"What if Forensic Science is Pseudoscience" Topic

link

Well worth the read.

I've always been skeptical of "scientific" claims of "odds".
It wouldn't matter if I sat on a jury, since jurors are not allowed to ask questions like "Can I see the math?"
I could only be a hold out juror.

I've had a lot of experience in manufacturing. One prime goal is that every single part made must match all the others. Would wire strippers be any different from die-cut and sewn armrests? Only in the precision of the tooling. Craftsman cannot put out wire strippers, none of which match others made with the same tooling.

Interesting article.
And, no. I won't be making comparisons to other "fake science".

I remember reading a fantasy story or two where the authors tried to put magic on the same footing as science. One brought up the Laws of Similarity and Contagion. Sounds like John W Campbell writers.

So, is this how it starts? When behavior profiling becomes unpopular (political), the blindfolded re-labeling of an entire field must begin? :)

Dan

Back a long time ago I attended a series of lectures by
an attorney who had been a Carter appointee to the USSC
as Chief Clerk.

I mention it in this thread because he stated (this was
in the early 1980's) that 'eyewitness testimony is
ALWAYS suspect' (emphasis his) and 'while fingerprints
may not lie, their interpreters can AND DO' (again,
emphasis his).

Hence I am not surprised at the article, nor the
conclusions drawn therein.

Any technique in crime-solving (including behavioral
analysis) which relies upon human interpretation or,
worse yet, human speculation of any sort, should
at a minimum be disallowed in cases where a suspect's
life is at risk, to include mandatory life-sentences.

OTOH, we must be very careful and acknowledge that NOT
ALL FORENSICS deserve to be labeled 'pseudoscience'.

Indeed. And the fact that, just like forensics, many other things that involve probabilities are still vital and science-based.

Dan

Any technique in crime-solving (including behavioral
analysis) which relies upon human interpretation or,
worse yet, human speculation of any sort, should
at a minimum be disallowed in cases where a suspect's
life is at risk, to include mandatory life-sentences.

OTOH, we must be very careful and acknowledge that NOT
ALL FORENSICS deserve to be labeled 'pseudoscience'

I'll admit that I went for the punchy headline after my preferred title hit the character limit. But the article does make a distinction between techniques developed and tested in an academic setting, vs techniques created for the express purpose of getting convictions.

Martin: "vs techniques created for the express purpose of getting convictions."

I thought the entire purpose of forensics was to help catch the bad guys. :)

Dan

No forensic uses techniques developed for scientific inquiry. DNA sequencing wasn't invented to catch bad guys (it was invented to find out who was the baby daddy on Jerry Springer) and a gas spectroscopy wasn't invented to find out where the bad guy farted last either.

I've thought about the 'other side' of the forensics
issue.

If forensics should be viewed askance on establishing
guilt, how about when establishing innocence ???

The 'absence of evidence is not evidence of absence'
idea.

There were so many half-truths in the article I gave up about half way through. I love the part where the author doesn't like judges deciding what constitutes good evidence. I'm guessing he wants an expert of his choosing deciding what's good. The truth is that if you have enough evidence you can find an 'expert' that will say whatever you want.

Prime example. I was in court recently with a young man I caught doing 90+ mph in a 55 zone. This is a jailable offense here. His lawyer tried to argue that my radar wasn't accurate, tried to argue it wasn't line of sight, and intimated he was sending me an article that proved his point. I'm still waiting for the article. I know for a fact he doesn't believe any of that, because he is a state senator and has never introduced legislation to take radar away from police because it's unreliable. But, since he was getting paid, that's what he argues.

I love the part where the author doesn't like judges deciding what constitutes good evidence. I'm guessing he wants an expert of his choosing deciding what's good

Yes, becasue judges can at best be uninformed about science, at worst nig idiots.

Louie Gohmert was a judge, Roy Moore was a Judge, since Judgeships often is political there is no guarantee they know the difference between a donkey and frog.

I love the part where the author doesn't like judges deciding what constitutes good evidence. I'm guessing he wants an expert of his choosing deciding what's good.

Would you be OK with quantum physicists as arbiters of traffic court? Just because someone is learned in one topic does not mean that they are equipped to tackle others. This is PhDitis/Nobelitis on steroids.

Bite mark evidence? Not as reliable in practice under lab conditions than proponents state during expert testimony. They can't get the species right at a high enough rate to be considered reliable, let alone difference between different suspects…

I love the part where the author doesn't like judges deciding what constitutes good evidence.

Judges are best equipped to make legal decisions, such as the legality of the obtaining the evidence and determining the admissibility of the evidence in court. That's their job. They are not the best to determine the scientific veracity, and accuracy of the evidence.

Prime example

Doesn't seem to be an example of your statement, "The truth is that if you have enough evidence you can find an 'expert' that will say whatever you want." at all. A defense attorney is not an expert in radar detectors.

I was in court recently with a young man I caught doing 90+ mph in a 55 zone. This is a jailable offense here. His lawyer tried to argue that my radar wasn't accurate, tried to argue it wasn't line of sight, and intimated he was sending me an article that proved his point.

Sorry to say, in some respect he is correct. The further away you are from the vector of the car's movement, the more in error your radar detector is. Simple geometry. But he is also wrong. If you are standing along the same vector as the car's movement, then your reading is pretty accurate. If you are away from that vector, the relative velocity to you decreases. That means that the car is actually moving faster than your machine's reading indicates. A bad argument on the lawyer's part. Maybe that's why you didn't get the paper.

