Fingerprints were first used as evidence to solve a criminal case in 1892, when investigators in Necochea, Argentina matched a bloody thumbprint on a bedroom door to Francisca Rojas, who then confessed to murdering her two children. That case is the recognized starting point for forensic fingerprinting, though the science behind it had been building for more than two centuries, and courts in England and the United States did not accept fingerprint evidence until years later.
The 1892 Rojas Case in Argentina
In the village of Necochea, two young children were found stabbed to death. Their mother, Francisca Rojas, accused a neighbor named Velásquez of the killings. Juan Vucetich, an Argentine police official who had been developing his own fingerprint identification system, sent an investigator to the scene.1National Library of Medicine. Juan Vucetich and the Origins of Forensic Fingerprinting
The investigator found a bloody thumbprint on the bedroom door. When Rojas provided her own ink prints for comparison, the match was clear. Confronted with the evidence, she confessed to killing her children, reportedly to improve her chances of marrying a boyfriend who disliked them. Rojas was sentenced to life imprisonment.1National Library of Medicine. Juan Vucetich and the Origins of Forensic Fingerprinting
The case did something no earlier use of fingerprints had done. Ancient Babylonian clay tablets and Qin Dynasty document seals had used prints to authenticate signatures and contracts, but no one had used a print to identify a specific unknown person who had committed a crime. The Rojas case proved that a stray mark at a scene could overturn a suspect’s story and establish guilt.
The Science That Made 1892 Possible
Three researchers, working independently, built the foundation that let Vucetich’s investigator interpret that thumbprint.
William Herschel: Proof That Prints Don’t Change
In July 1858, British colonial officer William Herschel had a local contractor in Jungipoor, India press his inked palm onto the back of a road-building contract. Over the following decades, Herschel collected prints from people of all classes. By 1877, as Magistrate and Collector at Hooghly, he required fingerprints on registered deeds and on jail warrants to prevent identity fraud among prisoners and pensioners.2Project Gutenberg. The Origin of Finger-Printing His decades of records gave him proof that a person’s fingerprints stayed the same over many years.
Henry Faulds: The First Forensic Proposal
Scottish physician Henry Faulds was working at a hospital in Tokyo when he began collecting and comparing prints. When Tokyo police arrested a man for burglary, Faulds showed the suspect’s prints did not match those left at the scene; when a second suspect was arrested, Faulds confirmed the match. In 1880, he published a letter in Nature explicitly proposing that fingerprints could identify criminals. He was the first to articulate the forensic use in a scientific publication.
Francis Galton: The Statistical Foundation
Neither Herschel’s administrative records nor Faulds’ insight had the rigor to persuade courts. Sir Francis Galton supplied it. His 1892 book Finger Prints analyzed the fine details (minutiae) of ridge patterns and calculated that the odds of two prints being identical were roughly 1 in 64 billion. He also devised the first workable classification system, organizing prints by loops, whorls, and arches. Galton’s work appeared the same year as the Rojas conviction, and together they marked the point at which fingerprinting could claim both a courtroom success and a scientific basis.
How Fingerprint Evidence Spread to Courts
The Henry Classification System
A single case does not create a legal practice. Fingerprinting needed a filing system that worked at police-force scale. Edward Henry, a British administrator in Bengal, developed one in the 1890s with Indian assistants including Khan Bahadur Azizul Haque, who devised much of the mathematical framework. Their system allowed 1,024 primary groupings based on pattern types and was adopted across British India in 1897 after a government committee found it superior to body measurements.3NY DCJS (Division of Criminal Justice Services). The Fingerprint System
Scotland Yard, 1901–1902
Scotland Yard established its Fingerprint Bureau in 1901. Fingerprints appeared as evidence in English criminal courts the following year. The first notable Scotland Yard conviction based on fingerprint evidence involved a June 1902 burglary where the accused had left a thumbprint on a freshly painted window sill.4Journal of Criminal Law and Criminology. Finger Prints and Finger Printing: An Historical Study
The First U.S. Conviction: People v. Jennings
American courts followed. On the night of September 9, 1910, Thomas Jennings broke into a home in Chicago’s Morgan Park neighborhood and fatally shot homeowner Clarence Hiller during the burglary. Jennings left four fingerprints on a freshly painted porch railing. At trial, police removed the railing itself and had experts enlarge photographs of the prints so the jury could compare them against Jennings’ prison records. Four expert witnesses, including a U.S. government investigator trained at Scotland Yard, agreed the prints were his.5State of Illinois Office of the Illinois Courts. Illinois Supreme Court History: Fingerprints
Jennings appealed to the Illinois Supreme Court, arguing that no statute authorized fingerprint evidence and that it had no basis in common law. Chief Justice Orrin Carter rejected both arguments in People v. Jennings, 252 Ill. 534 (1911), concluding that “there is a scientific basis for the system of finger-print identification and that the courts are justified in admitting this class of evidence.”5State of Illinois Office of the Illinois Courts. Illinois Supreme Court History: Fingerprints Jennings was executed in 1912. By the end of the 1920s, every state court in the country had followed the reasoning of the Jennings decision.
