Black Box Thinking

by Matthew Syed

  Neufeld questioned him on what had become, by this stage, an unshakable belief. If Bromgard is guilty, Neufeld asked, how could McGrath explain the presence of semen from a different man in the victim?

  Kathryn Schulz quotes from the transcript of the exchange:

  McGrath: The semen could have come from multiple different sources.

  Neufeld: Why don’t you tell me what those multiple sources are?

  McGrath: It’s potentially possible that [the victim] was sexually active with somebody else.

  (The victim was 8 years old.)

  McGrath: It’s possible that her sister was sexually active with somebody else.

  (Her sister was 11 at the time.)

  McGrath: It’s possible that a third person could have been in the room. It’s possible. It’s possible that the father could have left that stain in a myriad of different ways.

  Neufeld: What other different ways?

  McGrath: He could have masturbated in that room in those underwear . . . The father and mother could have had sex in that room in that bed, or somehow transferred a stain to those underwear . . . [The father] could have had a wet dream; could have been sleeping in that bed; he could have had an incestual relationship with one of the daughters.

  The transcript runs on for another 249 pages of similar outlandish claims.

  “So we have four possibilities,” Schulz writes. “The eight-year-old was sexually active; her eleven-year-old sister was sexually active while wearing her sister’s underpants; a third party was in the room (even though the victim had testified to a single intruder); or the father had deposited the semen in one perverse way or another.”

  There was, of course, a fifth possibility, but it required McGrath to accept the evidence for what it was, rather than what he wanted it to be. Bromgard was innocent. The state of Montana eventually paid Bromgard $3.5 million in damages. And McGrath failed in his attempt to ban publication of the exchange with Neufeld.

  What was going on? The only way to make sense of this exchange is through the prism of cognitive dissonance. Many prosecutors see their work as more than a job; it is more like a vocation. They have spent years training to reach high standards of performance. It is a tough initiation. Their self-esteem is bound up with their competence. They are highly motivated to believe in the probity of the system they have joined.

  In the course of their investigations, they get to know the bereaved families well and quite naturally come to empathize with their trauma. And they want to believe that in all those long hours spent away from their own families pursuing justice, they have helped to make the world a safer place.

  Imagine what it must be like to be confronted with evidence that they have assisted in putting the wrong person in jail; that they have ruined the life of an innocent person; that the wounds of the victim’s family are going to be reopened. It must be stomach churning. In terms of cognitive dissonance, it is difficult to think of anything more threatening.

  As Richard Ofshe, a social psychologist, has put it: “[Convicting the wrong person is] one of the worst professional mistakes you can make—like a physician amputating the wrong arm.”21

  Just think of how desperate they would be to reframe the fatality. The theory of cognitive dissonance is the only way to get a handle on the otherwise bewildering reaction of prosecutors and police (and, indeed, the wider system) to exonerating DNA evidence. “It is almost like a state of denial,” Scheck says. “They just couldn’t see the new evidence for what it was.”

  In an adversarial system you would expect any new evidence secured by the defense to be looked at with healthy skepticism by prosecutors. You would expect them to give it scrutiny and to look at the wider context to be sure it stacks up. But in case after case contested by the Innocence Project, the sense of denial from many prosecutors and police went a lot further.

  Nothing seemed to budge them from their conviction that the man who had been sent to prison was guilty. Even after the test had been performed. Even after the conviction had been overturned. Even after the prisoner had been released from jail. The problem was not the strength of the evidence, which was often overwhelming, it was the psychological difficulty in accepting it.

  The reframing exercise often took a distinctive path. First the prosecutors would try to deny access to DNA evidence in the first place. When that strategy was batted away by judges, and the test had excluded the convict as the source of the DNA, they would claim that it had not been carried out correctly.

  This didn’t last long, either, because when the test was redone it would invariably come back with the same result. The next stage was for the prosecutor to argue that the semen belonged to a different man who was not the murderer. In other words, the victim had had consensual sex with another man, but had subsequently been raped by the prisoner, who had used a condom.22

  This is the domino effect of cognitive dissonance: the reframing process takes on a life of its own.

  The presence of an entirely new man, not mentioned at the initial trial, for whom there were no eyewitnesses, and whom the victim often couldn’t remember having sex with, may seem like a desperate ploy to evade the evidence. But it has been used so often that it has been given a name by defense lawyers: “the unindicted co-ejaculator.”

  It is a term that usefully captures the power of cognitive dissonance.

  Schulz quotes from a fascinating interview with Peter Neufeld of the Innocence Project:

  We’ll be leaving the courtroom after an exoneration and the prosecutor will say “We still think your client is guilty and we are 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” . . . 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.”

