How it really works
A print on a glass is rarely complete. It's usually part of a fingertip, overlaid on another similar print, smudged and uneven. A computer searches a database and comes up with several dozen similar candidates, then a person takes over, looking at branching points and line endings and making the decision.
Fingerprints are unique, so a match proves guilt.
The uniqueness of a pattern and the reliability of comparing a smudged fragment are different things. The question isn't whether fingerprints are unique, but how accurately an expert can identify a match from a fragment.
A revealing case took place in 2004. After the Madrid bombings, a fingerprint was found on a bag, and US experts confidently linked it to a lawyer from Oregon who had never been to Spain. Two weeks later, Spanish police found the print's actual owner.
The review showed not negligence, but ordinary mechanisms. The match seemed good, and then context came into play: the experts knew that the person fit the other evidence and began seeing a similarity where there wasn't one. Each subsequent reviewer knew the previous reviewer's conclusion.
What keeps a print from being proof
Little is known about the frequency of errors, and that's a separate problem. In several tests where experts were given pairs whose correct identities were already known, the share of false matches was small, but not zero. For a long time, such tests weren't conducted at all, and accuracy was declared to be 100%.
This has led to changes in the rules. Examinations began to be conducted blind, without knowledge of the investigation's theory, and reports are now being urged to replace the phrase “full match” with an honest assessment of probability. A fingerprint remains strong evidence, but it is no longer a verdict on its own.
Finish reading in the Miqo app
1 more minute, then a quiz at the end. Find it under “Crime → Investigation and Evidence”
Point your phone camera here to open the App Store or Google Play.
Sources
- The mistaken fingerprint identification in the Brandon Mayfield case in 2004 was reviewed by the US Department of Justice inspector general. The question of measurable fingerprint-analysis accuracy was raised in a 2016 report by the President's Council of Advisors on Science and Technology