The Mare of the Rue Saint-Nicaise: Police Evidence in 1800 Paris

At about eight in the evening on December 24, 1800, the carriage of Napoleon Bonaparte, then First Consul of France, left the Tuileries for the opera, where Haydn’s oratorio The Creation was to be performed. In the narrow Rue Saint-Nicaise it passed a small cart carrying a barrel. Seconds later the barrel exploded. Bonaparte’s carriage was already clear; the windows of the carriage following with his wife Josephine were shattered. Bonaparte went on to the opera.¹² The people in the street were less fortunate. The next day the official Moniteur reported three women, a shopkeeper and a child killed and fifteen injured; a later account published under the name of the police minister Joseph Fouché put the toll at some twenty dead and fifty-six wounded.²³

What happened next is one of the better-documented police investigations of its era, because the government printed it. In the spring of 1801 the Imprimerie de la République published a two-volume stenographic record of the trial, prefaced by the reports of Fouché and of the prefect of police, Dubois, which list day by day who was questioned and what was found.⁴⁵ Read today, it shows something close to forensic method, decades before the word was used in its modern sense: physical evidence preserved, an expert description circulated, witnesses confronted with suspects, expert analysis of gunpowder, comparison of handwriting, and a doctor called to account for a wound he did not report.

A political answer first

The government did not begin with evidence. Bonaparte’s secretary Bourrienne, writing years later, recalled that the First Consul announced that same night that the attack was “the work of the Jacobins,” the radical republicans, some of whom had already been arrested that autumn for plotting against him.¹² Fouché, whom Bonaparte regarded as a former Jacobin leader, doubted it.¹ His own report concedes that the police had been following agents of the royalist chouan leader Georges Cadoudal for weeks, that two of the plotters’ circle were police informers, and that the suspects had vanished after an unrelated arrest alarmed them.⁴

Politics moved faster than the inquiry. A consular decree of January 4, 1801, confirmed by the Senate the next day, ordered 130 people, labelled Jacobins, deported without trial.¹ Separately, a military commission condemned and executed Chevalier and Veycer, men accused of building an earlier “infernal machine” for a Jacobin plot.³ Bourrienne’s later judgment was blunt: “both the innocent and guilty were punished; with this difference, however, that the guilty at least had the benefit of a trial.”¹

Reading the debris

The trail that led to the real bombers began in the wreckage. Fouché wrote that although the blast seemed to have scattered or destroyed every trace, “tous les débris dont la rue était semée, furent conservés et interrogés” (all the debris strewn in the street was preserved and questioned).⁴ The trial record lists the official reports: the commissioner of the Tuileries division had the dead horse and the fragments of cart and harness carried to the prefecture of police on the night of the explosion, and another commissioner wrote up a description of the animal the next day.⁴

The most useful object was the horse. On 4 nivôse (December 25), Dubois had a precise description of the mare printed and posted around Paris, drawn up by Jean-Baptiste Huzard, a veterinarian and member of the National Institute.⁴ Huzard was a leading figure in French veterinary science, an inspector general of the veterinary schools who assembled a library of some 40,000 volumes on the horse and animal medicine.⁶ A specialist’s written description turned a carcass into something that could be recognized by people who had handled the living animal.

It worked within two days. On 6 nivôse (December 27) a grain dealer named Lambel identified the mare as one he had sold, with a cart, for 200 francs on 29 frimaire (December 20) and described the buyer. The same day the farrier Jean-Baptiste Legros recognized her as an animal he had shod for four years, and a scrap dealer named Brunet recognized the remains of the cart as one he had sold to Lambel.⁴ At the trial Legros explained that he had washed the dead horse at the prefecture and found the old wounds he had once dressed.⁴ Lambel added a detail that would matter later: the buyer carried a boxwood snuffbox engraved with a man on horseback.⁴

A chain of witnesses

From the horse the police worked outward. A carriage hirer named Thomas and his wife reported that on the evening of December 20 three men had brought the mare and cart to a coach house on the Rue de Paradis, presenting themselves as traveling merchants bound for Laval with brown sugar. The concierge, the cab driver who had seen the lease arranged, the owner who had rented out the coach house for 25 francs, and some twenty tenants of the building added descriptions. A cooper and his assistant on the Rue de l’Échiquier recalled fitting hoops to two barrels, one with iron hoops, and a wine merchant had sold an empty cask. On 9 nivôse (December 30) the coach house was searched for traces of powder.⁴

The physical evidence and the testimony checked each other. A tenant named Roché said one of the “merchants” had asked him to drill a third hole in each shaft of the cart to shorten the harness. When Lambel came to inspect the wreckage, he had been puzzled to find three holes where his cart had had two.⁴

The method also had a negative test. Between 11 and 14 nivôse (January 1 to 4, 1801), the police brought 223 people already under arrest before the witnesses. None was recognized.⁴ Given that the government had just deported 130 alleged Jacobins, this result, recorded in the prefect’s own report, is quietly significant.

