Fingerprint Analysis

A defendant's fingerprints match those found at a crime scene. His goose is cooked, right? After all, no two people have the same fingerprints.

Or do they? Nobody really knows. The statistical data needed to analyze the chances that there are two identical fingerprints haven't been collected. But no matter how astronomical the odds turn out to be, fingerprints are no longer quite the courtroom clincher that they used to be.

The reason stems from two court cases on opposite sides of the Atlantic. In Scotland, forensic investigators in the late 1990s found what they said was a police officer's thumbprint in the house where a woman had been murdered. Subsequent events showed that the fingerprint was neither from the officer nor from a thumb (it was from a forefinger). While everyone acknowledges that crime-scene prints are typically small, smudged, and hard to identify, the Scotland fiasco points up the fact that matches are made by people--and people sometimes make mistakes.

Partly on that basis, federal judge Louis H. Pollak sent shivers through the American justice system last January. In U.S. District Court in Philadelphia, he ruled that experts from the FBI wouldn't be allowed to state unequivocally that any crime-scene fingerprint belonged to one of three alleged drug dealers accused of being involved in four murders. The feds appealed, of course, and persuaded Judge Pollak to reverse himself in March.

Still, it's now part of U.S. legal lore that the criteria for declaring two prints to be a match vary widely. As Judge Pollak learned during his hearings, Italy demands at least 16 identical features or patterns, called Galton points. But France requires only 12. The FBI doesn't have any such standard; instead, it relies on two or more fingerprint examiners who agree there's a match.

In persuading Judge Pollak to reverse his January decision, the FBI insisted that its fingerprint-matching procedure has never incorrectly nailed a suspect. But how would we know? Law-enforcement agencies were just as confident of the evidence that convicted scores of people who were later exonerated when modern DNA evidence became available.

Are there people in prison who were convicted on the basis of faulty fingerprint IDs? Could be. It's an uncertainty we'll have to live with for a while longer.

By Otis Port

    Before it's here, it's on the Bloomberg Terminal.