On Sunday, 60 Minutes aired a segment that questioned the reliability of fingerprints as evidence. The reason, said a public defender, was that never has there been a study that checked the reliability or a standardized set of method to evaluate fingerprints in the United States. Several countries, apparently, have stricter methods... I think in Italy an examiner must see something like 15 points of similarity. There is no requirement in the U.S. Lesli Stahl interviewed an FBI investigator/forensic expert who demonstrated how it's done and he found about 3 or 4 similarities then said this is the same print. She asked how sure he was, and he replied "100%." To me, that would indicate that fingerprint analysis is not a science... since I've always accepted that there are no sure things in science. But can this be true? Is there no study that has been conducted on the reliability of fingerprints as evidence? I would think that this would be the one source of information that would be referred back to time and again in court to validate an examiner's method. The show also included an interview with a man who was convicted of a crime he didn't commit, based upon fingerprints left at the scene. The defense attorny for the guy got several expert witnesses that were certified by the International Association of Identification, or IAI, which only professionals who are already working in the field can take. The examiners who looked at the guy's prints were not certified... the judge did like every other judge in the country.. he accepted fingerprint analysis as a infallible science and sent the guy up. He finally got off and several FBI examiners testified that the bloody print they had did not belong to him. I don't know if this will spark any discussion, but I found the whole thing fascinating. I suspect that it isn't an issue of whether fingerprints can be reliable, but rather a matter of whether the examiner is qualified. You don't have to be certified to examine prints, which is good, since the 60 Minutes stated in the show that a full half of the people who took the test fail it. It only has 15 points of question. I think a study would do a couple things... it would lend credibility to fingerprint evidence, which might be in danger if more and more mistakes are found and defense attornies begin challenging the proceedures more (it only takes one case to become a precedent). Also, it would establish or standardize a methodology that could be accepted in court as valid and also improve the "science" altogether by improving the methods of examiners.