I think the criticism of “forensic science,” generally in the linked Popular Mechanics article is overblown. Much forensic science is very good.
Fingerprints can be matched by computer. The only real dispute there is partial print matches. There was a scandal regarding very poor fingerprinting techniques; there have also been a couple of incidents of outright fraud. But if the prints match, dude, it’s you. And there are many competent fingerprint examiners. I’ve never seen a computer mismatch out of thousands of examples.
I have some expertise in collision reconstruction. It’s certainly true that some techniques used are not as good as others; expressing solid confidence in pedestrian throw is probably a bad idea. But collision reconstruction based on critical speed scuff marks and various other methods are solid physics.
Forensic accounting is valid science. Forensic chemists test for drugs and alcohol with very high accuracy. Properly done ballistics testing is good science. Hair sample comparisons are good science, if not oversold. DNA is good science, but not if you screw up your Bayesian analysis.
Some people testify to silly things. Some people make mistakes. Some people are willing to say things they know aren’t true. Some scientists are underqualified. Some fields—like forensic odontology—lack the rigor of others, and should not be allowed in court barring a prior showing that the person can do what they say they can do.
I think the criticism of “forensic science,” generally in the linked Popular Mechanics article is overblown. Much forensic science is very good.
Fingerprints can be matched by computer. The only real dispute there is partial print matches. There was a scandal regarding very poor fingerprinting techniques; there have also been a couple of incidents of outright fraud. But if the prints match, dude, it’s you. And there are many competent fingerprint examiners. I’ve never seen a computer mismatch out of thousands of examples.
I have some expertise in collision reconstruction. It’s certainly true that some techniques used are not as good as others; expressing solid confidence in pedestrian throw is probably a bad idea. But collision reconstruction based on critical speed scuff marks and various other methods are solid physics.
Forensic accounting is valid science. Forensic chemists test for drugs and alcohol with very high accuracy. Properly done ballistics testing is good science. Hair sample comparisons are good science, if not oversold. DNA is good science, but not if you screw up your Bayesian analysis.
Some people testify to silly things. Some people make mistakes. Some people are willing to say things they know aren’t true. Some scientists are underqualified. Some fields—like forensic odontology—lack the rigor of others, and should not be allowed in court barring a prior showing that the person can do what they say they can do.
But the idea that forensic science is “mostly created by cops who were guided by little more than common sense” seems quite misguided to me.
The sentence doesn’t seem misguided so much as it being used as a complaint. Try:
“Biology was mostly created by nature lovers who were guided by little more than common sense.”
“Computer science was mostly created by math geeks who were guided by little more than common sense.”
“History was mostly created by story tellers who were guided by little more than common sense.”
I’ve heard that, until relatively recently, forensic arson investigation was actually complete nonsense.