The story of the West brothers is the stuff of legend—at least in the world of forensic science. In 1903, a man named Will West was processed at a penitentiary in Leavenworth, Kansas. There, he met an identical man, William West, who’d checked into the same prison in 1901.
But, the legend goes, there was a problem: The two men weren’t related in any way.
True, they looked extraordinarily similar. The “West brothers” had similar Bertillon measurements—a system used to compare the dimensions of certain bony parts of the body—but miraculously, they didn’t share a mother or father.
The prison had a problem since they needed to tell the two virtually identical men apart. One of the officers asked for a fingerprint comparison, and history was made on the spot; by looking at the men’s prints, jailers could figure out which West was which.
Suddenly, law enforcement offices around the United States had a powerful, infallible tool. Fingerprint analysis quickly became a common tool at crime scenes, leading to countless arrests and at least 4,000 episodes of CSI: Miami.
Unfortunately, there’s a problem with the story we just told: It’s not really true.
Despite its importance, most tellings are based on some fairly shoddy science and downright ridiculous assumptions.
We spoke with Bill Schade of the International Association for Identification, an organization that certifies fingerprint analysts and other forensic experts. Now retired, Schade worked as a latent fingerprint manager for 46 years, and he has no problem dispelling some of the most common myths of his former trade.
First of all, the West brothers’ case has a simple explanation.
It’s probably the first thing that went through your mind while reading it: The two men were probably brothers. One fingerprint procedural manual (link opens a PDF) noted, “Recent opinions suggest that Will and William West were related, therefore [explaining] the closeness in measurements.”
“That case is almost urban folklore,” Schade tells Urbo. “That was one of the first cases where fingerprints were used to differentiate between people—until then, they relied on photography.”
The West brothers’ case did help to elevate the public’s understanding of fingerprinting, which may have helped the technique spread. However, the technology was already gaining traction with investigators, and there was nothing borderline paranormal about the situation.
“That was the shining moment for fingerprints when they suddenly became accepted,” Schade says. “In 1903, they started to use it in the New York State prison system, so the case is what everyone points to as the moment of acceptance. …But I’ve heard people say they aren’t sure if it’s real, or that it’s become enhanced over the years.”
In other words, if the West brothers hadn’t become the most famous early case for fingerprinting, another set of twins probably would have done so. Still, twins or not, the case highlighted the shortcomings of forensic science in the early 20th century.
“It’s so easy to confuse individuals when all you’re doing is an eyewitness identification or measuring the length of their elbows to their fingertips or other things like that,” Schade says. “Those were rudimentary systems.”
But while fingerprint examiners of the early 20th century could help police find valuable evidence—probably while wearing monocles and riding bicycles with giant front wheels, since, you know, it was the early 20th century—the science had a long way to go.
Criminals don’t tend to leave fingerprints all over the place.
Watch any police procedural, and you’ll see cops put some dust on a doorknob or end table, only to find a set of perfectly preserved fingerprints. That’s not how it works in real life; just ask anyone who’s ever experienced a home robbery.
“When the cops arrived, I expected them to find fingerprints pretty quickly, since that’s what you see on TV,” Brad, a 28-year-old writer living in St. Louis, tells Urbo. Two men broke into his house in early 2018, and while they weren’t exactly careful (a neighbor’s surveillance camera provided evidence that led to their arrest), they did remember to wear gloves.
“The police didn’t find a single print,” Brad says. “They just got this inky black dust everywhere.”
Even if the criminals hadn’t worn gloves, they might not have left anything behind.
“The idea that touching something leaves behind an identifiable fingerprint, each and every time—those of us in the business know that’s not the case,” Schade says. “There’s many factors that affect whether or not the impression left behind can, in fact, be identified.”
Latent fingerprints—the kind that police take with ink prints when they capture criminals—are pretty easy to identify. At crime scenes, however, most fingerprints are fragmentary; they’re not clear or easy to work with.
No, “CSI” isn’t real, and fingerprinting isn’t perfect.
Schade says that shows like CSI have led people to believe that fingerprinting works as definitive evidence.
“Police dramas clearly raise expectations [of forensic experts],” he tells us. “Most jurors are expecting to see scientific evidence in court. And if police don’t process a crime scene for fingerprints, people think you’re being negligent. …I think that’s a good thing in many ways. People deserve that type of service from their local law enforcement. But now we have to be careful that we don’t overstate our conclusions.”
