Sunday, January 30, 2011
Forensics, is it science?
In our current criminal justice system, we rely a lot on something referred to as Forensic science, but is forensic science really science? In order to determine this we have to break down what elements must be present in order for science to truly be scientific. Science relies on sound use of the Scientific Method, this is the primary way for those who practice science to identify something to study and make a hypothesis, then run many experiments or gather a lot of data, after which conclusions can be documented and re-tested by others. It is important to note, these conclusions may not be factual, they could be a result of scientific bias. Bias occurs when a scientist is running tests to attempt to prove something when there are perceived high stakes. The scientific method attempts to minimize bias in the testing phase. One of the easiest ways to alleviate bias is to open your work up for peer review. This is where other scientists can test your hypothesis, and conclusions then either support or refute them. Science is rarely if ever unanimous, so there is always dissenting opinions, and you must be prepared to deal with the bias of other scientists. If a compelling body of evidence, or research done by many scientists confirms a hypothesis, it can then considered a plausible theory. If you are looking to science to prove something, it is not the right field for you, science is never proof, it is simply evidence of a theory that could be proven wrong in the future. While many theories have practical applications, such as the theory of lift for an aircraft, there is still compelling argument about whether lift is simple Newtonian physics, or if it is the Bernoulli effect. In case you're wondering, the FAA doesn't officially recognize Newton's 3rd law an an explanation of lift, even though the Bernoulli effect has been invalidated as a major contributor of lift by many scientists.
The process to begin Forensic detection is normally that a prosecutor, or investigating police meets with a forensic “scientist” and explains the situation and what they want to find out. Occasionally a defense attorney will attempt to use a forensic scientist, however the burden of proof lies on the prosecutor. In America our legal system was founded on the presumption that people accused are innocent until proven guilty in a court of law, this is why it is normally the prosecution utilizing forensics, it is to satisfy their burden of proof. When the prosecutor approaches the forensic investigator, they will normally give a back story about what they think this person did, and occasionally add irrelevant details outlining why this person is a very bad person and needs to be in jail. This establishes bias very effectively, the forensic scientist now has a one sided backstory they are attempting to prove. According to the Association of Forensic Scientists, forensic scientists are not supposed to be an arm of the prosecutor, nor are they supposed to be an arm of law enforcement, to do so would be an ethics violation.
So how come I have met with over 20 forensics scientists, and they all describe their job field as “law enforcement”? When I ask these so-called scientists if their tests have ever been reviewed by other scientists, they all said no. The most effective way to eliminate a single bias is to have a theory tested by multiple scientists, so why is peer review ignored by forensics? A few of the more honest forensic scientists I have talked agreed that they would never put anyone in jail if it was peer reviewed, because there would almost always be dissenting opinions. After years of peer reviewed documents on DNA, it has been found that DNA is an effective manner in which to test individuals for certain crimes, like rape, and murder. So what happens when the prosecutor doesn't get the answer they are looking for? Just ask Jerry Hobbs in Lake County Illinois, they keep the person in prison for years until another person is caught who matches the DNA test, just so the prosecutor can have his number of convictions very high.
What about fingerprints, those are accurate right? While we haven't physically observed 2 fingerprints that are exactly the same, that doesn't mean there aren't any. What's worse is that when observing fingerprints, you're looking for points of minutia, when you get a partial print, one forensic scientist might declare a match, while another might say it's inconclusive. There isn't currently any peer reviewed documents concluding the probability of matching partial prints, or even how much of a print is needed to make a match. What about ballistics? Currently there is no scientific basis for measuring ballistics. When bullets hit objects, they break, deform, shear off. Unless the bullet only hits soft tissue, and isn't an expanding round, you can't be sure. Telling someone they have a ballistic match on a handgun means they think they know what caliber it was, and if they have enough of the bullet they might be able to tell you how many grooves were in the gun, thus limiting them down to the bullet being fired from 1 of 22% of all the 9mm guns that exist. Bite marks? Normally bite marks are comparing 2 pictures to see if they match... art, is not science. There really is no science behind bite marks. So DNA is pretty much all we have.
The majority of the corruption in forensics, is from the prosecutors and police. While the forensic scientist who maybe got their degree online from the university of Florida might be wary of biting the hand that feeds them, many times the forensic scientist doesn't even get the correct evidence to test based on corruptions that are outside of their control. In conclusion, in forensics people use science, but that does not make them scientists, and currently our forensic scientists are more of an unethical arm of law enforcement, than they are part of the scientific community. When you think about the lives they affect, destroy or exonerate, it is hard to imagine this kind of power going unchecked by something as simple as a little ethical peer review.