  Some of these contortions would be almost comical if the subject matter were not so serious. In an investigation by Andrew Martin of the New York Times dozens of surreal explanations were uncovered:

  In Nassau County on Long Island, after DNA evidence showed that the sperm in a 16-year-old murder victim did not come from the man convicted of the crime, prosecutors argued that it must have come from a consensual lover, even though her mother and best friend insisted she was a virgin. In Florida, after DNA showed that the pubic hairs at the scene of a rape did not belong to the convicted rapist, prosecutors argued that the hairs found on the victim’s bed could have come from movers who brought furniture to the bedroom a week or so earlier.23

  Of course, the prosecution has a duty to test the claims of the defense. After all, it is possible that the semen in a rape victim was deposited by someone else who was not the murderer. Exploring the context is reasonable and, in many circumstances, necessary. They are only doing their job.

  But notice the contrast here. When prosecutors are assessing evidence at the beginning of a case, DNA is held up as the most powerful evidence there is. That is why it has helped to secure so many convictions. But once prosecutors have secured a conviction, exonerating DNA evidence suddenly becomes highly suspect. Why is this? Festinger would have found it pretty easy to explain: DNA evidence is indeed strong, but not as strong as the desire to protect one’s self-esteem.

  There may also be external incentives at work in the behavior of prosecutors as Brandon Garrett, a law professor at the University of Virginia, has pointed out. “Legal scholars looking at the issue suggest that prosecutors’ concerns about their political future and a culture that values winning over justice also come into play,” he said in an interview with the New York Times. “They are attached to their convictions, and they don’t want to see their work called into question.”24

  But often the sca
le of denial went way beyond any of this. As Barry Scheck told me: “I am not a psychologist, but it seems pretty obvious that some prosecutors just couldn’t bring themselves to accept that they had got it wrong. It was just too raw.”

  And this brings us back to Juan Rivera. You’ll remember that, as a nineteen-year-old, he was convicted of the rape and murder of an eleven-year-old girl on the basis of a confession signed in the middle of a psychotic episode during a four-day interrogation. You will also remember that the DNA test excluded him as the source of the semen found inside the victim.

  “When the DNA results came back showing that Juan Rivera was absolutely not the person responsible for the rape of Holly Staker everyone assumed that that was the end of the case,” Larry Marshall, professor of law at Stanford University, has said. “It was the classic exoneration.”25

  But that is not how it seemed to state prosecutors. They came up with a new story to account for the DNA evidence, a story very different from the one they had presented at the original trial. Holly, an eleven-year-old child, had had consensual sex with a lover a few hours before the attack, prosecutors claimed. This accounted for the semen. And Rivera? He had happened upon Holly after intercourse had taken place. Rivera may not have deposited the semen, they claimed, but he did murder her.

  “It was a grotesque way of squaring the new evidence with their unshakable belief that Rivera was guilty,” Steven Art, one of Rivera’s lawyers, told me. “But it was also totally inconsistent with the overwhelming evidence that Holly had been raped, quite brutally. There were signs of vaginal and anal trauma and stab wounds in her genitals.”

  The prosecutor’s new story may have seemed outlandish and improbable, but the consequences were very real. Rivera did not leave prison for another six years. In a retrial in 2009 the jury discounted the DNA evidence. The power of a signed confession and the graphic nature of the murder were simply too strong to ignore.

  I asked Rivera, who was eventually released in 2012 after a fourth trial, what it was like to sit in his cell while the system resisted the exonerating evidence. He was understandably emotional. He said:

  When the DNA result came back, I was so happy. It showed that I had been telling the truth all along. It showed to the community that I was not a rapist or a murderer. It was an incredible relief.

  But when my attorneys came into my cell to tell me the result there was always a fear at the back of my mind that it wasn’t over. I knew the prosecutors would resist the new evidence. I had this sense of dread that they would find a way of keeping me in prison. But even I was shocked at the new story they came up with. There didn’t seem to be anything that could convince them that I hadn’t done it.

  The nineteen years in prison took an extraordinary toll. “I got stabbed twice and endured three attempted rapes,” he said. “People wanted to hurt me; they thought that I was a child rapist. But perhaps the toughest thing of all, was knowing that I was innocent. No matter how often they twisted the story to fit in with the new evidence, I could at least hold onto that truth.”

  V

  The criminal justice system takes evidence seriously. You could almost say that the entire system is founded on the notion that evidence is sacrosanct and that the best way of arriving at the right answer is to examine it without prejudice. Verdicts are likely to be flawed otherwise. But if trained prosecutors lose their bearings because of a fear of failure, what hope is there for the rest of us?

