How Exonerations Happen and What DNA Evidence Changed

In 1992 two New York lawyers named Barry Scheck and Peter Neufeld started a legal clinic at the Benjamin N. Cardozo School of Law in Manhattan. The idea was narrow and, at the time, a little unusual: law students, supervised by the two lawyers, would take on cases of prisoners who claimed they were innocent and whose cases had biological evidence, such as blood or semen, that could be tested with the new science of DNA. They called it the Innocence Project. Within a few years the two men would be famous for something else entirely, as members of the defense team in the O.J. Simpson trial, where they attacked the handling of DNA evidence. But the work they started at Cardozo turned out to be the more durable legacy. Case by case, it showed that American courts had convicted innocent people far more often than almost anyone in the system had been willing to admit.

In short: DNA testing, first used to free a wrongly convicted American in 1989, revealed that eyewitness errors, false confessions, flawed forensic testimony and official misconduct had sent innocent people to prison and death row. Exonerations still depend on a slow combination of new evidence, dedicated lawyers, cooperative or reluctant prosecutors and courts. The lessons have produced reforms in lineups, interrogations and forensic labs, but the system remains better at convicting than at admitting mistakes.

A test invented in an English laboratory

The science arrived from an unexpected place. In 1984, a British geneticist named Alec Jeffreys at the University of Leicester realized that certain regions of human DNA varied so much between people that they could serve as a kind of genetic fingerprint. Within a few years the technique was used in an English murder investigation, where it first cleared a young man who had confessed to a killing he did not commit and later identified the actual killer.

American law enforcement quickly embraced DNA as a tool for identifying the guilty. Defense lawyers soon realized it could do the opposite. Evidence collected years earlier and stored in police property rooms or court files could be tested to see whether it matched the person convicted.

In 1989, an Illinois man named Gary Dotson became the first person in the United States exonerated by DNA testing. He had been convicted of a rape in 1979. The accuser had recanted years before, but courts had not freed him. DNA testing finally showed that he could not have been the source of the evidence.

The first man freed from death row

Kirk Bloodsworth was a former Marine and a waterman on Maryland's Eastern Shore. In 1984 a nine-year-old girl named Dawn Hamilton was raped and murdered near Baltimore. Witnesses described a man, and Bloodsworth, who had no criminal record, was identified. He was convicted and sentenced to death in 1985.

He insisted he was innocent, and after years of appeals and a retrial, he learned about DNA testing from a book about the English case. Testing of evidence from the crime, arranged by his lawyer, excluded him. He was released in 1993, the first person in the United States to be sentenced to death and later exonerated through DNA evidence.

A detail from his case shows how slowly the system can move even after the truth is known. The person who actually committed the crime was not identified until 2003, when the DNA profile was run through a database and matched another man, who was already in prison for a different crime. Bloodsworth had spent years free but still under the shadow of suspicion before the case was finally solved.

Why innocent people are convicted

As DNA exonerations accumulated, researchers could study them as a group. Because DNA proved innocence so clearly, these cases offered something rare: a set of convictions known to be wrong, which could be examined to see what had gone wrong.

Eyewitness misidentification appears in most DNA exonerations. Human memory does not work like a recording. It is reconstructed each time, and it is vulnerable to stress, the presence of a weapon, cross-racial identification, suggestive police procedures and the confidence that grows when a witness is told she picked the right person.

The case of Ronald Cotton and Jennifer Thompson became the best-known example. In 1984 in Burlington, North Carolina, Thompson, a college student, was raped. During the attack she deliberately studied her attacker's face so she could identify him. She later picked Ronald Cotton from a photo lineup and a live lineup and testified against him with complete confidence. He was convicted. In 1995 DNA testing excluded Cotton and identified another man, Bobby Poole, who was already in prison. Cotton had served more than ten years. He and Thompson later met, became friends and wrote a book together about memory and justice.

False confessions appear in a substantial share of DNA exonerations. It is one of the hardest facts for jurors to accept: why would anyone confess to something they did not do? But long interrogations, exhaustion, youth, intellectual disability, fear, false claims by investigators about the evidence and the promise that confessing will end the ordeal can all produce false confessions. The five teenagers convicted in the 1989 Central Park jogger case gave confessions after lengthy questioning; their convictions were vacated in 2002 after another man confessed and DNA matched him.

Flawed forensic science contributed to many wrongful convictions. Some disciplines, such as microscopic hair comparison, bite-mark analysis and some forms of arson investigation, were presented in court as more certain than the science supported. A 2009 report from the National Academy of Sciences found that many forensic methods had not been rigorously tested. In 2015 the FBI acknowledged that its examiners had given flawed testimony about hair comparisons in the large majority of the trials it reviewed.

Official misconduct, including withheld evidence, coached witnesses and false testimony, appears in many exonerations, particularly in homicide cases.

Jailhouse informants who testify that a defendant confessed to them in custody, often in exchange for leniency, are another recurring factor.

DNA cases are only the visible part

DNA evidence is available in only a fraction of criminal cases. Most crimes leave no biological evidence that identifies the perpetrator, and in many old cases the evidence was lost or destroyed.

The National Registry of Exonerations, founded in 2012 and now run with support from several universities, tracks exonerations of all kinds, not just those involving DNA. Its data show that DNA cases are a minority of known exonerations. Many others come from recanted testimony, newly discovered witnesses, proof of police or prosecutorial misconduct, or reinvestigations by prosecutors themselves.

