In the long run, the more dangerous criminal is the government itself if unchecked. Yes, it is frustrating to have criminals go free because police or prosecutor fucked up -- but government holds both the power of law and the power of (sanctioned) violence. If even they cannot be expected to follow their own law, why are they expecting others to follow it?
Your hypothetical is clear: the defendant's counsel would ask the judge to throw out the case, and the judge would comply. To make it a bit less hypothetical: what if the compelling piece of evidence was obtained through torturing a witness?
(edit: also, a "dangerous serial killer" is not so until a judge has convicted him/her. You are begging the question by presuming guilt)
> what if the compelling piece of evidence was obtained through torturing a witness?
The standard reply is that evidence obtained from torture can't be compelling because it is by definition unreliable. I've never been happy with that, since some claims can easily be independently verified. (For example, if I wanted credentials for a bank account belonging to Bill Gates, and I tortured him for those credentials, I wouldn't be too worried over the fact that someone under torture will be willing to say anything to stop it, and thus I can never really know if he was telling the truth about his credentials -- I can just try to take money, and if it doesn't work I'll keep torturing him.)
Once we stipulate that the compelling evidence obtained from torture is known to be true, I think you'll find that a lot of people still consider this a tough question.
> edit: also, a "dangerous serial killer" is not so until a judge has convicted him/her. You are begging the question by presuming guilt
No, a dangerous serial killer is one regardless of conviction. It's fair game to have whatever facts you want in a thought experiment.
> The standard reply is that evidence obtained from torture can't be compelling because it is by definition unreliable
You're conflating the reason evidence from torture is inadmissible with one of the arguments against torture.
Evidence from torture can't be admitted because we won't condone torture, ever (or, that's the theory, the real world is more, erhm, nuanced). The fact that intelligence from torture is unreliable is just one of a suite of arguments against torturing people in the first place, not a universal fact which is true for any information obtained through torture.
> since some claims can easily be independently verified
That's basically the Jack Bauer school of torture: You have a bad guy, and you know for a fact that he knows where the bomb is, and if/when he tells you, you can confirm immediately. If that situation occurred in the real world, I would assume that most people would bite their tongues and accept torture. But there is no evidence that it does. It's "What was your role in planning the attack? Who is the person in this picture? Who knows where Osama is hiding?" [source: Zero Dark Thirty] -- Sure, eventually, you can sort-of confirm these things, but torture really isn't effective if you have months-long feedback cycles.
>> The standard reply is that evidence obtained from torture can't be compelling because it is by definition unreliable
> You're conflating the reason evidence from torture is inadmissible with one of the arguments against torture.
No, I'm complaining that in much of what I read on this topic, the argument I stated is presented in the terms I stated it in: no evidence given under torture can ever be trusted, because people under torture will say whatever you tell them to say. I find that aggravating, because it is clearly wrong.
It's misleading to say it's one of a suite of arguments presented against torturing people in the first place; it is one of two, the other being that we shouldn't torture people because it is invariably morally wrong.
I don't think it's a coincidence that arguments that we should never torture people no matter what rely so heavily on different premises of invariability. But while the moral-incorrectness invariability claim can be defended, the untrustworthiness invariability claim can't be.
> a dangerous serial killer is one regardless of conviction.
But how do you know whether the person presented in the dock is a dangerous serial killer?
> It's fair game to have whatever facts you want in a thought experiment.
The whole point of a trial is to establish facts. In reality, we cannot ever guarantee things like this; we can only ever say how convincing the evidence is for a particular history of events.
TBH, if we want an example, OJ Simpson is good: he was eventually found to have wrongfully killed his wife in civil court, but at his initial murder trial he was acquitted due to (among other things) poor evidence handling by the LAPD.
There were widespread accusations of attempts to 'frame' OJ and of police racism. It was both right and necessary to discard the questionable evidence, otherwise a large part of the public would have believed that he was wrongly convicted. Evidence is not just for the courtroom, it's to rule out allegations of improper conviction and unfair treatment by the general public.
