quote: More than 12 years later, the state's DNA databank provided what Berry and police could not: a suspect. And yesterday, for the second time in a month, a rapist in a decade-old case was brought to justice in Northern Virginia by the use of DNA. An inmate in West Virginia was sentenced last week in Prince William County to two life sentences for raping a chemistry teacher 14 years ago, and a Stafford man was convicted yesterday in Berry's case by a Fairfax jury, which also recommended two life terms.
The jurors who convicted Angel M. Anderson, 54, did not learn that Anderson is awaiting trial in Stafford County on another set of sexual-assault charges, and they were not told that the Stafford arrest created the link to the attack in Groveton 13 years ago.
But the jury was told that Anderson was convicted of rape once before, in Fairfax in 1979. He was sentenced to 18 years, and state records show that he was paroled after serving 10 years. Less than two years later, he attacked Berry.
Now, I don't think this conviction was a mistake, especially given the man's past convictions. But, as DNA registries grow, and non-criminal DNA is added to the mix, eventually someone will match who is innocent of the crime. And unless he has an airtight alibi for a particular day and time that might be years ago, that person will be convicted.
quote:A concrete example can make it clear why this reasoning is fallacious. Suppose there is a one-in-a-million chance of a match given that the accused is innocent. The prosector says that means there is only a one-in-a-million chance of innocence. But in a community of 10 million people, one expects about 10 matches by pure chance, and the accused is just one of those ten. That would indicate only a one-in-ten chance of guilt, if no other evidence is available.
...
We start with a thought experiment. I have a big bowl with one thousand balls, some of them made of wood, some of them made of plastic. I know that 100% of the wooden balls are white, and only 1% of the plastic balls are white, the others being red. Now I pull a ball out at random, and observe that it is actually white. Given this information, how likely is it that the ball I pulled out is made of wood? Is it 99%? No! Maybe the bowl contains only 10 wooden and 990 plastic balls. Without that information (the a priori probability), we cannot make any statement. In this thought experiment, you should think of the wooden balls as "accused is guilty", the plastic balls as "accused is innocent", and the white balls as "the evidence is observed".
posted
DNA evidence has also freed many, many, many innocent people.
DNA forensics should be thought of as a tool, a useful tool, but nonetheless a tool that is only as good as the person doing the tests, the tests being done, and the tools being used. You only have to think of that jackass in Texas to see what I mean.
Posts: 13123 | Registered: Feb 2002
| IP: Logged |
posted
Oh, luckily false negatives have a much lower relative rate of occurence (to true negatives) than false positives do (to true positives).
If someone's DNA doesn't match, unless the evidence against them is absolutely overwhelming (say, perfectly clear video of them committing the crime and a couple of Supreme Court justices out for their nightly stroll as witnesses ), that should constitute reasonable doubt.
Its only when the DNA does match that we have to go through all the bother of examining the other evidence .
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
posted
Oh, I'm absolutely in favor of DNA matching as evidence in court. What scares me are situations where it provides the only evidence at all.
One prior conviction for a sex crime is enough to assuage my fears in a particular case. But I've seen little understanding of the underlying math by prosecutors, let alone potential jurors.
Couple this with the CSI phenomenon (which almost never touches on this issue at all), and I'm really scared of widespread DNA databases.
posted
In a population as large as the United States, just a match and the fact the guy is a sex offender . . . would seem still within the realm of reasonable doubt to me. I mean, what percentage of our population are sex offenders? Based on this page, plus a more than reasonable inflation of arond seventy thousand for those not under current care/custody/monitoring, the number seems to be about .1% of the population. So the likelihood that at least one of the 9 people (say) who were falsely identified with DNA is a sex offender is .9% That would definitely be a reasonable doubt to me! Not to mention that we don't even know if the attacker actually was a previous sex offender.
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
posted
If the DNA hasn't degraded and there's enough of it you can fingerprint using minisatellite primers. A minisatellite is basically a set of repeats (like CGCGCGCGCGCG) that occurs throughout the genome. The occur in different places in different people and are all of different lengths. The odds of two people having the same pattern is about 1 in 10^5 to 1 in 10^8 (certain populations seem to have more homology than others, there's more homology within families, etc).
On the other hand, if there's only a minute amount of DNA, or if that DNA is degraded you make a set of primers and determine whether both the suspect DNA and the crime scene DNA had the same pattern. My textbook here doesn't give me a convenient number for false positives, but it's safe to say that it's a lot more unreliable than the minisatellite method.
The problem is, DNA from old crimes is quite likely degraded, damaged or lost and so option B is the one that can be used. If your entire case is built off of random PCR primers I don't know that I could ever convict someone. Could you convict on randomly selected parts of a fingerprint matching with a whole fingerprint? And, of course, to the press and (I assume) the general public, it's all just DNA fingerprinting.
Posts: 3243 | Registered: Apr 2002
| IP: Logged |
In a different recent case the guy matched a description from the victim given after the assault. The guy had priors, and was known to be in the area at the time. Not great, but far more reliable than just one match.
posted
Assuming the original sample hasn't been contaminated, a suspect who knows he or she is innocent can always request another DNA sample be taken and the test run against the fresh sample.
That said, nothing is infallible. Some situations are less error-prone than others.
posted
The thought occured to me that it was merely a false connection between Profile A and Case Record B and a fresh sample of the suspect's DNA when compared to the Case Record B would resolve the matter.
But if the Case Record B is corrupted, contaminated or otherwise lost, the results could be irredeemably skewed.
Printer-friendly view of this topic
![[Smile]](smile.gif)
![[Wink]](wink.gif)
), that should constitute reasonable doubt.
