Jeff Sessions Is Keeping Junk Science in America's Courts

Trump's attorney general is a threat to the rule of law he's tasked with upholding

Attorney General Jeff Sessions announced earlier this week that he’s suspending the National Commission on Forensic Science. Credit: Jabin Botsford/The Washington Post/Getty

Much of the "forensic science" used to convict people of crimes in the United States turns out not to be science at all. After a number of scandals showed forensic techniques developed by prosecution experts to be either flawed or completely bogus, the Obama administration took steps toward comprehensive reforms to address the crisis of junk evidence and wrongful convictions. But this week, Attorney General Jeff Sessions announced he is suspending those efforts.

It's hard to overstate how bad the junk forensics problem is. Take microscopic hair analysis: The FBI found that 26 of its 28 such analysts overstated the likelihood of matches over a 20-year period. As of March 2015, the FBI had found erroneous testimony in 96 percent of the 268 cases it reviewed in which hair analysts testified against defendants. In 33 of those cases, the defendant received the death penalty – 14 were executed or died in prison prior to the review. FBI Director James Comey also recommended governors order reviews of cases in which hair analysis was used to convict defendants, because the FBI trained hundreds of examiners in its faulty techniques, and those examiners testified in state cases.

Then there's bite-mark analysis, which isn't scientifically valid, but has been used to convict 25 innocent people who have been exonerated and an unknown number who haven’t. Forensic dentists can't even determine whether an injury is a human bite mark with any consistency, let alone whether a particular human inflicted it. 

And the FBI abandoned unreliable examinations of bullet lead meant to show crime-scene bullets matched other bullets owned by defendants only after providing such testimony in more than 2,500 cases, over a period of decades.

Even when the science is sound, junk practitioners have botched simple blood tests and mixed up DNA samples; even fingerprint analysts have shockingly high error rates

In 2013, the Department of Justice and the National Institute of Standards and Technology formed a commission to engage independent experts to improve the reliability of forensic evidence. The National Commission on Forensic Science, made up of scientists, judges, attorneys, researchers and law enforcement officials is tasked with making recommendations to the DOJ concerning the validity of forensic science, how quality control in forensic labs can be improved and what protocols should govern evidence seizure, testing and analysis.

But Sessions has had enough with these meddling scientists. He plans to scrap the commission and replace it with an in-house adviser and maybe some other commission – presumably a more prosecution-friendly one. He and his friends at the American Association of District Attorneys who supported the move have indicated they prefer that standards for prosecutions are left up to prosecutors. The Washington Post reported that Sessions has also ordered the FBI to stop its review of past cases involving suspect evidence.

This is not the first time Sessions has demonstrated he isn't bothered by convictions based on bad evidence. As a U.S. attorney in Alabama, he argued against the reversal of Larry Randall Padgett's conviction for murder, even though prosecutors had withheld evidence indicating Padgett was innocent. During the presidential campaign, Sessions attested to Trump's "law and order" cred by pointing to an ad Trump took out in 1987 calling for the execution of the Central Park Five before the teens had even been tried. Sessions' praise for the ad came years after the Central Park Five had been exonerated by DNA evidence.

Sessions also rejected the findings of a 2009 report of the National Academy of Sciences that Congress had commissioned in light of DNA exonerations showing hundreds of people had been wrongly convicted in cases involving inaccurate forensic testimony; Sessions said he worried that examining validity of evidence used in courts would "leave prosecutors having to fend off challenges on the most basic issues in a trial."

Mike Ware, executive director of the Innocence Project of Texas, has represented many clients wrongfully convicted on the basis of junk science, including the recently exonerated San Antonio Four; he previously ran the Conviction Integrity Unit of the Dallas district attorney's office. Ware, whose state is ahead of the curve in dealing with junk forensics, says Sessions' move harkens "back to the old ways of getting people convicted with quasi-science.

"For people like Sessions, it worked perfectly well in the old days," Ware says. "Whenever there is a move to inject improved methods, protocols and techniques, there is always a backlash among prosecutors and police."

Though it should have been apparent for decades that the "expert" testimony of retired cops and moonlighting dentists didn't comport with the scientific method, it is undeniable that prosecutors use garbage evidence to lock up innocent people in light of hundreds of DNA exonerations.

But those DNA exonerations will not solve the problem of junk science in courts. Though it is now known that bite-mark analysis is junk, it hasn't been banned from courts. Furthermore, some forms of junk science can't be refuted by DNA – sometimes the problem isn't that they got the wrong guy, but that no crime occurred. The San Antonio Four were locked up for decades for a sexual assault that never happened, based on the debunked theory that discolorations on the hymen indicate trauma; parents and caregivers have been convicted of murder in "shaken baby syndrome" deaths of infants who may have fallen or suffered from undiagnosed conditions; and shoddy arson investigations have led to murder convictions for likely electrical fires.

In recent years, new technologies have only presented new opportunities to use poorly analyzed evidence to win convictions. Defendants may lack the resources for experts who can analyze GPS data or information subpoenaed by the government from telecommunications companies. Ware is currently representing a client who was convicted on the basis of forensic video analysis. "I believe it is a sound science," he says, "but there are a significant number of charlatans in it who can put together a dog and pony show that will fool judges and juries."

One reason junk science is such a menace is that once it starts to be accepted in court, it's very hard to undo the damage. Defendants generally only have recourse post-conviction, if ever. "Junk science still has a shelf life of five or six years before it starts being exposed as unsound science at best or fraud at worst," says Ware. "Meanwhile, someone is locked up based on bullshit."

Keeping innocent people from being deprived of their freedom is the most basic responsibility of the criminal justice system, so you'd think addressing America's egregious and well-documented junk evidence problem would be among the attorney general's top priorities. Instead, Sessions seems to be actively working to protect prosecutors from standards and oversight that would prevent future wrongful convictions, while blocking discovery of those that have already occurred. This makes it difficult to conclude anything but that Sessions doesn't care whether the people imprisoned in the United States are factually guilty. He must believe that either the wrongfully convicted, who are disproportionately minorities, are undesirables who are probably guilty of something else – as Trump has suggested of the Central Park Five – or that they are justifiable collateral damage in a "law and order" society. Like Trump, who has repeatedly abused his power by accusing his foes of criminal conduct without any evidence, Sessions doesn't appear to think facts and fairness are his responsibility. That makes both America's president and its top prosecutor serious threats to the rule of law they are tasked with upholding.