Instead
of the probabilistic approach, fingerprint experts claim that their work
achieves 'individualisation'. This is the belief that a competent fingerprint
match - after all the verification and quality assurance - identifies a single person
in the whole world with certainty. In
other words they say they can always distinguish between the real donor of a
latent mark and the person in the world who has the fingerprint which is most
similar to it. This is claiming an
error rate of zero for competent work. If we can be sure that this is true and
we can be sure that every fingerprint case is executed to the highest standards
then it could be argued that statistical observations about fingerprinting are
irrelevant. However, case 2, proposing a previously unknown crime to explain an
identification is so dangerous that it should never be done, it has no safety
margin.
I
am not aware of any research that has proved or disproved the claim of
individualisation.
I
must point out that in the 'Town Fingerprint Project' the imaginary fingerprint
experts achieved the very low error rate of 1 in 25 billion latent print to
inked print comparisons, this is four times the population of the world. This produced an error rate of 1 in a
million identifications yet this translates to no certainty at all in a Shirley
McKie type situation (case 2). Even in
a normal crime-led situation (case3) this does not mean there is only 1 in a
million chance that the accused did not leave the latent print. This is because you need to get all
the identifications in a case right, including all the hundreds of
eliminations, to avoid prosecuting the wrong person. (check out the Prosecutor's
Fallacy).