Do we execute innocent people?
“Our book challenges readers to consider the evidence we have carefully arrayed—and to test each phrase in the book against all of the relevant evidence on the point to which readers can quickly link on the web site—and decide for themselves whether our criminal and capital justice systems are reliable enough to keep innocent people from being executed.” — James S. Liebman
This week our featured book is The Wrong Carlos: Anatomy of a Wrongful Execution, by James S. Liebman and the Columbia DeLuna Project. In today’s guest post, James S. Liebman gives an account of the origin of The Wrong Carlos as a research project and book, and explains how he hopes readers will read and react to the story of Carlos DeLuna’s execution.
Be sure to enter our book giveaway for a chance to win a free copy of The Wrong Carlos!
Do we execute innocent people?
James S. Liebman
Do the three dozen American states that authorize death as a punishment for murder execute innocent people? That is the fundamental question at the heart of The Wrong Carlos: Anatomy of a Wrongful Execution, a book coauthors and I published last week with Columbia University Press.
I began thinking about this question in 2000 and 2002, when colleagues and I issued two studies of rates of serious error found by courts in U.S. capital cases: Broken System I: Error Rates in Capital Cases, 1973-1995 and Broken System II: Why Is There So Much Error in Capital Cases and What Can Be Done About It?. The studies and a follow-up article documented judicial findings of serious error in over two-thirds of all U.S. capital cases that courts reviewed between 1973 and 1995. Nearly all of those findings involved the kinds of legal errors known to undermine the accuracy of the determination that the defendant committed the crime and that he or she deserved to die for it.
Our studies sparked a heated public debate over two competing interpretations of the findings. Did courts’ discovery of so much accuracy-undermining error prove that the system worked to winnow important mistakes out of the system before executions occurred? Or did the studies reveal so much smoke in the form of serious legal error that there had to be a fire somewhere—cases in which the courts failed to catch serious mistakes, allowing innocent people to be executed.
By 2006, this debate had reached the U.S. Supreme Court. In a dissenting opinion in a case called Kansas v. Marsh, Justice John Paul Stevens expressed concern that states’ capital justice systems were not reliable enough to avoid executing the innocent. Justice Antonin Scalia responded that Stevens “does not discuss a single case—not one—in which it is clear that a person was executed for a crime he did not commit. If such an event had occurred in recent years, we would not have to hunt for it; the innocent’s name would be shouted from the rooftops by the abolition lobby.”
In late 2003, the first of a rolling set of Columbia law students joined me in searching for an execution to study that might shed light on this debate. Roughly following the logic of serial robber Willie Sutton, who famously chose to rob banks because “that’s where the money is,” we decided to look for an execution case to study in Texas, which has executed more people by far in recent times than any other state. We focused on cases involving what many observers believe to be the most alluring yet least reliable type of evidence: eyewitness testimony. Carlos DeLuna’s case came to light because his 1983 conviction and 1989 execution rested mainly on a single eyewitness identification. At trial, DeLuna had testified that it was not he, but a man named Carlos Hernandez, who had knifed convenience store clerk Wanda Lopez to death.
Initially, a posthumous inquiry into Carlos DeLuna’s case hardly seemed likely to support doubts about our criminal justice system, because DeLuna’s guilt seemed open and shut.
While pumping gas at a Shamrock station in Corpus Christi on the evening of February 4, 1983, Kevan Baker noticed a man inside the gas station store wrestling with the female store clerk. After hesitating to get involved, Baker decided to go to the woman’s aid and approached the door, locking eyes for a few seconds with the fleeing assailant as he exited, threatened to shoot Baker, and ran off behind the store. Forty-five minutes later, after a massive manhunt, police arrested Carlos DeLuna, shirtless and shoeless, hiding in a puddle of water under a pickup truck. When police brought DeLuna back to the Shamrock station, Baker identified him as the man with whom he had locked eyes.
Less than seven years later, after a trial and a full round of appeals, DeLuna was executed by lethal injection at a prison in east Texas. Unlike other cases, such as Gary Graham’s and David Spencer’s, which garnered considerable pre-execution attention based on claims that they were innocent, the run up to DeLuna’s execution was swift and seamless—the case appeared to be as routine and error-free as any could be.
As a law professor friend responded when I described the case to him, DeLuna’s some other dude named Carlos did it” defense looked absurd on its face. Fueling that view, a respected prosecutor in the case had assured the jury that “Carlos Hernandez” was a “phantom”—a fabrication by Carlos DeLuna—and the federal judge who reviewed DeLuna’s conviction some years later agreed that this “Carlos Hernandez” probably never existed.
Nevertheless, a private investigator I knew named Peso Chavez was on his way to Corpus Christi on another matter, and I figured that, before giving up on the case, it couldn’t hurt to ask Chavez to spend a few spare hours looking into DeLuna’s claim that Carlos Hernandez was the real culprit.
Chavez’s initial tantalizing discoveries—read the book to find out what they are—altered the whole tenor of the case. They led us to conduct a painstaking investigation and to collect and analyze what I believe to be the most complete set of materials on a U.S. criminal case ever assembled. The result is The Wrong Carlos and its accompanying web site.
Our book challenges readers to consider the evidence we have carefully arrayed—and to test each phrase in the book against all of the relevant evidence on the point to which readers can quickly link on the web site—and decide for themselves whether our criminal and capital justice systems are reliable enough to keep innocent people from being executed.
Only in the Epilogue do we offer our own conclusion. Your judgments may differ. But we believe your view of the magnitude—and comfort level with continuing to give our justice system the benefit—of the doubt, will never be the same once you read the book.