The Other Police Violence

Misconduct in interrogations and trials has sent hundreds of innocent Black men to prison.

An illustration of forensic evidence.
The New York Post / Getty / Katie Martin / The Atlantic

What Richard Miles experienced at the hands of the police was not captured on cellphone video, did not involve a dramatic altercation with seven bullets to the back or a knee on the neck. His experience was slow, almost invisible, but still devastating—more like a cancer than a heart attack. Miles, who spent 15 years in a Texas prison for a murder he did not commit, told me the images that are horrifying the public today are only a starting point in a sequence of events that has ruined the lives of hundreds of innocent Black men who landed in prison because of police malfeasance.

For Miles, the loss amounts to years spent in a cell, the death of his father before he was cleared, the stigma of having spent time in prison, and others’ suspicion that he really is a killer. “All of those wounds, I still walk with mentally,” he said. “We have all died by the misconduct of people in authority.”

In a new study published by the National Registry of Exonerations, four researchers analyzed 2,400 exonerations from 1989 to 2019. (The study defines exoneration as an instance when a prisoner has been officially cleared based on new evidence of innocence.) Perjury or false accusations (which were lumped together in the study) are the single most-common source of such convictions, but police misconduct ranked second—responsible for nearly 850 of the convictions examined in the report—ahead of mistaken witnesses, jailhouse informants, bad forensic testimony, and rogue prosecutors. Researchers found that police bent or broke the rules in 35 percent of the cases—tampering with witnesses, coercively interrogating suspects, fabricating evidence, hiding exculpatory evidence, and lying at trial. Samuel Gross, a professor emeritus at the University of Michigan Law School and the study’s editor, says that when you also take into account the behavior of prosecutors (and some of the cases included both prosecutorial and police misconduct), government officials’ misconduct contributed to more than half of all bad convictions. “And it probably occurs in other cases, because we don’t know about a lot of misconduct that did occur,” Gross adds. “By its nature, it’s concealed.”

The more serious the crime and the punishment, the more likely it is that police and prosecutors behaved improperly or illegally, says Gross, who also founded the National Registry of Exonerations, which tracks wrongful convictions. Official misconduct contributed to 72 percent of cases in which the person was wrongfully convicted of murder, compared with 32 percent of nonviolent crimes. “There’s more at stake, and they care more about getting a conviction,” he says. “It’s one thing to let a drug dealer off, but it’s another thing to let a murderer off. The correct and virtuous impulse to get a conviction can lead people to cut corners for what they believe is getting the right result.”

“There’s a ton of pressure to close those kinds of crimes,” agrees Brendan Cox, a retired police chief in Albany, New York. Police want to bring closure to the victim’s family and security to the community, by taking a killer off the streets. This creates a paradox. “For the police, we are sworn to uphold the Constitution, so it’s our job to actually ensure people’s rights,” he says. “But it’s then also our job to try to make sure that we bring somebody to justice. And some of the things that we’re supposed to do, that we absolutely should do, that we must do, fly in the face of getting that justice sometimes.”

One particular form of misconduct has led to more bad convictions than any other: concealing exculpatory evidence. Although prosecutors have largely been blamed for such strategic omissions, police also hid evidence, which then contributed to the convictions of 172 innocent people. One of them was Richard Miles.

On May 16, 1994, witnesses at a Texaco station in Dallas saw a Black man shoot two men sitting in a car, killing one and wounding the other. Within moments, police picked up 19-year-old Miles, walking home after spending the evening with friends. One witness thought he looked like the shooter; five others said there was no resemblance. Police corroborated Miles’s alibi; even so, he was charged with murder and attempted murder.

“The only thing that I can think of is the police wanted to close out a case, and they picked out the first Black male they saw,” Miles told me. “They knew that they would be able to fit this person into a case. And that’s what they did.”

Miles was convicted and sentenced to 60 years in prison. A dozen years later, Centurion Ministries, a nonprofit group that reinvestigates questionable convictions, took up his case and found that police had suppressed key evidence: A woman had told police that her ex-boyfriend had confessed to the murder, and another person had sworn that a different man, not Miles, was the shooter.

“What did they do with that information?” asks Jim McCloskey, the founder of Centurion and the co-author of When Truth Is All You Have, his memoir of the innocence movement. “They buried it. They didn’t give it to the DA, and they certainly didn’t give it to the defense. They did no investigation of that information whatsoever. That is clear misconduct and suppression of evidence leading to another likely suspect.”

Miles’s conviction was vacated in 2009; after 15 years, he walked out of prison, exonerated. This is police violence of another name. And Miles’s story resembles those of other police-violence victims in another way, too: the racial dynamics of the case. Gross and his researchers identified official misconduct in 78 percent of the cases where Black people were wrongly convicted of murder, versus 64 percent for white exonerees. The gap grew when an innocent person received a death sentence: 87 percent of the cases involving Black death-row inmates, versus 68 percent for innocent white inmates. All of the officers in Miles’s case were white.

An underlying theme to these mistakes is that the behavior is not necessarily illegal. Police operate with great latitude as they make decisions about whom to pursue, and, once they’ve trained their sights on a suspect, how to gather the evidence. This is particularly true behind closed doors, when they are talking to witnesses and suspects. How many details about the crime can they mention to a witness? How much pressure can they exert to ensure the witness testifies? How long can police hold a suspect to get a confession? How much pressure, how many threats or promises, can they make to persuade the suspect to come clean? “There’s way too much gray area,” Cox, the former chief, says. “Way too much is left up to everybody’s interpretation.”

