Errors of judgment

View full image

Thirty-two years ago, in 1979, a jury in Illinois convicted Gary Dotson of aggravated kidnapping and rape and sent him to prison for 25 to 50 years. A decade later, in 1989, Dotson became the first prisoner in the United States to be exonerated by DNA. Gary Dotson spent ten years in prison for a crime he did not commit.

Since Dotson was cleared, another 265 men, most convicted of crimes involving rape or sexual assault, have been exonerated by DNA. Of course, only a small fraction of criminal cases—under ten percent—involve DNA. But by studying the DNA cases, we can learn something applicable to the vast number of criminal cases in which DNA cannot set the record straight. Brandon L. Garrett’s sobering new book, Convicting the Innocent, is the most thorough effort to date to examine these criminal cases and understand how and why we make mistakes. A law professor at the University of Virginia, Garrett exhaustively reviewed 250 of the cases involving DNA exonerations to identify patterns. He writes elegantly and clearly, and he identifies reforms that I believe could significantly reduce the number of wrongful convictions in this country.

Consider a case typical of those Garrett discusses: a woman is kidnapped and raped. She tells police two perpetrators attacked her, where the attack occurred, and generally what the men look like. Police arrest a suspect. At a traditional lineup, the victim identifies a man as one of the assailants. She subsequently identifies him at trial. Juries are often persuaded by eyewitness identifications, and the jury convicts this defendant and sends him off to prison. Years into the prisoner’s sentence, however, new DNA technology reveals three genetic profiles in the crime scene evidence. The woman told police at the time that she had not had consensual sex in several months. One of the profiles belongs to the victim. The other two must therefore belong to the rapists. Both profiles exclude the man convicted of carrying out the attack. We jailed the wrong man.

How police and prosecutors err when identifying the suspect, and how juries err in convicting the wrong man, are the central concerns of the book, and Garrett’s exploration of the mistakes in these cases leads him to suggest a number of reforms. We know, for example, that police lineups are frequently flawed and suggestive, and that we can significantly improve the accuracy of identifications with small procedural changes.

For one, Garrett recommends that police replace their typical five-man lineups with a procedure in which witnesses see one person at a time, and the accompanying officer does not know which one is the suspect. This technique has been shown to reduce the incidence of mistaken identifications dramatically. Correcting eyewitness errors is particularly important in that, as Garrett points out, mistaken eyewitness identifications played a role in an astounding 76 percent of cases involving DNA exonerations.

Garrett takes aim at dubious forensic evidence that deserves the name of junk science. He points out that 75 of the exonerees were convicted, in part, on the basis of microscopic hair comparisons. Another 7 were convicted in part on the basis of bite-mark analysis. The National Academy of Sciences has indicated that neither of these methods is based on rigorous science, yet they remain admissible as evidence across the country.

Garrett also discusses the phenomenon of false confessions, another flawed type of evidence common in the exoneration cases. To someone who has never been interrogated, the notion that a person would confess to a crime he did not commit is nearly incomprehensible. Garrett meticulously describes how this occurs and shows that, in at least 16 percent of cases in which DNA exonerated a prisoner, that prisoner had made a false confession. Investigators, sometimes knowingly and sometimes perhaps inadvertently, share crime details that find their way back into the suspects’ statements. Garrett proposes that these investigations be recorded from start to finish. In some jurisdictions they are, but in most of the country, juries have no way of knowing what facts about the crime might have been provided to the suspect by the police. Garrett illustrates the dangers of false confession with a death penalty case—the case of Ronald Jones. In 1985, Jones confessed to a rape-murder in Illinois. Despite recanting at his trial, he was convicted and sentenced to death. Twelve years after the murder, DNA proved Jones was not the murderer.

In one respect, Jones’s case is somewhat unusual; of the 266 DNA exonerations, only 17 have involved death row inmates. But it would be a mistake to conclude that errors in capital cases are uncommon; the better explanation is that capital cases typically do not feature DNA. In fact, the critical lessons Garrett gleans from non-capital exonerations are routinely applicable in capital cases, where confessions, shoddy science, police lineups, and eyewitness testimony are common.

A criminal trial is sometimes said to be a search for truth. Garrett demonstrates that the tools most commonly relied upon in the courtroom are often antithetical to that ostensible goal. Garrett calls the 266 exonerations the “tip of the iceberg,” and he is undoubtedly right. What we do not know, however, is how big the iceberg is. How many innocent people have died in prison? How many have been executed? It would take an unusually ingenuous or disingenuous observer to say the answer is none. But the truth is, we have no idea. I’ve had clients who told me they were innocent and who were convicted, and executed, solely on the basis of eyewitness identification. Some of them might have been telling the truth. In the absence of DNA, we simply cannot know.

The one thing we do know is that we would make many fewer mistakes if states were to enact the reforms Garrett endorses. His riveting book is ultimately based on a value that prosecutors and police have in common with defense lawyers: when an innocent man goes to prison, the actual bad guy is still on the loose, and the criminal justice system has failed.