Error, and the Death Penalty
The case of Troy Davis, put to death by the state of Georgia despite what looks to me and many others like reasonable doubt, puts us face to face with the awful psychological consequences of being wrong. The harder you look into questions of guilt and innocence, the more you come up against the powerful emotions that surround the possibility of error. Some people will do almost anything to avoid admitting that they did the wrong thing. There seems, for some people, to be nothing worse in life. This horror operates at both the personal and institutional levels. People hate to be wrong themselves and will fight like Tasmanian devils to avoid admitting error. People also hate for the institutions they work for and believe in to be questioned. Rather than focus on getting at the truth, many policemen, prosecutors, and judges act to defend their faith in the judicial system. If re-opening old cases would show that the courts made a mistake, they would rather leave an innocent man in prison. There is a class of person that simply cannot live with uncertainty and must believe absolutely in the rights and wrongs of things. Unfortunately, such people are probably more likely to get into criminal justice careers than those inclined toward skepticism. The result is a system that too often opposes the pursuit of the truth, when that pursuit risks showing that somebody has made an awful mistake.
Institutional opposition to re-opening old cases was on display in the two most controversial recent death penalty cases, those of Troy Davis and Cameron Todd Willingham. Seven of the nine witnesses against Davis have recanted their testimony against him, and one of the other two has bragged on several occasions that he was the actual shooter. Despite this tide of uncertainty, the state of Georgia fought against allowing him a new trial. The state's argument at every step was, as Andrew Cohen put it, that "its interest in the finality of its capital judgments is more important than the accuracy of its capital verdicts." Fear of executing an innocent man should not, argued the state, be allowed to interfere with the smooth functioning of the system.
The case of Cameron Todd Willingham is well known now, thanks to David Grann's extraordinary story in the New Yorker. (If you haven't read this yet, do it.) Willingham's children died when his house burned down, and he was convicted of their murders after investigators thought they had found proof of arson in the house. Witnesses who had at first said Willingham was distraught and had to be restrained from running back into the flaming house changed their stories after he became a suspect, testifying at his trial that he seemed calm and distant. The arson investigators were relying on what amounts to firefighters' folklore about the signs of intentionally set fires, old ideas that have been rejected by scientists. The prosecution's case was bolstered by psychaitrists who testified that Willingham was a classic sociopath, a kind of testimony that is very controversial among both judges and psychaitrists. (One of the psychaitrists who testified against Willingham was expelled from the American Psychaitric Association three years later for repeated ethics violations that included reaching diagnoses without ever examining the prisoner.) So this was a case, not only of one wronged man, but of real problems in the Texas criminal justice system that one would think the state would like to sort out. But no. The state refused Willingham a new trial. An important report by arson expert Gerald Hurst, arguing that the original investigation of the fire was flawed and reached the wrong conclusion, was not even reviewed until after Willingham had been executed.
Willingham's death was not the end of the story. Because of controversy surrounding the forensic evidence in this and other cases, in 2005 the Texas legislature created the Texas Forensic Science Commission to investigate the possible misuse of such evidence in capital trials. In 2009, just when it looked like the commission might find that the evidence against Willingham was flawed, the chairman and three other board members were replaced by governor Rick Perry. Perry made no secret of his distaste for the commission's activities and accused the replaced chairman of exceding his authority. The new chairman says they will only look into general concerns and not review particular cases. It certainly looks like Perry launched this back-room maneuver to avoid having the state of Texas pronounce, officially, that it had made a mistake in a death penalty case. To Perry, how Willingham is remembered, the feelings of his relatives, and the actual integrity of the judicial system in Texas are less important than the system's reputation for infallibility, or just the need to accept its verdicts as the final word.
The personal attachment of many police and prosecutors to their conclusions is even greater, and stranger, than the reluctance of institutions to alter their verdicts. Peter Neufeld, co-director of the Innocence Project, has many strange stories about the reluctance of prosecutors to admit error. Even in cases when DNA evidence leads to an overturned verdict, the prosecutors cling to their belief in the guilt of the accused:
“We’ll be leaving the courtroom after an exoneration,” Neufeld says, “and the prosecutor will say, ‘we still think your client is guilty, and we’re going to retry him.’” Months go by, “and then finally the prosecutor comes back and says, ‘we’re agreeing to dismiss the charges, not because your client is innocent, but because with the passage of time it’s too difficult to get the witnesses.’ Or, ‘we don’t want to put the victim through the hardship of having to testify again after she got closure.’” Such prosecutors give up the case, but not their conviction that they are correct. To the bitter end and beyond, Neufeld says, “There’s a whole category of prosecutors and detectives who still say ‘I can’t tell you how, I can’t give you a logical explanation, but there’s no doubt in my mind that your guy is guilty."In a case from Billings, Montana, the Innocence Project won a reversal for Jimmy Ray Bromgard, convicted of the 1987 rape of an 8-year-old girl. After semen and pubic hairs found in the girl's underpants turned out not to be Bromgard's, the court set him free. Bromgard then sued the state of Montana, and during that suit Montana Attorney General Michael McGrath was deposed about the case. McGrath refused to accept the DNA evidence. First he tried to argue that Bromgard was a chimera, that is, a person with two sets of genes resulting from the fusion of fraternal twins in the womb. He was not. Then McGrath began to offer increasingly wild theories about how the semen and pubic hair of someone other than the rapist could have ended up in the victim's underpants:
McGrath: The semen could have come from multiple different sources.