I know for a fact he doesn't believe any of that, because he is a state senator and has never introduced legislation to take radar away from police because it's unreliable.

Well, there is "unreliable" and then there is "understanding the limitations". As far as legislation goes, many dubious technologies and methodologies are still used. Polygraph tests and drug dogs are still staples of law enforcement, despite tons of evidence showing both techniques are seriously unreliable. And some places are still employing "psychics"! Now we are in the realm of pseudo-science.

Radar does have limitations and will be universally replaced by LIDAR eventually. Due to the speed of light, the vector discrepancies will become moot.

But, since he was getting paid, that's what he argues.

All lawyers do that, don't they?.

gunfreak, if you can give me examples of either of those men making decisions that were actually against the constitution, I'll be happy to hear them. My guess is that you know nothing about their judicial history, but you don't like their politics. Like many on the left, you may believe that because you disagree with someone then they shouldn't be allowed to be judges.

Martin, I don't think I made myself clear, or perhaps you're being deliberately obtuse. The writer, who is clearly a leftist, doesn't think judges should be allowed to make legal decisions based on the evidence before them. There's no doubt that judges make the wrong decision at times, I know I've seen it happen plenty of times, but they have to go with what they are given.

let me use this as an example: There's a new technology that allows police to tell who committed a murder within 12 seconds. Now, it is controversial, it is unproven, and there are arguments that it works, and some that it does not. Who do you want making the decision that it should or should not be accepted? A group of scientists chosen by The Innocence Project? A group of scientists chosen by a police union? Or a judge who listens to both sides and makes the best decision he/she can? I think I know who the author would want, and I know who I would want.

That's all I was trying to point out. While our system is not perfect by any stretch of the imagination, it is the best in the world.

"The further away you are from the vector of the car's movement, the more in error your radar detector is. Simple geometry."

Yes, ,it's what is called the cosine effect. If the target is more than ten degrees out of the center of the beam it will not show the correct speed of the target. However, it will show a slower speed than the target is actually going and it works in favor of the violator. As you noted. I'm sure I didn't get the paper because he was trying to get me to say the accuracy of my equipment would be effected by outside forces. It is not.

I do not see lidar replacing radar anytime soon. Lidar doesn't work from a moving platform at this time, and I don't see how that can be changed with current technology. It doesn't, in my experience, work well through glass. You have to have a window down or be standing outside. I have no intention of keeping a window down or standing outside in 10 degree weather! Since you have to aim it at a specific target, much as one aims a firearm, I don't see how it can be used while moving.

I don't have a problem with lawyers arguing for their client, I have a problem with them putting out information they know to be false. Remember, they're not under oath. In a similar vein to the radar incident, I've had a lawyer argue that the center red light on a pickup truck isn't a brake light, it's a cab light. Even though there's no doubt he knows that that light is a brake light and all pickup trucks have been required to have them since, I believe, 1989.

I've had a lawyer argue that the center red light on a pickup truck isn't a brake light, it's a cab light. Even though there's no doubt he knows that that light is a brake light and all pickup trucks have been required to have them since, I believe, 1989.

Lol. I think that's called cognitive dissonance. For anyone else it would be a somewhat debilitating mental disorder. For lawyers, it's an asset. But we are getting off topic here.

As for the cosine error, it's odd the lawyer would bring that up. Errors in measurement due to being off vector, usually would benefit the speeder. However radar guided anti-aircraft systems have built in cosine error correction in their software. Perhaps that would trickle down to the traffic cop one day.

Didn't know about LIDAR and glass/windshields. Thanks.

gunfreak, if you can give me examples of either of those men making decisions that were actually against the constitution, I'll be happy to hear them. My guess is that you know nothing about their judicial history, but you don't like their politics. Like many on the left, you may believe that because you disagree with someone then they shouldn't be allowed to be judges.

Well one of them was forcibly removed because of breaking the constitution…..

Also, the question here isn't the constitution, DNA analysis didn't exist in the 18th century.

The question here is that a judge has to Decide what is a valid scientific technique or not, and if you're an idiot, that will be hard.

I noticed, in a Google search, that many speeders are trying to get out of their speeding tickets by claiming the "line of sight" errors in the Radar detector give an incorrect reading. Therefore, their case should be thrown out. Actually it is somewhat true as explained above, but the readings are downwardly scaled the more off vector the detector is. The relative velocity of the car is measured less than the car's actual movement.

Isn't this proof that Judges are not the best arbitrators of scientific issues? Could anyone in one of these traffic situations, explain the cosine error? Can any of them remember what cosine even means?

You can't use a broad brush on all judges. I'd be remiss if I didn't mention the great work done by John E Jones III. He is famous for his presiding over the Kitzmiller vs Dover School District trial of 2005.

His summation may be of interest to some. It shows he clearly understood the scientific issues of the case.

PDF link

Yes, of course judges can be good no matter their political leanings. The problem is that in a lot of American courts the judgeship is an elected possion or is placed there by who ever is running the white house at that moment. In other words its a crap shoot if the judge is capable in any subject be it science or any other topic.

And it's not an American problem, norwegian judges to ignore expert opinion/testimony.