Tuesday, August 18, 2009

Things Change

One of the most important elements of any criminal prosecution in America is the identification of the suspect beyond a reasonable doubt.

This has been a concern for law enforcement officials for millenia. In many ancient societies, criminals were branded or otherwise multilated in order to identify them.

A frenchman, Eugéne François Vidocq (1775-1857), of the Sûreté was the first person to actually try a system that used more scientific (and less brutal) methods to identify criminals. He simply proposed memorizing how people looked using policemen with above average memories - it wasn't much but it was a start. This was aided by the use of photography.

Another frenchman Alphonse Bertillon (1853-1914) developed a more rigorous system using physical measurements of the criminal. Starting around 1870, these measurements were rendered into a formula. About thirty years later the Bertillon System was discredited when it proven to be unreliable in a case in Kansas.

Fingerprints have been also been used to verify authorship or ownership for thousands of years. The Persians used them in the 1500's on government documents and the fact that no two fingerprints seemed alike was observed and a treatise written in 1686 by Marcello Malpighi, a Bologna , Italy, university professor.

The use of fingerprints as a means to ascertain personal identity slowly formed during the nineteenth century. In 1888 Sir Francis Galton, cousin to Charles Darwin, began to investigate the use of a criminal identification system based on fingerprints. The first criminal prosecution to actually use fingerprints to identify a suspect occurred in 1892 in Argentina. Both the Bertillon system and a rudimentary fingerprint classification system were in use in Argentina at the time.

In 1900 England adopted the Henry fingerprint classification system and in 1901 the Fingerprint Branch was created in Scotland Yard.

In the United States, the first government agency to utilize fringerprints for identifying criminals was the New York State Bureau of Prisons in 1903. In 1924 the FBI was ordered by Congress to create a fingerprint identification division. By 1946 over 100 million fingerprint cards were on file.

Now persons who are arrested are fingerprinted by special booking scanners. The defendant just palces his hands on the plastic lenses and a laser scans his prints. The prints are then stored in automated systems (AFIS) that classify them and search for hits.

Fingerprints are still a valuable tool, but the use of DNA has opened up entirely new identification techniques. DNA testing was haled as the new standard for criminal identification; states now routinely take DNA samples from everyone arrested for a fingerprintable offense (misdemeanors and felonies). This was widely greeted as progress, as the most accurate, infallable scientific method for determining the identity of a criminal or victim.

Until now.

Israeli scientists have shown that DNA evidence can be manufactured and used to implicate innocent people. In fact, for someone who has access to the equipment, it is easier to fabricate DNA evidence than it is to fake a fingerprint.

Of course this is not a situation where your typical crackhead will whip up a batch of fake DNA and smear it all over the prostitute he just killed. The equipment and skill set needed for this is still fairly sophisticated and beyond most citizens.

But I wonder about governments and the expensive laboratories they maintain...

1 comment:

Empoprises said...

(Disclosure: I work for an AFIS vendor, but these views are my own.)

As with any criminal technological advance, there needs to be a corresponding advance on the law enforcement end.

In this particular case, the paper was written by someone from a company, Nucleix, that claims to be able to detect fake DNA. "Nucleix’s test to tell if a sample has been fabricated relies on the fact that amplified DNA — which would be used in either deception — is not methylated, meaning it lacks certain molecules that are attached to the DNA at specific points, usually to inactivate genes."

But...I read a blog comment that stated, "But of course any competent molecular biologist will point out that it is simple to use a simple CpG methyltransferase treatment to methylate the amplified DNA." Yeah, simple.

But presumably law enforcement could in turn detect that simple CpG metawhatever treatment, and so forth.

So in effect, DNA becomes like currency and other items - ways need to be developed to determine if they are real or fake.

Then again, several people have pointed out that it's easier to steal someone's hair and plant it at a crime scene, rather than to go through the trouble to fabricate DNA.