by Dan Sperber
“Indeed, sir! May I inquire who has the honour to be the first?” asked Holmes with some asperity.
“To the man of precisely scientific mind the work of Monsieur Bertillon must always appeal strongly.”
“Then had you not better consult him?”1
Who was this Monsieur Bertillon, superior even to Holmes as an expert detective? Alphonse Bertillon was, at the end of the nineteenth century, one of the most respected policemen and forensic scientists of the world. He had developed a scientific method for recording the identity of criminals based on anthropometry: a precise measurement of various traits, from the size of the left foot to the length of the right ear.2 The most famous component of this system, still used today, is the mug shot, a standardized way of photographing people who have been arrested.
It is for his expertise in the new domain of photography that Bertillon is contacted by investigators from the French army in early October 1884. His task is to take a series of pictures of an extremely sensitive document, a letter known as the bordereau, written by a French officer spying for the Germans. Bertillon obliges. A few years later, this tiny piece of paper will turn him into the most ridiculed man in France.
On October 13 Bertillon is asked to perform one more task, one that lies outside his area of expertise. In order to identify the bordereau’s writer, he must tell whether the writing on another letter matches that of the bordereau. After a few hours, Bertillon concludes that the same person very likely wrote both letters. That, however, is not conclusive enough for the lieutenant-colonels Henry and du Paty de Clam, who lead the inquest. They want a definitive indictment. And so, two days later, they visit Bertillon once again. Now they give him more information: they have evidence that Captain Albert Dreyfus is the spy who leaked French secrets to the Germans. All they need is a final proof: that the bordereau bears his handwriting. Actually Henry and du Paty de Clam have no other piece of evidence, but Bertillon does not know that, and with the extra confidence afforded by anti-Semitism, he happily starts working with the premises that the Jew Dreyfus is guilty.
Bertillon’s mind works tirelessly with a single purpose: proving that Dreyfus wrote the bordereau. Here’s what he has to work with: two letters—the bordereau and a sample of Dreyfus’s writing—that have some similarities but also marked differences. These differences are sufficient for real experts to conclude that the two letters have not been written by the same person. But Bertillon is smarter than that. Only by imagining what clever deceptions Dreyfus has devised will this connoisseur of the criminal mind be able to prove the traitor’s guilt.
Bertillon wonders: What kind of spy would write such a compromising message in his own hand? (The real spy, as it turns out, but no one knows this yet.) In Bertillon’s mind Dreyfus, a spy, and a Jew to boot, is too shrewd to make such a glaring mistake. He must have disguised his hand. This explains the differences between Dreyfus’s normal writing and the bordereau.
But now Bertillon has another problem: How to account for the similarities? Why hasn’t that shrewd spy simply used a completely different writing? To answer this question Bertillon comes up with his chef-d’œuvre, the keystone of his system: the theory of the auto-forgery.
Imagining what a shrewd spy might do, Bertillon realizes that transforming one’s writing would work only if the potentially incriminating document were found in a nonincriminating place. Then Dreyfus could use the disparities to claim that he was not the author of the bordereau. However, if the letter were discovered on Dreyfus’s person or in his office, he could not simply claim that it wasn’t his. Instead, this master of deception would have to say that he was being framed, that someone had planted the bordereau. But if someone were to try to frame Dreyfus, surely they would be careful to reproduce his handwriting. And so Dreyfus set out to imitate his own handwriting—he engaged in auto-forgery. Frank Blair, a Chicago lawyer writing at the time of the affair, offers a sarcastic summary of Bertillon’s reasoning:
In short the differences between Dreyfus’s natural hand and that of the bordereau, admitted by Bertillon, were, according to him, artfully put in by Capt. Dreyfus to throw off suspicion; while the absolute similarities were put in to enable him, in a proper case, to claim they were traced from his own handwriting, and therefore, done by someone else.3
If Dreyfus’s handwriting looks like that of the bordereau, he is guilty. If it does not, he is guilty. The blatant flaw in the reasoning doesn’t deter Bertillon, who states in conclusion: “I have arrived at a set of observations and comments that embrace all the facts with a comprehensiveness so perfect that the conclusions impose themselves indisputably.”4 Partly on the strength of Bertillon’s “evidence,” Dreyfus is arrested, tried, and sentenced to cashiering and life deportation—in a minuscule cell, on a minuscule island, far away from France.
