The Crimes of Paris: A True Story of Murder, Theft, and Detection (23 page)

What Bertillon was certainly guilty of — and this was quite characteristic of him throughout his career — was obstinately refusing to admit, once he had passed judgment, that he might be wrong. Through several courts-martial and trials over the next five years, Bertillon would be called on to testify. Offered a chance to change his mind by expanding on his earlier warning that a forgery might have been attempted, Bertillon decided that the forger was — Dreyfus himself! His rationale for this startling conclusion became increasingly convoluted. Bertillon posited that Dreyfus had adopted the formation of certain letters from the handwriting of his brother and wife in an attempt to disguise his own script. In support of his convoluted claim, Bertillon produced a schematic chart that made no sense to anyone who saw it. General Auguste Mercier, the minister of war who had staked his reputation on convicting Dreyfus, brought Bertillon to the president of the republic, Jean Casimir-Périer, to explain his reasoning. Casimir-Périer told a confidant that Bertillon was “given to an extraordinary and cabalistic madness.… I thought I had an escapee from La Salpêtrière or Villejuif [asylums] before me.” The word the French president used to describe the detective was “madman.”
³⁷

The officers who were pressing for Dreyfus’s conviction, however, portrayed Bertillon’s judgment as a scientific certainty. Bertillon himself, swept along by the flattery, declared, “The proof is there and is irrefutable. From the very first day, you knew my opinion. It is now absolute, complete, and admitting of no reservation.”
³⁸
He was never to soften that self-confident assertion, despite later evidence that overwhelmingly contradicted it.

Several more highly publicized trials marked the course of the affair, Bertillon testifying at all of them. At each, he violated the very principle that he tried to instill in those who adopted his system: to see as truly and objectively as possible. Abandoning that only subjected him to more ridicule. At the 1898 trial at which Émile Zola was tried for libel after he accused the minister of war of framing Dreyfus, Bertillon’s performance had drawn hoots of laughter from Zola’s supporters. There too, he brought as part of his evidence diagrams and charts and enlarged photographs to defend his system, but they only confused everyone who saw them.

An observer at the trial, an official of the Ministry of Foreign Affairs, wrote in his journal his impressions of the participants: “I nearly forgot the
handwriting experts,
absurd creatures incapable of agreeing on the most elementary deductions, flinging their extraordinary theories in one another’s faces and running down and abusing one another like Molière’s doctors. But one of them deserves special mention — Bertillon, who is a madman, a maniac, armed with the crafty, obstinate, and powerful dialectic that is the characteristic of interpretative psychosis.”
³⁹
Zola’s defense lawyers mocked Bertillon so cruelly that he became tongue-tied. The judge rescued him by declaring, “Let us say that the witness does not want to speak.”
⁴⁰

At Dreyfus’s second court-martial, Bertillon once again stunned objective observers with his labyrinthine presentations. When the defense attorney cross-examined him, he punctured Bertillon’s self-assurance, drawing laughter from the spectators. An English reporter described the scene: “Now and again M. Bertillon’s voice rose in hateful shrieks. There were interludes when he clenched his fist and struck the bar, swearing that Dreyfus was the traitor. The voice rang out with passion and excitement. You beheld in him the man vain unto madness with confidence in his atrocious phantasies. He was at last taking his revenge for all the insults of those who had called him a fit subject for an asylum of the insane.”
⁴¹

In April 1904, at the Cour de Cassation, three of France’s most eminent scientists reviewed Bertillon’s work. Using the cutting-edge scientific instruments of the time, the trio took a fresh look at the
bordereau.
They tested Bertillon’s argument that the writing had to have come from a person using the type of grid used in drawing military maps and utilized a
macro-micromètre,
developed for astronomy and the mapping of the skies, to check the handwriting and measure the distances between and within the letters. Their instruments were far more exact and technologically advanced than those of Bertillon, and in a sense, he was outmeasured. Stating their conclusions based on statistical probability, they demolished Bertillon’s argument and concluded that Bertillon’s work was devoid of any scientific value.