Finding Carbon and Saint-Réjant

The descriptions converged on one man. Information gathered around Cadoudal himself identified him as François-Jean Carbon, known as “le Petit-François,” a veteran of the royalist war in the west. On 25 nivôse (January 15) police searched the flat of his sister, Mme Vallon, and found a half-full barrel of fine powder, cartridges and four blue smocks of the kind the witnesses had described. Tenants of the building mentioned a slight scar near his left eye, which made the descriptions more precise. Her daughters eventually said their uncle was hiding with former nuns on the Rue Notre-Dame-des-Champs, where he was arrested on 28 nivôse (January 18). Fifteen witnesses recognized him. Faced with the material evidence and the confrontations, he confessed and named his accomplices.⁴

One of them, Pierre Robinault de Saint-Réjant, had escaped his lodgings hours earlier. Under his bed police found letters, including one in disguised handwriting that blamed the failure on poor gunpowder and on a grenadier’s horse that had shoved the man lighting the fuse against a wall.⁴ A daughter of his landlady testified that before the attack Saint-Réjant had tested scraps of tinder and loose powder on her mantelpiece, timing them with his watch; he wanted the powder to catch in two seconds and found it took twenty or more.⁴ He was brought in on 8 pluviôse (January 28), after an agent of Cadoudal whom Fouché had left at liberty led the police to his new hiding place. Bourrienne records that on January 31 Fouché brought Bonaparte proof that the attack was royalist, too late to stop the deportations.¹

Powder, paper and a doctor’s silence

Three further kinds of evidence went beyond eyewitness identification. Dubois had the powder from Mme Vallon’s flat analyzed by the administrators of the national powder and saltpeter service, who reported that it had not been made in the national powder works and matched powder seized from royalist prisoners in the Temple. Two health officers also removed a fragment of iron from the leg of an injured architect named Trepsat, showing that the bomb had been packed with scrap metal.⁴

Handwriting was examined by professional “experts écrivains.” At the trial one of them, Augustin Oudart, said that the disguised letter showed some resemblances to known writing of Joseph Picot de Limoëlan, Carbon’s fellow conspirator, but not enough to attribute it. A note passed to Carbon in his hiding place, however, was in an undisguised hand, and comparing it with Limoëlan’s known writing he said “il n’est pas possible de douter de l’identité” (it is not possible to doubt the identity). He also compared a letter signed “Gédéon” with one signed by Georges Cadoudal and found the same pen habits.⁴ The care to separate a firm identification from a weak one is notable.

The last strand was medical. Saint-Réjant had been near the blast. The physician Basile-Jacques-Louis Collin, called to him that evening, told police he found the patient spitting blood and bleeding from the nose with great difficulty breathing, though Saint-Réjant claimed to have had a fall. Later visits found trouble with his eyes and deafness in his left ear.⁴ In court Collin defended himself in professional terms: “Le but que se propose un médecin, est de recueillir les renseignemens propres à assurer son diagnostique” (a physician’s aim is to gather the information needed to secure his diagnosis), and once treatment succeeded, his task was over.⁴ The court disagreed. It convicted him under a pre-revolutionary police ordinance of November 8, 1780, which required anyone practicing surgery in Paris to report the names and addresses of wounded persons they treated, and the nature of the wounds, to the local commissioner. He was fined 300 francs and given three months in a house of correction.⁵

Verdict and limits

The jury deliberated for twenty-four hours. On 16 germinal IX (April 6, 1801) Saint-Réjant and Carbon were sentenced to death. Carbon’s sister, the women who had sheltered him, and the couple who had lodged Saint-Réjant were acquitted of taking part in the plot, but each received three months in a house of correction for failing to register their lodgers with the authorities, as a law of 1796 required. Collin, too, was acquitted of complicity before being punished for his silence. The Tribunal de cassation rejected their appeals on April 19.⁵ They were executed on April 20, 1801. Limoëlan escaped to the United States, and Cadoudal was executed after a later plot in 1804.²

The trial also recorded the cost of the scheme to bystanders. A bread seller named Peusol testified that she had been told her fourteen-year-old daughter had been given twelve sous to mind the cart; the girl had died in the explosion, and the presiding judge called it an atrocity to have exposed her.⁴

It would be easy to overstate the science. The decisive break came partly from informers inside Cadoudal’s circle, the confession came from a man who, by Fouché’s account, hoped his revelations would save him, and the government that praised its police had already deported people against whom, as Bourrienne admitted, there was not “even the shadow of a proof.”⁴¹ The memoirs later published under Fouché’s name present the outcome as his triumph, and they should be read as advocacy.³ Still, the printed record shows something striking: a police force that kept a dead horse as evidence, asked a veterinary scientist to describe it, had seized powder analyzed by the state powder administration, weighed handwriting with stated degrees of confidence, and held a doctor to a legal duty to report the wounded. Much of what later generations would call forensic investigation is already visible in the wreckage of the Rue Saint-Nicaise.

Footnotes

1. Bourrienne, Memoirs of Napoleon Bonaparte, vol. 5 (Project Gutenberg 3555), chapter on the infernal machine. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

2. Sánchez Arreseigor, National Geographic History, December 20, 2019. ↩ ↩² ↩³ ↩⁴

3. Memoirs of Joseph Fouché (Boston, 1825), pp. 132-134 and 138-139. ↩ ↩² ↩³

4. Procès instruit… contre Saint-Réjant, Carbon, vol. 1 (1801): Fouché’s report, pp. v-xi; Dubois’s report, pp. xii-xxxiv; list of procès-verbaux, pp. xxxv-xxxviii; testimony of Collin, pp. 144-146, Peusol, pp. 311-312, Legros, pp. 328-329, and Oudart, pp. 330-332. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰

5. Procès instruit…, vol. 2 (1801): jury’s declaration, p. 315; judgment of 16 germinal IX, pp. 328-339; Tribunal de cassation, 29 germinal IX. ↩ ↩² ↩³

6. La Bibliothèque Mondiale du Cheval, “Huzard père, the man with 40,000 books.” ↩