A year later, Lieutenant-Colonel Georges Picquart takes over as head of military intelligence. Reviewing the files associated with the Dreyfus case, Picquart discovers conclusive evidence that the real culprit is the officer Ferdinand Walsin Esterházy. His handwriting matches that of the bordereau perfectly. When Bertillon is presented with this evidence, even he has to admit that Esterházy’s writing is strikingly similar to the writing on the bordereau. But Bertillon’s mind never falls short of reasons to sustain his views; Bertillon claims that “the Jews have been training someone for a year to imitate the writing”—to become a scapegoat, presumably.5 The higher echelons of the military are similarly unmoved. They already have their culprit; another one would just mess up their story. Thanks to Picquart, Esterházy is tried. Thanks to the generals’ influence, the guilty man is proven innocent, the innocent man proven guilty once more, and the whistle-blower disgraced. Dreyfus stays on his island.
It will take a major social and political upheaval, brought about by Jean Jaurès, Leon Blum, Georges Clemenceau, and, most famously, the novelist Emile Zola and his J’accuse, for the army to reopen the case. In 1899, the original decision is quashed and a new trial is held in Rennes. Bertillon, again, is one of the expert witnesses. His task has become more difficult, as he must now prove that the bordereau is not an undisguised note left by the sloppy spy Esterházy but the carefully designed product of Dreyfus’s devious mind. Bertillon is up to the task. Apotheosis of Bertillon’s system, the deposition runs for more than fifty pages of dense text, plus pictures and exhibits.
Of that bordereau, Bertillon peruses every word, measures every letter, photographs every wrinkle. He sees patterns everywhere. When the thirteenth line of the bordereau is superimposed on the thirtieth, three letters are aligned.6 When the word intérêt is taken out and repeated and the two copies put end to end, they measure 12.5 millimeters, a unit size on military maps.7 Even more damning, a standard subdivision of this unit, 1.25 millimeters, is found everywhere in the word: “length of the t’s cross: 3 [units of 1.25 millimeters]; length of the acute accent: 1; width of the circumflex, one and a half, and height of the final t: 4, etc.”8 Coincidences? Impossible. The bordereau must be the work of a master craftsman who used several templates and a military-issue ruler to create one of the most complex forgeries of modern times.
Such considerations can leave no place for doubt and so, after ten hours of deposition(!), Bertillon gives a forceful conclusion: “In the set of observations and concordances that form my demonstration there is no place for doubt, and it is made strong by a certainty both theoretical and material that, with the feeling of responsibility born of such an absolute conviction, I affirm with all my soul, today as in 1894, under oath, that the bordereau is the work of the accused. I am done.”9
It is hard to tell how impressed the court is with Bertillon’s arguments. In any case, the judges find Dreyfus guilty of treason once again, although with “mitigating circumstances.” This nonsensical verdict reflects more the need to uphold the status quo than the merits (or lack thereof) of the case. Dreyfus’s innocence is plain for anyone to see. Refusing to wait for yet a new trial that may never
happen, Dreyfus consents to be pardoned by President Loubet on September 19, 1899, at the cost of accepting the guilty verdict. He will have to wait seven more years for his final rehabilitation: being reinstated to his former rank in the army.