The scientists’ testimony was crucial in clearing Dreyfus of all charges. Perhaps it was surprising, then, that Bertillon was not correspondingly disgraced and discredited, like some of the military officers who had concocted false evidence against Dreyfus. In fact, Bertillon kept his job, and his prestige does not seem to have declined appreciably.

viii

As impressive as the Bertillon identification system was, it had two flaws. First, it had little use at the crime scene itself: criminals did not leave their measurements behind. Second, its effectiveness was tied to accuracy and diligence. The system’s detractors proved correct when they asserted that police officers required very careful training to make measurements that were precise enough to be effective. Those two deficits were addressed in a new system that began to be developed shortly after Bertillon’s: fingerprinting.

Perhaps it was the humiliation he suffered when he ventured beyond strictly scientific matters that led to Bertillon’s other great failure — his refusal to recognize the importance of fingerprinting. It was particularly poignant, for had he reacted differently, he could have turned this failure into his greatest success.

From antiquity, it has been known that the patterns on the tips of fingers could be used as identification; ancient Chinese documents were sometimes signed with thumbprints. Mark Twain mentions thumbprint identification of a criminal in
Life on the Mississippi,
written in 1883. But the modern use of fingerprints is said to date from a letter written by Henry Faulds, an English physician working in Japan, to the journal
Nature
in October 1880. Faulds based his observations on his examination of Japanese pottery, where the makers had left fingerprints in the clay. He suggested that the prints never changed throughout a person’s life, and he also took the first step toward the classification of fingerprints by describing the three categories of whorls, loops, and arches. This suggestion prompted a response from Sir William Herschel, a British official in India, who since 1858 had been using fingerprints affixed to paper as a means of identification for the residents of Bengal, the territory where he served. Sir Francis Galton, a British anthropologist, took notice of Herschel’s work and began to investigate further; his books and articles, published in the early 1890s, popularized the subject of fingerprint identification. Galton — an adherent of bertillonage — realized that fingerprints could be important as an identification tool, but he failed to grasp their real significance: that they could connect a criminal to the scene of the crime.

The first country in which fingerprints would be used to solve a crime was Argentina, owing to the influence of Juan Vucetich. Born in Croatia, Vucetich had immigrated to Argentina in 1881. He went to work at the Central Police Department in La Plata, where he impressed his superiors with his intelligence and hard work. Two years later he became head of the Statistical Bureau and was given the responsibility of organizing the Department of Identity. The country had been using the anthropometric system since 1889, so Vucetich had to apply bertillonage to the prisoners and suspects. When Vucetich read about fingerprinting, he immediately saw that it would revolutionize police work. Fingerprints, he realized, were easier to obtain and more accurate than Bertillon measurements — but his superior told him to stick to anthropometry.

Pursuing a dual track, Vucetich developed his own system of fingerprint classification while he was filling out anthropometry cards. He identified four common fingerprint traits: arches, prints with a triangle pattern on the right side, prints with a triangle pattern on the left side, and prints with triangles on both sides. Vucetich represented these patterns with the first four letters of the alphabet. He called the system dactiloscopia.
⁴²

In 1892, a sordid murder case gave Vucetich the opportunity to show the worth of his system. In Necochea, a coastal town two hundred and fifty miles south of La Plata, Francisca Rojas informed the police that a man named Valásquez had killed her two children in a frenzy of jealousy. Taken into custody, Valásquez admitted that he loved Francisca, even that he had threatened her, but denied that he had killed her children. Even when police tried to beat a confession out of him, he did not change his story. The local authorities had Valásquez tied up and laid next to the bodies of the dead children overnight in the hope that guilt would make him confess. But the next morning the suspect still claimed he was innocent.

An informant told the police that Rojas, the children’s mother, was involved with a young lover who said he would marry her if not for her children. Focusing now on Rojas, the local police spent a night outside her house rattling windows and making ghostlike sounds, supposed to mimic an avenging angel. (Such was the advanced state of police investigation in Argentina at the time.) That had no effect either. The police now looked for help from the regional headquarters of La Plata. Vucetich sent Carlos Álvarez, one of his most trusted men, to try to solve the case.