The FBI’s Identification Division
Federal adoption came in 1924, when the FBI established its Identification Division under Acting Director J. Edgar Hoover. The division consolidated 810,188 fingerprint files from the federal penitentiary at Leavenworth, Kansas, and the National Bureau of Criminal Identification.6Federal Bureau of Investigation. FBI Marks 100 Years of Fingerprints and Criminal History Records It gathered prints from police agencies nationwide and manually searched them on request for matches to suspects and crime scene evidence.7Federal Bureau of Investigation. FBI Criminal Justice Information Services Division Celebrates 100th Anniversary of National Fingerprint Repository That repository has grown continuously for a century and remains the backbone of the FBI’s biometric identification services.
Why Fingerprints Beat Body Measurements
Before fingerprinting took over, the dominant identification method was Bertillonage, developed by French police clerk Alphonse Bertillon in the 1880s. It relied on eleven precise measurements of a suspect’s head and body, chosen because Bertillon believed these dimensions became essentially fixed after age twenty.8PMC (PubMed Central). Reconsideration of Bertillonage in the Age of Digitalisation
The system had fatal weaknesses. Measurements were sorted into just a few broad categories per dimension, so matching depended heavily on subjective judgment. Officers in different cities might measure slightly differently, creating mismatches. Fingerprints could be recorded quickly, compared objectively, and filed in a system that scaled to millions of records. Once the Henry system proved itself, Bertillonage was abandoned in most jurisdictions by the early 1900s.
How Reliable Is Fingerprint Evidence Today?
For most of its history, fingerprint evidence carried an almost unquestioned aura of certainty. Galton’s statistic and decades of courtroom use created a perception of infallibility. That perception took serious hits starting in the early 2000s.
The Brandon Mayfield Misidentification
The most high-profile failure involved Brandon Mayfield, an Oregon attorney wrongly linked to the 2004 Madrid train bombings. Spanish authorities sent the FBI digital images of a partial latent print found on a bag of detonator caps. An FBI database search produced Mayfield as a candidate, and three experienced FBI examiners plus a court-appointed expert all confirmed the match.9Federal Bureau of Investigation. Statement on Brandon Mayfield Case The print actually belonged to an Algerian national named Ouhnane Daoud.
A Department of Justice Inspector General review found that the unusually high similarity between Mayfield’s prints and the latent confused the examiners, but the real failures were human. Examiners worked backward from Mayfield’s known prints, unconsciously adjusting their interpretation of ambiguous features to fit the match they expected. They gave excessive weight to tiny details that turned out to be image distortions, and they rationalized away differences rather than following the standard rule that a single unexplained discrepancy should kill a match.10U. S. Department of Justice, Office of the Inspector General. A Review of the FBI’s Handling of the Brandon Mayfield Case The FBI apologized. The case exposed confirmation bias as a systemic risk in fingerprint examination.
The NAS and PCAST Reports
Broader scientific criticism followed. A 2009 report by the National Research Council found that fingerprint identification had been used for decades “without the necessary scientific research-based underpinning.” The report concluded that most forensic disciplines other than DNA analysis lacked the validated research to “consistently, and with a high degree of certainty, demonstrate a connection between evidence and a specific individual.”
In 2016, the President’s Council of Advisors on Science and Technology drilled deeper into error rates. The council found that only two properly designed studies had ever measured the accuracy of latent fingerprint analysis, both conducted after 2011. The report also highlighted that the rules for declaring a match had historically been neither set in advance nor uniform among examiners. Some required a specific number of matching features; others used no fixed numerical standard. Some would discount apparent differences if enough similarities existed; others followed a strict one-discrepancy rule.11Executive Office of the President. Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods
Fingerprints Under Daubert
Federal courts assess whether fingerprint evidence is reliable enough to present to a jury using the framework established by the Supreme Court in Daubert v. Merrell Dow Pharmaceuticals (1993). Under Federal Rule of Evidence 702, expert testimony must be based on sufficient facts, use reliable methods, and apply those methods reliably to the case.12Legal Information Institute (LII) / Cornell Law School. Rule 702 – Testimony by Expert Witnesses Courts weigh whether the technique has been tested, subjected to peer review, has a known error rate, follows maintained standards, and has gained general acceptance. Fingerprint evidence has survived Daubert challenges so far. Defense attorneys increasingly cite the NAS and PCAST findings to argue that examiners should present conclusions with appropriate caveats about error rates rather than claim absolute certainty.
The thumbprint on the Rojas door opened a line of evidence that courts around the world have used ever since. The questions the twenty-first century has raised are not about whether that evidence belongs in a courtroom, but about how examiners should state what it proves.