Tuesday, July 22, 2008

How to lie with forensics

I've posted on the many problems with the 'science' used to convict people. The process of criminal prosecution is best regarded as a form of theater used to dole out penalties on the appearance of objectivity, befuddling fact finders enough to remove reasonable doubt. Most of the 'science' is crap that wouldn't pass muster in any real field of research.

Now, a crime analyst in Arizona named Kathryn Troyer, doing searches against DNA databases, searches that researchers are usually not allowed to do (and which Troyer was eventually told not to do once officials found out how dangerous her research was!), has found that the nine-loci tests commonly used to convict people do not lead to unique identifications. You know how prosecutors are always spouting off about how there is only a one in so many billion chance that another person could have the same DNA? Bullshit. From the LA Times story:
"In the 1990s, FBI scientists estimated the rarity of each genetic marker by extrapolating from sample populations of a few hundred people from various ethnic or racial groups. The estimates for each marker are multiplied across all 13 loci to come up with a rarity estimate for the entire profile.

These estimates make assumptions about how populations mate and whether genetic markers are independent of each other. They also don't account for relatives.

Bruce Weir, a statistician at the University of Washington who has studied the issue, said these assumptions should be tested empirically in the national database system.

'Instead of saying we predict there will be a match, let's open it up and look,' Weir said.

Some experts predict that given the rapid growth of CODIS, such a search would produce one or more examples of unrelated people who are identical at all 13 loci.

Such a discovery was once unimaginable."

The FBI's research which forms the basis of all the confident proclamations of prosecutors has never actually been tested against the existing databases, and what evidence there is seems to prove that the astronomical certainty commonly used to describe DNA proof is simply untrue.

Bonus: the FBI is using every dirty trick to prevent exculpatory searches, including in one case bluffing to cut the state of Maryland off from the national CODIS database if it allowed searches in its own state databases that could be used to impugn DNA evidence! The threat was used to allow prosecutors to craft an affidavit to have the court block the search. The FBI even claimed that the search itself could corrupt the entire state database! The judge didn't buy it, the search went forward, the FBI didn't cut Maryland off, the database wasn't corrupted, and the search disclosed:
"In a database of fewer than 30,000 profiles, 32 pairs matched at nine or more loci. Three of those pairs were 'perfect' matches, identical at 13 out of 13 loci."

The FBI tried much the same trick in Illinois, where it also failed, and where the search tuned up lots of matches which were officially impossible. They should empty the jail cells of people convicted solely on the basis of DNA and fill them up with FBI officials, who have been allowed to get away with scientific obfuscation, bluffing, and outright lying, without any punishment at all.