In other words McGrath, and many others like him, would say absolutely anything rather than admit to the apparently obvious fact that he and the original verdict were wrong.
The part of the legal system I know the most about is forensic science, and the same problems are rampant in crime labs. Every scientist would tell you that the only way to know how accurate crime labs really are would be to carry out regular blind tests. They all refuse this. They would rather send innocent men to jail with faulty results than find out how many mistakes they make. When independent scientists have done checks, they have found disturbing error rates. Studies of finger print matching find false positive rates as high as 30 percent. The FBI has only tried one experiment on fingerprinting that the public knows about, when they sent one set of prints to 39 different labs. Thirty thought it matched the suspect, but nine (23 percent) thought it did not. They have never used this method again. Because they refuse to do these checks, the methods of forensic scientists are not really scientific at all, and their work could not get published in the lowest tier of science journals. When challenged, forensic scientists prattle on about all the safeguards they have in place, which is nice but has nothing to do with science. The only scientific way to insure the quality of the labs would be blind tests conducted by independent investigators. It is the result of the experiment that matters, not people's assertions about how careful they are. And if the lab directors really believe their labs are so good, why are they afraid of being tested? Of course a serious program of blind testing would cost millions of dollars, maybe billions. But the radiocarbon labs I use to date charcoal from my archaeological sites are regularly tested in this way, and can it really be that archaeological results are more important, and worth a higher degree of care, than tests that might convict a man of murder? No. The labs will not test themselves because they refuse to confront the possibility that they are sometimes wrong. To them the system's reputation for perfection, or their own belief in their own infallibility, are more important than the truth.
Our refusal to confront our mistakes is, of course, not just a problem in the courts. Political pundits are notoriously reluctant to admit error, as are many politicians. Even many scientists, who ought to be the most devoted to the data and the most willing to change their views, become irrationally attached to their old positions. The scientists who proposed that a comet impact in North America 13,000 years ago caused the Younger Dryas cold spell, an idea now widely mocked, have refused to back down:
Such intransigence has been seen before in other cases of grand scientific claims. Sometimes those theories were based on data irregularities. Other times, the proponents succumbed to self-delusion. But typically, advocates become so invested in their ideas they can’t publicly acknowledge error. . . . all of these phenomena may be in play, apparently creating a peculiar bond of desperation as the theory comes under increasing attack.In science, though, the side with the better evidence will eventually prevail. For Cameron Todd Willingham and Troy Davis, it is too late. So it is in the field of criminal law that we must push hardest against the all-to-human refusal to face our mistakes.
Since difficulty admitting error is a very common trait all across the political spectrum, I have been wondering why support for the death penalty is not. In America, it is strongly associated with conservatism and especially with Christian conservatism. Those people working to re-open the cases of Willingham and Davis, and the people who provide the funding for the Innocence Project, are overwhelmingly liberal or libertarian. I wonder if it this divide is related to religious faith. Is there a psychological tendency that includes both a need to believe in God, as described in infallible holy writ, and a need to believe that the judgments of lawcourts are sacrosanct? Does skepticism about God correlate with a lack of interest in the awesome majesty of the court? Do people who think the truth is more important than the reputation of the system also think science is a better way to study the universe than reading the Bible? As I wrote at the start of this essay, I have a sense that the kind of person who needs to believe in the infallibility of the Bible is more likely to insist that judicial verdicts be allowed to stand unchallenged. Some people can live with ambiguity and uncertainty, and others cannot. Religious fundamentalism is at its root a mindset that cannot stand uncertainty. Since the search for truth cannot be carried out without strenuously confronting the possibility of error, it falls on all of us who care about truth and justice to fight to keep fundamentalist thinking away from the lawcourts.
September 30, 2011
If have done this, says my memory. I have not done it,
says my pride. Eventually, memory yields.