy & Governance, is assessing how raising the city’s minimum wage to $15 is affecting employers, workers and the local economy.

Killer whales
The UW Center for Conservation Biology’s study on pregnancy failures in the orca population in southern Puget Sound linked the whales’ problems to a scarcity of salmon and nutritional stress.

Ocean traffic
Tugboats and commercial crab fishermen often clashed about the placement of crab pots because they could damage shipping traffic. The UW-based Washington Sea Grant now leads the effort to provide information on tugboat and barge lanes between Cape Flattery and San Francisco.

Vision testing
The routine eye test for school children, an eye chart that has been in use since the Civil War, can’t detect whether a child’s eyes work together, track properly or change focus easily. UW Bothell Professor William Erdly convened a symposium that led to a new law requiring that near-vision testing be included with the standard eye test.