These days, well-trained fingerprint experts will often refuse to make definitive statements about defendants.
“People see it so much on TV, where they’re very quick to say, ‘Oh, that print belongs to him. It must be him and nobody else,’” Schade says. “When in fact, that’s outside the bounds of science. Science will never allow you to be 100 percent with a zero error rate.”
“The problem is that you can not leave a print. It doesn’t mean you didn’t touch the object, it just means you didn’t leave a print.”
So how did we get the idea that fingerprint analysis is infallible? Probably from fingerprint experts.
“We sold that idea for a long time,” Schade says with a laugh. “In fact, we probably believed it ourselves. The process was absolute—the examiner who was trained would always be able to make a decision. We were pretty confident that we’d get it right, and history has proven that, yeah, maybe it’s 99.5 percent of the time, we did get it right. But there have been mistakes.”
The tools have changed, but fingerprint analysis still requires a knowledgeable expert.
Schade says computers have greatly aided fingerprint science, but there’s no magical program for the partial prints found at crime scenes.
“A big misconception is that the machine does all the work,” Schade says. “You know, that we find a print, we put it up on the computer, magic happens in the background, and a name pops out. The computer tells who left that print. Unfortunately, it’s not quite as simple. The systems are getting more and more powerful all the time, but it’s the people who do the work. The system is a tool.”
“If I gave you a nail gun, would you be a carpenter, could you build a house?” he asks. “You need to know how to build a house—the nail gun is a tool. And so the computers and the different visualization systems, they’re all tools that you put in the hands of a fully trained analyst.”
The key term is “fully trained.” Schade says that many police departments have fingerprint “experts” who work with latent prints—those ink prints we mentioned earlier—and that those individuals might be completely unqualified to handle prints from an actual crime scene.
“Just because you know how to do fingerprinting at the jail, that doesn’t mean you can go into the lab and identify prints and testify to that in court,” he says.
Those poorly trained analysts have brought the entire field under scrutiny. Today, organizations like the International Association for Identification are working to train and certify fingerprint experts to rigorous standards.
“People talk about fingerprint science and say that it’s not really science—that it’s some law enforcement technique that’s never stood up to scientific scrutiny,” Schade says. “The truth probably lies somewhere in the middle.”
“… The most notorious mistakes have made people say, ‘Well, wait a minute, the emperor has no clothes.’ So let’s get a little more realistic here. Let’s scrutinize the results we bring to court. Some prints are no-brainers, but others are very difficult to see and compare. And we need to bring that out in the interest of transparency, not only to the prosecution, but to the defense and defense experts.”
Some cases don’t require an extremely high level of scrutiny.
“One of the downsides [of the CSI effect] is we tend to lump all fingerprint evidence together,” Schade explains. “All forensic evidence has to meet the highest protocols. That’s true when someone’s life is in danger, but if a kid was caught in a stolen car, maybe that doesn’t need the highest level of scrutiny that something like a homicide case would require.”
“I don’t mean to minimize that,” he adds. “I know that if it were me or my kid on trial, I’d want the highest scrutiny, but realistically, it’s not necessary.”
Over the last century, several high-profile cases have shown that, even with high levels of scrutiny, fingerprint science is still limited. A low-quality fragmentary print might be impossible to accurately identify, but prosecutors might still use it as evidence for a conviction. The Brandon Mayfield case (link opens a PDF) is a notable example. In 2004, the FBI arrested and held Mayfield in connection with the Madrid train bombings, using fingerprint evidence as part of their justification for the arrest. The bureau’s computer database had identified Mayfield as one of several possible matches; later, he was proven not guilty.
“Everybody likes to point out [in court] that the FBI made this major mistake,” Schade says. “That’s old news already—there are other cases we can point to and say, ‘Hey, something went wrong.’ But it’s small numbers compared to the number of cases looked at every day. Everybody should be under scrutiny, and if you make a mistake, you deserve to be criticized.”
Still, by most measures, fingerprint analysis is an extremely effective tool. One National Institute of Standards and Technology study found that computerized systems that automatically match latent fingerprints are accurate more than 99 percent of the time when matching prints from two or more fingers. While fragmentary fingerprints are less trustworthy, they’re still a valuable part of an investigation.
“I think the science of fingerprint pattern evidence stands up to scrutiny,” Schade says. “We’ve done more good than harm. The errors that were brought to light—it was the science that brought them to light. They weren’t found through other means.”