The Bertillon in All of Us
Bertillon offers a fascinating study in the use of reasoning to defend preexisting beliefs. He seems to have been truly convinced by his own arguments. When three experts were tasked with evaluating his system, they found an “incomprehensible jumble” “totally devoid of scientific value” whose absurdity was “so blatant that one is hard put to explain the length of its exposition.”10 Yet they also concluded that the very “naiveté with which [Bertillon] unveiled the secrets [of his system] would lead one to believe in his good faith.”11
It would be easy to regard Bertillon as a madman—many of his contemporaries did. But that would overlook his otherwise successful professional career, how he rose through the ranks and devised new ways to catch criminals—hardly what you would expect of a lunatic. And lest we feel too smug, every aspect of Bertillon’s thinking has been reproduced in the laboratory, showing how reasoning can lead everyone on the wrong track. These experiments have replicated—on a smaller scale, fortunately—the mental processes occurring in Bertillon’s mind. Unambiguously, they point to reasoning as the culprit.
When Bertillon mentions the perfect comprehensiveness with which he embraces all the facts, he exhibits clear symptoms of overconfidence. According to the intellectualist approach, reasoning is supposed to make us doubt our own beliefs, especially when they rest on foundations as shaky as Bertillon’s. How can reasoning lead to overconfidence instead?
Asher Koriat suggested an answer more than thirty years ago.12 In one of his experiments, participants had to answer general knowledge questions, such as, “Does Corsica belong to France or Italy?” and to specify how confident they were. Participants were overconfident. If they thought, say, that they would be right in 80 percent of the cases on average, they might have the correct answer only six times out of ten.
While some participants were not given any special instructions, others were told to give reasons supporting their answer. This had no effect on confidence because participants had been doing that all along: piling up reasons supporting their initial hunch. This is why they were overconfident. And it’s not as if they were completely unable to think of reasons why they might be wrong—when asked to provide such reasons, they obliged, and became less overconfident. But the myside bias stopped them from spontaneously engaging in this more objective reasoning.
Between his first, not fully conclusive report and the deposition in Rennes fifteen years later, Bertillon has not only become even more confident, his beliefs have also undergone severe polarization. Dreyfus comes out more cunning, the conspiracy wider, Bertillon’s testimony more critical than ever. Reason had had years to push Bertillon toward such extremes, but the roots of polarization can be observed in a much briefer time. When participants were made to think about someone for a few minutes, they ended up liking him more if they had liked him at first, and liking him less if they hadn’t liked him at first.13 In this brief interval of time, they had piled up reasons supporting and reinforcing their initial impression.
Initially, the only evidence Bertillon had of Dreyfus’s guilt was the alleged resemblance between his handwriting and that of the bordereau. When Picquart showed Bertillon Esterházy’s handwriting, which exactly matched the bordereau, Bertillon should have immediately changed his mind. But between his first encounter with the case and Picquart’s intervention, Bertillon had built an unyielding scaffold of reasons. This scaffold upheld his original contention in spite of overwhelming evidence against it, a process known as belief perseverance.
The earliest experimental demonstration of belief perseverance was done by a team led by Lee Ross in 1975.14 Participants were asked to distinguish between real and fake suicide notes, and were told how well they’d done. They were then left to think about their performance for a little while. During this time, they thought of many reasons why the feedback made sense: they had always been very sensitive, they knew someone who was depressed, and so on. Then the participants were told that in fact the feedback had been completely bogus, bearing no relationship whatsoever with their actual performance. But it was too late. Participants had found many reasons to buttress their beliefs, so that even when the initial support was removed, the beliefs stood on their own. Participants who had received a positive feedback thought they had done better than those who had received negative feedback, even after they’d been told the feedback was bogus.
Reasoning Poorly Together
Bertillon offers a sad example of how “starting with a mistake, a remorseless logician can end up in bedlam.”15 He kept accumulating reasons for his initial belief in Dreyfus’s culpability, with a blatant lack of self-criticism, until he reached grandiose and absurd conclusions. Bertillon’s mind seems boundlessly fertile, but it is only one mind. Imagine what people can do when they gang up to find reasons supporting their beliefs.
