CSI: the real world

[dehro]

the TV shows in it's various incarnations (and many similar shows) tell us that science is not a matter of opinion and that evidence, when enough boffins apply their grey matter to it, always points at the culprit, when there's enough evidence to begin with.
case (http://www.youtube.com/watch?v=ULATYrEpiJM) in point (http://www.youtube.com/watch?v=QVX68dO7SPs)

how is it possible then, that in the real world science in courtrooms does indeed become a matter of opinion?
there are plenty of court cases, real ones, where the various expert witnesses, be they experts in automotive, security, firearms, psychology or fingerprints, take a good hard look at the same exact evidence and come up with 2 opposite conclusions..
how? why?
I'm not debating legal validity of expert witness or proof within different judicial systems.. I'm talking about hard data, psychological evaluations (which I realize are very much open to interpretation), and physical bits of evidence.

I realize that a defence attourney will always try and find an expert that gives credit to his current defense strategy..but facts remain facts, right?
when 2 different guys put the same bit of evidence into the same machine they should obtain 2 identical results..right?
and if they didn't, one would be able to prove the other one is lying..right?
yet in several high profile cases, I've seen opposite experts badmouthing each other and saying the opposite side has made mistakes and so on..
for instance one side saying there's enough physical evidence to prove something conclusively (say, DNA material).. and the other side says there is not.. do they not rely on the same science books and machinery?

has anybody here got any experience or indeed expertise to explain how this is possible?

[Serpentine]

Textbook case of the CSI Effect (http://en.wikipedia.org/wiki/CSI_effect).

Take the classic, supposedly hard-and-fast, undeniable piece of evidence of, say, fingerprints. The fact is, fingerprint evidence is none of those things. Among other factors:
1. The idea that "every fingerprint is unique" is, at best, unconfirmable. You would have to fingerprint every single person on the planet to know that for sure, and they certainly haven't done that.
2. Fingerprints are fine, detailed and very similar to one another. Even a slight smudge can completely change the look of it, and how many fingerprints at a crime scene do you think are absolutely perfect?
3. Following from #2, fingerprint identification is actually quite subjective, down to the subjective interpretation of the analyst - as far as I'm aware, the programmes used on CSI to identify fingerprints are close to pure science fiction, or at best not even close to as reliable as the shows suggest.

That's just off the top of my head and from half-remembered articles. You can find plenty more information elsewhere - this article (http://heinonline.org/HOL/LandingPage?collection=journals&handle=hein.journals/ajcl31&div=7&id=&page=) (if you can get at it...), for example, looks like it could cover a good part of it.

Science, evidence gathering, technology and everything else just simply isn't as easy and objective as these shows make out. Everything is subjective and subject to error and/or interpretation: the gathering of the evidence, the interpretation of the results, right down to the basics of just what test to do.
For example: some years ago there was a high-profile case. The details may be off-forum, but it involved a historical event. The question was "is there any evidence that there were toxic gases pumped into these rooms?" To answer it, bits of stone from the walls were examined. In an attempt to "maintain objectivity", the scientists performing the experiment weren't told where it was from: they were told "look for this substance in these rocks", and that's about it. So, they ground them up, looked for the substance, and found negligible levels of it, which was taken as evidence against the use of the toxic gases. Now, when the scientists found out what their work had been used for, they were horrified, and insisted that their results couldn't be used as evidence at all: the substance they were told to look for, if it had been in gas form, would have stayed on the surface of the stone rather than being spread throughout it. If they'd been given more information, they would have known to only look at the outside of the rocks, rather than grinding up the whole thing. Because they did the wrong sort of test, the results were useless.

As an aside, it has been proposed that CSI and the like be classified as science-fiction, because the technologies shown are so very far removed from reality.