The registry's database has grown to include thousands of cases since 1989. Those are only the ones that have been discovered. No one knows the true number of innocent people in prison, and anyone who claims a precise figure is guessing.

How an exoneration actually happens

The path from a claim of innocence to freedom is long, and it rarely moves in a straight line.

1. A claim reaches someone who will listen. Innocence organizations receive many more requests than they can investigate. They screen for cases where new evidence might exist.
2. Investigators look for new evidence. That can mean finding stored biological evidence, re-interviewing witnesses, finding alternative suspects, or reviewing forensic testimony against current science.
3. Testing or new proof is sought. Every state now has a law allowing post-conviction DNA testing in some circumstances, but the rules vary, and the Supreme Court held in 2009, in District Attorney's Office v. Osborne, that there is no general federal constitutional right to such testing.
4. A court must be persuaded. Even strong new evidence must be brought through procedural rules about deadlines and the kinds of claims allowed. In Herrera v. Collins (1993), the Supreme Court suggested that a claim of actual innocence, standing alone, does not ordinarily entitle a prisoner to federal relief.
5. Prosecutors decide whether to fight. Some agree that a conviction should be vacated; others contest new evidence for years. Whether charges are dismissed or retried after a conviction is vacated is the prosecutor's decision.

That fifth step has changed in an important way. In 2007, the Dallas County District Attorney's office under Craig Watkins created what is generally recognized as the first conviction integrity unit, a team within the prosecutor's office dedicated to reviewing claims of wrongful conviction. Many other prosecutors' offices have since created similar units. The best of them have produced many exonerations; others exist mostly on paper.

What freedom looks like the day after

Exoneration stories usually end at the courthouse door, with a hug and a camera flash. For the person walking out, that is where a different set of problems begins.

Someone who has spent ten or twenty years in prison comes home to a world that has changed. Phones, banking, job applications and housing searches all work differently. Family members may have died. Children have grown up. Many exonerees have no savings, no recent work history and no credit record. Some have health problems from years in prison.

There is a cruel irony in how support is structured. People released on parole after serving a sentence for a crime they did commit often have access to re-entry services: a parole officer, transitional housing, job programs. People who are exonerated, because they are no longer under any criminal sentence, may walk out with nothing at all. Some leave with a criminal record that still shows up on background checks until a separate process clears it.

Compensation, where it exists, can take years. In many states the exoneree must file a claim, prove innocence again under a separate legal standard, and wait for a payment that may be capped or reduced. Some states exclude people who pleaded guilty, even when the plea was made under pressure and later shown to be false. Civil lawsuits against police or prosecutors face legal doctrines such as qualified immunity and absolute prosecutorial immunity that often block recovery.

Innocence organizations and exoneree networks have stepped in to fill some of these gaps, offering social workers, peer mentors and emergency funds. Exonerees themselves have become some of the most persuasive advocates for reform, testifying before legislatures about what went wrong and what they needed when they came home.

The argument about finality

Defenders of the current rules make a serious point. Criminal cases must end at some point. Victims and their families deserve to know that a verdict will stand. Witnesses' memories fade, and evidence disappears, making retrials harder over time. Endless appeals clog courts and can be used to delay just punishments.

That argument has force. But finality is a value only when the result is right. A legal system that is more interested in protecting its judgments than in correcting them has its priorities upside down. DNA evidence showed that convictions supported by confident eyewitnesses, confessions and expert testimony could still be wrong. Here is our view: the rules about deadlines and procedure should bend when there is credible new evidence of innocence, because the alternative is knowingly keeping an innocent person in prison.

The reforms that came out of the cases

The good news is that the system has learned some lessons.

Better lineups. Many police departments now use "double-blind" lineups, in which the officer running the lineup does not know which person is the suspect, and record the witness's confidence immediately.

Recorded interrogations. Many states and the federal government now require or encourage recording entire custodial interrogations, so courts can see how a confession was obtained.

Forensic oversight. Some states have created forensic science commissions to review disputed methods and laboratories.

Evidence preservation. Laws in many states now require biological evidence to be preserved, sometimes for as long as the convicted person is in custody.

Compensation. Most states and the federal government now have laws providing compensation to people wrongly convicted, though amounts and eligibility rules vary greatly, and some people exonerated receive little or nothing.

Many of these reforms passed with bipartisan support, because a wrongful conviction is also a failure to catch the actual criminal. In many DNA exonerations, the real perpetrator committed more crimes while an innocent person sat in prison.

What you can do

• Support innocence and legal aid groups. Innocence organizations operate in most states, often through law schools. You can find groups by state in our nonprofit directory under justice reform and legal aid.
• Ask your state legislators about reform. Laws on evidence preservation, recorded interrogations, eyewitness procedures and compensation are mostly state laws. Look up your legislators and their contact details on our representatives page.
• Ask prosecutor candidates about conviction review. When you vote for a district attorney, ask whether the office has a conviction integrity unit, how it is staffed and whether it publishes its results. Registration deadlines and voting rules for your state are on our voting page.
• Serve on a jury with an open mind. If you are called, remember what the research shows about eyewitness confidence and confessions. Weigh the evidence, not the certainty with which it is presented.
• Follow the larger reform story on our criminal justice reform page.

The people freed by DNA did not get back the years they lost. What they gave everyone else was proof that the system can be wrong, and the knowledge to make it wrong less often. It is up to the rest of us to use that knowledge.