Conspiracy theories often start small, questioning accepted facts: Why does the American flag allegedly planted on the moon appear on the photos to be waving in spite of the lack of wind? Could the World Trade Center really have crumbled on 9/11 the way people say it did? Some of the people who raise these questions go online, discover pamphlets, find kindred spirits. Soon enough the doubt escalates, alternative answers are found, and pointed questions turn into full-blown paranoia. Officials must be lying. The government has to be in on it—NASA, the CIA, the FBI, the NSA. The conspiracy has to be global, pushed by the United Nations, the Bilderberg group, and more often than not, some form of “international Jewry,” heirs to the anti-Semitic conspiracy theories of Bertillon’s days.16
While the development of vast conspiracy theories involves hundreds of people, smaller groups can also be led astray by reason. In the 1960s, the Yale psychologist Irving Janis started investigating when and why small groups make poor choices. He examined in detail the process that led to disastrous decisions such as the failed attack on Cuba launched by the American government in 1961—the infamous Bay of Pigs invasion. For Janis, the culprit was groupthink, the failure of group members to criticize each other’s suggestions and to consider alternatives.17
The problems caused by a lack of dissent have also been captured in the laboratory, where psychologists have accumulated evidence of group polarization. Put a bunch of people together and ask them to talk about something they agree on, and some will come out with stronger beliefs. Racists become more racist, egalitarians more egalitarian.18 Hawks increase their support for the military; Doves decrease it.19 When you agree with someone, you don’t scrutinize her arguments very carefully—after all, you already accept her conclusion, so why bother? When like-minded people argue, all they do is provide each other with new reasons supporting already held beliefs. Just like solitary reasoners, groups of like-minded people can be victims of belief polarization, overconfidence, and belief perseverance.20
Evolution Doesn’t Care How Good We Feel
Reasoning can lead to outlandish territories—Bertillon developing his system, a conspiracy theorist thinking that lizard men control the earth. It is only natural to dismiss people who adopt such beliefs as crazy or stupid—or, more politely, to suggest that they suffer from cognitive limitations.
The facts do not support this interpretation. Bertillon was neither a madman nor a dunce. Few conspiracy theorists suffer from psychosis or cognitive impairment.21 Moreover, people who cannot be suspected of any mental deficiency share the same plight. Linus Pauling went from seeing vitamin C as a remedy for the common cold to hailing it as a universal cure. The escalation of the Vietnam War was decided by “extraordinarily intelligent, well-educated, informed, experienced, patriotic, and capable leaders.”22
This spells trouble for the intellectualist approach. Not only does reasoning fail to fix mistaken intuitions, as th
is approach claims it should, but it makes people sure that they are right, whether they are right or wrong, and stick to their beliefs for no good reason. Historical examples attest that these are not minor quirks magnified by clever experiments, but real phenomena with tragic consequences.
Psychologists sometimes use a distinction between cognitive and motivational explanations. People who do or believe something wrong either must be the victims of a cognitive failure or must have been motivated to go astray. Since cognitive failures cannot explain the surprising outcomes of reasoning, a sensible move is to offer a motivational account, as Ziva Kunda did when she defended the prevalence of motivated reasoning. According to her, when reason leads people astray, it is because of a “wish, desire, or preference” to reach a preordained belief, whether it is accurate or not.23 People may want to believe something for many reasons, but the most common is hedonic: because it makes them feel good. There would be a desire to believe that leads people to “give preferential treatment to pleasant thoughts and memories over unpleasant ones.”24
At its most extreme, a feel-good account could simply claim that people adopt whatever belief is pleasurable—an extreme form of wishful thinking. “Wishful thinking” is a phrase commonly used to describe someone’s belief when it seems to be grounded not on evidence that the belief is true but on the desire that it were. Moreover, in ordinary conversation, “It’s wishful thinking” is often given not just as description but as explanation: “Why does John believe he is popular? It’s just a case of wishful thinking—he wishes he were!”
From a psychological point of view, and even more from an evolutionary one, wishful thinking is something to be explained rather than an explanation. Our beliefs are supposed to inform us about the world in order to guide our actions. When the world is not how we would want it to be, we had better be aware of the discrepancy so as to be able to do something about it. Thinking that things are the way one wishes they were just because one so wishes goes against the main function of belief.