[dehro]

heu...just to clarify..I'm not suggesting CSI to be gospel or indeed anything like realistic..the whole, I'll have your DNA tests results ready by lunchtime is laughable..
and yes, I know bits of evidence are never presented in pristine laboratory conditions..unless the murderer happens to be a robot-laboratory-assistant who hasn't fled the scene..
but one would still think that if there is any sort of agreement as to what the best procedure is to evaluate a bit of evidence, this procedure should be followed by both parties..and yield the same results.. and that if such an agreement is not in place, then any sort of result of non-homogeneous procedures should be invalidated as proof on account of not being verifiable in a commonly accepted manner.

the event that sparked this thread is something that is in the news here in Italy (and possibly in UK and USA)..but I didn't want to make this thread about a single case or it might violate forum rules.. but the argument that is being made is that the evidence was not conclusive then and that the tests that said it was were wrong..
except that with the time passing, evidence mislaid or possibly improperly handled over time etc etc..well.. new tests are bound to give slightly different results, whether the initial tests were done to satisfaction or not..

so..is there any point left in analyzing physical evidence at all..if all one has to do is to wait for the appeal, hire a new expert who will say something different and try to invalidate the evidence used?

[erikun]

but one would still think that if there is any sort of agreement as to what the best procedure is to evaluate a bit of evidence, this procedure should be followed by both parties..and yield the same results.. and that if such an agreement is not in place, then any sort of result of non-homogeneous procedures should be invalidated as proof on account of not being verifiable in a commonly accepted manner.
There is a problem on both sides of this example.

The first is that evidence is rarely (I would assume) embodied in a single physical object that can be tested by anyone at their leisure. A bloodstain on a wall will be photographed, measured, sampled, and perhaps checked with some UV tests to see if there is anything else nearby. They do not cut out the section of the wall to preserve the evidence. A house that has been broken into and ransacked will not be perserved until the court date - information will be collected, photos will be taken, and whoever owns the home will be allowed to clean it up and live in it once again.

As for the alternative, if one party finds any conclusion that contradicts the other, then the entire evidence is thrown out? :smallconfused: That sounds like a way to throw out nearly any sort of evidence you could present, especially anything as subjective as witness testimony. If the only way to prosecute a criminal would be to catch them on tape with a clearly recognizable identification in the video, I'd think that it would become nearly impossible to prosecute even the most obvious of crimes. And even that has the possibility of being thrown out of court...


so..is there any point left in analyzing physical evidence at all..if all one has to do is to wait for the appeal, hire a new expert who will say something different and try to invalidate the evidence used?
This is why both sides get to hire experts, and why we use thinking humans rather than boolean computers to make the decisions. If a hired expert is trying to obfuscate the issue with confusing statistics and scientific gibberish, the opposing expert can call them on it and the jury can choose to not factor it into the decision.

Well, that is the idea anyways...

[dehro]

If a hired expert is trying to obfuscate the issue with confusing statistics and scientific gibberish, the opposing expert can call them on it and the jury can choose to not factor it into the decision.

Well, that is the idea anyways...

ha..but how is a jury going to determine which one of the two experts to believe?
either the jury is made of experts, or it's going to be just as much guesswork as not having an expert in the first place would be

I have now confused myself

[LaZodiac]

ha..but how is a jury going to determine which one of the two experts to believe?
either the jury is made of experts, or it's going to be just as much guesswork as not having an expert in the first place would be

I have now confused myself

In the end, it's based on who is move convincing. For the most part it's all politics, which is quite annoying.

Though if there is a situation were all the evidence points to you not doing it, but we also have 5 eye witnesses who are 100% in tune with eachother, THEN we have a problem that simply politics won't fix.

[Wreckingrocc]

You'd actually be surprised how little politics factor into it at all. If there isn't sufficient evidence to believe an investigation, it's thrown out, and the case is evaluated without consideration. Precedent encourages juries in cases with consistent amounts of evidence to always convict similarly.

Really, it seems like the major issue in the judicial system is the "single trail" rule. All too often, juries are left with little to no evidence on either side, and are forced to convict based on the loose information they've got. In one controversial circumstance from about a week ago here in the states, there was a fiasco about conviction without actual genetic evidence; after the trials, the majority of witnesses also revoked their claims. It didn't do much.