Cox recalls a case from 1997, in which two of his detectives questioned two suspects until they eventually admitted to a murder. Nearly 20 years later, another man confessed to the crime, and the two men were released. “I think our techniques did show that we had some weaknesses,” Cox notes. The National Registry of Exonerations study found that police exacted false confessions time and again. For example, Richard Lapointe, a mentally disabled man, confessed to the murder of his wife’s grandmother after nine hours of questioning: “If you say I was there, I guess I was,” he told the detective, “but I don’t remember being there.” He was exonerated after 23 years in prison. And 16-year-old Bobby Johnson, who has an IQ of 69, was threatened with the death penalty until he confessed to killing a 70-year-old man; he was exonerated nine years later.

What is so disturbing about aggressive interrogation tactics—even many of the legitimate ones—is not merely that they can lead to the imprisonment of an innocent person, but that they pervert the entire process, rendering the pursuit of justice a farce. Consider the case of Davontae Sanford. Early in the morning on September 17, 2007, the 14-year-old was awakened by gunshots two blocks from his house, in a rough neighborhood of Detroit. Police responded to the quadruple murder, and found Sanford in his pajamas outside his home. What happened next was perfectly legal. Police violated no laws when they interviewed and then aggressively interrogated Sanford for the better part of two days and nights. It was legal for them to talk with Sanford, who had been classified in school as learning disabled, without either a parent or a lawyer present. It was legal for them to lie about the evidence and tell him that blood from the crime scene had been found on his shoes, when they had not tested them. And it was legal, if close to the line, for police to let the teenager see crime-scene photos, which showed the layout of the room and the position of the bodies.

Megan Crane, an expert on interrogations, says police are allowed to psychologically manipulate suspects and lie about the evidence. They’re allowed to make vague threats (Unless you tell us what happened, things are going to be very bad for you), but are not supposed to make explicit ones (You’re going to be charged with murder unless you confess)—although they usually get away with it. They can’t make specific promises of leniency (If you confess, you can go home), but they can make vague ones.

“They can say, I just want to get you back to school, or I want to get you home to your mom—which is what they said to 14-year-old Davontae,” says Crane, who worked on Sanford’s post-conviction litigation as the co-director of Northwestern University Pritzker School of Law’s Center on Wrongful Convictions of Youth. “So there’s an implied promise, but that’s a pretty blatant implication that you’re going to go home. And let’s be honest, you’re not going home when you confessed to a quadruple murder. But a kid might not get that.” Sanford recanted his confession three weeks later during an interview with a psychologist, saying he felt pressured to confess.

If police had merely interrogated him in this manner, the misconduct might not have been discovered. But at trial police said that the boy had drawn a diagram of the crime scene—when in fact, a deputy chief had drawn the diagram and asked Sanford to draw the bodies, which he could do because the police had shown him crime-scene photos. The officer was never charged with perjury. Additionally, a hired gunman confessed to the quadruple murder in painstaking, corroborated detail, but the detectives never said a word. When the perjury and suppressed evidence were discovered, Sanford’s conviction was vacated, all charges against him dropped; he was officially exonerated and received compensation for the years of incarceration. He was 23 by then, having spent more than a third of his life behind bars.

(Sanford is suing the Detroit Police Department for violating his civil rights. The department has responded that a private lab in Pittsburgh retested Sanford’s shoes, and claims that blood matches DNA found at the crime scene, even though the lab noted that the “preliminary unconfirmed results should not be used for civil or criminal justice purposes.” Earlier tests by the state did not identify a match. Sometimes even an exoneration does not end the story.)

Some evidence suggests that certain types of misconduct are waning, in large part because the innocence movement and DNA analysis have demonstrated the pervasiveness of wrongful convictions. Gross’s analysis found that police are less likely to exact or threaten violence during interrogations than they were 20 years ago, and that there’s less improper questioning of children in child-abuse cases, and less forensic fraud. He says many wrongful convictions would be avoided if police taped the entire interrogation (currently 26 states require that interrogations be recorded) and improved the way eyewitnesses identify suspects, particularly how lineups are conducted (31 states have made some reforms).

But many, like Crane, the interrogation expert, think that police are “just getting smarter or savvier.” Even though police are no longer allowed to use physically abusive tactics to exact confessions, she believes they are still routinely pried from people involuntarily—otherwise, why would more than a quarter of innocent people cleared by DNA confess to a crime they didn’t commit? “It’s almost worse than the old days, when police could beat suspects into confessing, because it’s so much harder to prove,” she says. “You’re not going to have photos that document the bruises and cuts that resulted from the physical beating, because it’s become much more sophisticated.”

Police have little incentive to change, says Mark Godsey, a federal prosecutor turned innocence lawyer and the author of Blind Injustice, his book about flaws in the justice system. In the marketplace, a company that gets sued has a reason to change its practices. Not so with the police: Qualified immunity makes it nearly impossible to sue a police officer, he says, and police unions protect them even after their misconduct comes to light. “These two factors together have helped create a culture of power without accountability, which is always a toxic mix.”

That leaves a long road ahead. Bodycams and cellphones have raised awareness about police mistreatment of Black citizens. In a sense, exonerations are like video footage: They are hard evidence, after the fact, that something went wrong between arrest and conviction. The new study, the first close look at 2,400 exonerations, suggests that time and again, the injustice was perpetrated by the police.

Richard Miles is not surprised by the findings, and thinks the spotlight has been trained on prosecutors for too long. But the trouble almost always begins with the police officer who finds a suspect and presents the evidence to the prosecutor. “That’s the root,” Miles says. “If you don’t deal with the root, then you’ll never deal with the tree of segregation, racism, police misconduct, civil unrest. You’ll never deal with it.”