  Not all trials followed the pattern of Rivera or Bromgard, however. Many prosecutors accepted the strength of the DNA tests, and after suitable scrutiny accepted that wrongful convictions had taken place. Indeed, many support the work of the Innocence Project and recognize that these failures provide an opportunity to adapt the system. But the wider sense of denial has been unmistakable. Sometimes the system itself seems designed not to learn from mistakes but to bury them. Until recently, for example, many states denied access to DNA tests through so-called finality doctrines. These put a time limit on reopening old cases and, by implication, thwarted access to the very evidence that could prove that a wrongful conviction had taken place.26

  “The Innocence Project and other advocates have spent hundreds of hours just arguing against finality doctrines that are used to block inquiries that no fair person would resist,” Scheck has written.27

  Until 1999 New York and Illinois were the only two states that permitted DNA tests after conviction: they also, unsurprisingly, had the most exonerations. Today, all fifty states have statutes allowing post-conviction DNA testing, but many retain time limits. Others do not allow access to DNA evidence if the suspect originally confessed (like Rivera), even if the test could exonerate them.28

  And then there is the attitude of those at the top. It is remarkable that many of the highest courts around the world, including the Supreme Court in the United States, have effectively stated that they would retry cases only if it could be shown that there was a mistake in procedure rather than in fact. As William Renquist, the former chief justice, put it: “A claim of actual innocence is not in itself a constitutional claim.”*

  Think about that for a moment, because it has darkly comic overtones. Defective systems create errors even when procedures are followed. Think of United Airlines 173, where the pilots followed procedure but the plane crashed. It was precisely because of the evidence provided by the crash that procedures were altered (the introduction of Crew Resource Management, for example). That is one of the key ways in which progress happens.

  But the highest courts were refusing to listen to claims of factual innocence unless the original trials contained procedural errors. It meant that factual errors, created by procedural flaws, would not be investigated, still less addressed. For innocent people behind bars, it was a catch-22 of monumental proportions. And it revealed the breathtaking scale of closed-loop behavior within the legal system.

  In chapter 6 we will look at reform of the criminal justice system (and catch up with what happened to Juan Rivera). We will see that when wrongful convictions were investigated by the Innocence Project, systematic defects were revealed in everything from police procedures to forensic science. If these investigations had taken place earlier and the problems been addressed, hundreds of innocent people could have been spared wrongful conviction. As Barry Scheck has written:

  In the United States there are grave consequences when an airplane falls from the sky . . . Serious inquiries are made: what went wrong? Was it a systemic breakdown? An individual’s mistake? Was there official misconduct? Can anything be done to prevent it from happening again? . . . [But] America keeps virtually no records when a conviction is vacated based on new evidence of innocence. Judges typically write one-line orders, not official opinions, meaning that they don’t analyze what went wrong. Neither does anyone else.29

  Chapter 5

  Intellectual Contortions

  I

  The phenomenon of cognitive dissonance is often held up as a testament to the quirkiness of human psychology. It is easy to laugh when we see just how far we are prepared to go to justify our judgments, sometimes to the point of filtering out evidence that contradicts them. It is all part of the elusive trickery of the human brain, it is said, a charming if occasionally troubling aspect of our eccentricity as a species.

  But we can now see that it is so much more than that. So far in this book, it has been argued that progress in most human activities depends, in large part, on our willingness to learn from failure. If we edit out failure, if we reframe our mistakes, we are effectively destroying one of the most precious learning opportunities that exists.

  And the scariest thing of all is that we scarcely realize we are doing it. When, in the initiation experiment discussed in the previous chapter, the students who had been subject to the embarrassing initiation were told the real reasons they had found such a tedious discussion so fascinating, they wouldn’t accept it. “After each particip
ant had finished, I explained the study in detail and went over the theory [of cognitive dissonance] carefully,” Aronson has said.

  Although everyone who went through the severe initiation said they found the hypothesis intriguing and that they could see how most people would be affected in the way I predicted, they all took pains to assure me that their preference for the group had nothing to do with the severity of the initiation. They each claimed that they liked the group because that is how they really felt. Yet almost all of them liked the group more than any of the people in the mild initiation condition did.1

  This reveals a subtle difference between external and internal deception. A deliberate deception (misleading one’s colleagues, or a patient, or a boss) has at least one clear benefit. The person doing the deceiving will, by definition, recognize the deceit and will inwardly acknowledge the failure. Perhaps he will amend the way he does his job to avoid such a failure in the future.

  Self-justification is more insidious. Lying to oneself destroys the very possibility of learning. How can one learn from failure if one has convinced oneself—through the endlessly subtle means of self-justification, narrative manipulation, and the wider psychological arsenal of dissonance-reduction—that a failure didn’t actually occur?

  It is worth noting here, too, the relationship between the ambiguity of our failures and cognitive dissonance. When a plane has crashed, it’s difficult to pretend the system worked just fine. The failure is too stark, too dramatic. This is what engineers call a red flag: a feature of the physical world that says “you are going wrong.” It is like driving to a friend’s house, taking a wrong turn, and hitting a dead end. You have to turn around.