Keep in mind that the standard of proof in a civil trial is a lot lower. Being found liable in a civil trial doesn't mean he should have been found guilty at the criminal trial.
I agree that OJ probably did it based on what's known, I'm just nitpicking that one bit.
I'm saying that a more specific thought experiment, one where there is someone in custody, with clear (but tainted) video evidence of them committing multiple murders is in line with the more general thought experiment proposed upthread. So the various lawyers, the judges, the cops, they all have knowledge of the clear video evidence and the question is whether it can be used at trial or not.
So you could have a whole bunch of people really comfortable calling the suspect a murderer (because they've seen the evidence) and no path to a conviction.
I believe the problem there is that you can't define what "clear evidence" is, until the trial, when it's already been collected.
This gives policemen some really bad incentives to follow through on. Currently, the penalty for obtaining evidence illegally(or suppressing exculpatory evidence) is a slap on the wrist, so LEOs would have an incentive to collect as much evidence as possible, ignoring due process violations, because there's always a chance that evidence is ruled "clear evidence" at trial.
You would have to institute severe penalties for police officers, and then you are likely to end up with a law that has little effect in practice, except for corruption - LEOs will shy away from using it, except in cases where they know they can get away with it(bought prosecutor/judge, friendly jury, etc), or, admittedly, in cases where they really believe they're catching The Bad Guy(tm). The problem with this kind of law is that it has little effect for the majority of people(except for the headline that we caught one more Bad Guy(tm) per year using it), but it usually ends up as a tool to oppress minorities or those with deviant political and social views, as those are the ones a jury is most likely to be biased against.
Consider also that most criminal cases actually end with a plea bargain pre-trial. There is no jury to argue in front of in most of them, so all you'd have in front of you is evidence that the prosecutor is claiming to be "clear evidence"(still unproven), and your overworked PD telling you that the likelihood of the prosecutor not getting a conviction is in the single digits. You would have to allow judges to rule on whether something is "possibly clear evidence", but judges naturally favor the prosecution(most judges are ex-prosecutors, and they are usually on friendly terms with prosecution/police officers).
Due process is already trampled daily, without this "clear evidence" law that you propose. Police officers fabricate justifications for stops and searches out of thin air, prosecutors violate Brady by hiding exculpatory evidence(so much so that Chief Judge Kozinski highlighted it in a dissent - http://www.abajournal.com/news/article/epidemic_of_brady_vio...), trials take a stupidly long amount of time, PDs are exhausted, overworked and underpaid, etc. Your theoretical addition to it won't really make a significant impact - there's a million other ways for prosecutors to get a conviction, this will just be one extra tool in the toolbox. It won't be making the situation much worse, but it'd still be worse.
> No, a dangerous serial killer is one regardless of conviction.
True enough, but the point he raises is at what point do you justify illegal actions.
In your example, torture may or not bear fruit, depending on if your suppositions are correct.
The point is generally that in order for extreme measures to be taken, extreme justification must be given. And in general the justifications needed for torture would be so high that you would probably be able to convict without it, i.e. You can only torture someone if you knew beyond any doubt they dunnit.
The point of restricting the collection of evidence is to ensure that law enforcement go through a process with justification at each stage. This is to protect innocent/law abiding from law enforcement itself, as a lot of the acts they go through in the course of their business would be illegal if they were not law enforcement.
> And in general the justifications needed for torture would be so high that you would probably be able to convict without it
It's so high that IMHO the only possible place for torture in a civilized legal system is as punishment for torture by someone acting in a position of authority; if you're at the point where the moral calculus favors torture, you give absolutely zero fucks about what happens to yourself afterwards.
Your hypothetical is clear: the defendant's counsel would ask the judge to throw out the case, and the judge would comply. To make it a bit less hypothetical: what if the compelling piece of evidence was obtained through torturing a witness?
(edit: also, a "dangerous serial killer" is not so until a judge has convicted him/her. You are begging the question by presuming guilt)