Am I correct in understanding that the fingerprint matches on the bag by the FBI consisted of 11 and 18 points?
No. From chapter 4d:
"Latona's exhibits reflect that there were 9 points of similarity between Oswald's left index finger and the bag fingerprint, 15 points of similarity between Oswald's right palm print and the bag palm print, 11 points of similarity between Oswald's right palm print and the palm print on Box D, 13 points of similarity between Oswald's left palm print and the palm print on Box A, 10 points of similarity between Oswald's right index finger and the fingerprint on Box A, and 11 points of similarity between Oswald's right palm print and the lift from the rifle. And yes, you are correct. Only two of these would have been accepted by most American examiners, and none--not one--would have been accepted by a European examiner.
At least not in '63... Over the decades that followed, the FBI convinced experts around the world that they needn't count points, and that an expert can just "know" when two prints are a match based upon an individualized and instinctual algorithm built upon the number of similar points, and the rarity of these points (aka "hunch").
This was, of course, a recipe for disaster. It was only a matter of time, after all, before an "expert" or group of "experts" came to the wrong conclusion in a high profile case. The first crack in the dam came in 1997 when four Scottish experts found 16 points of similarity between a latent print found at a crime scene and the print of one of the detectives on the scene, even though the detective claimed she hadn't been in that room. This led to her termination, and a 1999 lawsuit in which she proved the "experts" had made a mistake and that the print was not her own. Now, this was a mistake in which 16 points were identified. By 4 experts. It seemed clear, then, that the FBI, with its looser standards, was capable of making a similar mistake.
It took five years for such a mistake to surface. In 2004, the FBI identified the left index fingerprint of Brandon Mayfield, an American Muslim, as the print of a terrorist behind an explosion in Spain. Even though the FBI could find no evidence Mayfield had visited Spain, or had even left the U.S., ever, he was imprisoned. The Spanish authorities, to their credit, rejected this identification, and kept searching. But the U.S. Government, feeling certain the FBI was correct in their identification, refused to release Mayfield. Weeks passed. Eventually, the Spanish authorities matched the print the FBI claimed was Mayfield's to a known terrorist, and the U.S. government agreed to Mayfield's release. He sued the government and was awarded 2 million dollars. Oops.
No, actually it was more than oops. The FBI''s embarrassing mistake led to its re-appraisal of the sanctity of fingerprint evidence, and to its softening its stance regarding the possibility of a misidentification. In doing so, for that matter, the FBI was finally acknowledging what the scientific community had been whispering for decades. The identification of a suspect's fingerprint at a crime scene isn't the sure-fire proof of guilt it was long claimed to be. It just isn't.
The Myth of Fingerprints (1937-2004)
1. No two fingerprints are alike.
2. Fingerprint examination is a precise science, and fingerprint examiners do not make mistakes.
3. Having one’s prints found at a crime scene is a sure sign of guilt.
The Reality of Fingerprints (2004- )
1. Some fingerprints are so similar that an expert can be fooled.
2. Misidentifications are commonplace.