Inspector Álvarez examined the crime scene and saw a stain on the bedroom door illuminated by the evening sun. A closer look showed it to be a bloody thumbprint. He cut it out with a saw and took the wooden piece to the station house. He called Rojas in and rolled her right thumb on an ink pad, then pressed the print on a piece of paper. Studying the two with a magnifying glass, he saw that the prints were the same. When he questioned Rojas again, she broke down and confessed. She was tried for murder and convicted.

Vucetich had followed the case and felt it proved his argument for the superiority of fingerprints. He wrote a friend, “I hardly dare to believe it, but my theory has proved its worth.… I hold one trump card now, and I hope I shall soon have more.”
⁴³
He did, and within a few years, Argentina became the first country to change from bertillonage to fingerprints. With this success, Vucetich’s prestige — and his method — spread throughout South America.

A skeptical Bertillon saw fingerprints as a threat to his own system, which, among other things, had brought him the respect he craved. In 1893 he wrote in his
Textbook of Anthropometry,
“Skin markings have insufficiently distinct gradations to serve as a basis for archives.”
⁴⁴
But starting the following year, he began to include some fingerprints on his file cards, taking the prints of the thumb and three largest fingers of the right hand, though the cards were still filed according to the anthropometric system.

Two failures made it clear that fingerprints were needed. In 1901, it was found that twin brothers curiously named Albert Ebenezer and Ebenezer Albert Fox — English criminals who provided alibis for each other — could only be differentiated by fingerprints, not by bertillonage. (Identical twins do not have the same fingerprints.) Two years later in the United States, another case showed the limitations of bertillonage. Two convicts at the U.S. Penitentiary at Leavenworth, Kansas, turned out to have identical measurements. Though unrelated, the men looked amazingly alike and even had the same name, Will West. The only way to tell them apart was through fingerprints.

Ironically, however, Bertillon is often credited as being the first European detective to apprehend a criminal based on fingerprint evidence alone. On October 17, 1902, he went to the rue du Faubourg Saint-Honoré to photograph a murder scene. Joseph Reibel, the servant of a dentist named Alaux, had been murdered in the dentist’s office. A desk and a cupboard with glass panes had been broken open, but so little had actually been stolen that Bertillon immediately guessed that the killing had not been the product of a robbery. While photographing the scene, Bertillon found a pane of glass with fingerprints. He took it to his lab to see if it would be possible to photograph the prints. Sure enough, by placing the glass against a dark background and using an arc lamp, he obtained an excellent reproduction.

Then, because he had some fingerprints on file, having collected them for eight years by that time, he started his assistants checking to see if any matches would turn up. Of course, this was a laborious process because Bertillon’s files were still arranged by physical measurements, and in the search for fingerprint identification, each card had to be examined. Remarkably, a card was found with the same prints (luckily, the killer of Reibel had used his right hand on the pane of glass; Bertillon still did not take prints from the left hand) — those of an ex-convict named Henri-Léon Scheffer, who soon turned up in Marseilles. He confessed and was sent to prison for life. Bertillon’s surmise that the crime was not motivated by robbery proved correct. Reibel and Scheffer had been homosexual lovers, and when Reibel was no longer interested, Scheffer had killed him in a moment of passion.

This latest feat of deduction brought Bertillon new acclaim. A caricature even appeared in
L’Assiette au Beurre,
a Paris newspaper, showing Bertillon peering through a magnifying glass at some prints left on the grimy wall of a public toilet. But in this instance, Bertillon was annoyed by the publicity. He failed to see that the case illustrated perfectly the value of fingerprints: left at the scene of a crime, they could identify the perpetrator as certainly as if he had left a calling card. Even so, as a result of the case, a legend (frequently repeated) grew that Bertillon had actually invented fingerprint identification.