Active thread

Only the replies to the one thread you selected
OP: Suppose that in a prison yard, there is clear evidence that 24 out of 25 prisoners participated in the murder of a guard. There is no other evidence regarding any of the individual prisoners. Can one prisoner of the 25, randomly chosen, be convicted of murder on this evidence?
Triton: I agree with Deimos, No
Cloud City: Because each has a 50/50 chance of being the one who did it versus the one who did not: we know that one did not.
Whirlpool: Yeah we can't convict
Loc Grim: that's not quite right probability-wise--each one has a 95% probability of being guilty
Alderaan: I agree we should not. We have knowledge that we WILL prosecute an innocent person
Athos: Wouldn't the chance be 24/25 that he is the one that did it versus did not
Whirlpool: Just like int he real world - we know some portion of the population are criminals. A fact about a group of people cannot (on its own) be used to apply to individuals within a group - It's like the composition fallacy or something
Loc Grim: but I still think there's a reasonable doubt for each one, even if that reasonable doubt is only 5%
Athos: so if we can generally use probability in other cases like civil ones why not here?
Triangulum: Guilt or conviction in criminal prosecutions needs to be about more than just probabilities. Sure, technically there is a 95% chance that any individual prisoner participated in the murder, and that maybe makes each one suspicious, but there should be an affirmative way to tie an individual to the crime before conviction. Clearly too much chance involved here
Triton: There is no specific evidence to say that this individual person did it- shouldnt be able to convict on mere probabilities. Separately, the 96% chance is less than my personal reasonable doubt standard.
Moon: Leave it up to the jury to decide if there is reasonable doubt or not but I believe it should be upheld on appeal as sufficient
Jupiter: No. Adjudicating this way, using a utilitarian calculus, seems fundamentally against the concepts of innocent before proven guilty, giving an individual a fair shake, and actually seeking out the truth. Not only is it a probabilistic fallacy (@Cloud City) but it's also morally dubious.
Alderaan: Absolutely agree with @Jupiter on why convicting in this case would be dangerous
Sateda: 50/50 is not quite right I think. Still, I was always told that it is better to set free 10 guilty people than convict one person who is innocent. What about 24 who are guilty? It does not matter. We know that 1 person is 100% innocent and that is enough for me to let them all go.
Abydos: He is innocent until proven guilty, that
Asuras: I dont think that I honestly believe that in most criminal trials there is actually less than a 5% of wrongful conviction, so this standard might actually be more favorable to the wrongfully accused
Asuras: That's true this ensures that an innocent person is convicted as opposed to creating a mere chance of it
Abydos: He is innocent until proven guilty, that's true. But that doesn't mean he can't be accused. We've certainly got probable cause. Let him bring forth evidence to prove his innocence. Let him make a convincing case he is the 1 out of 25 that did not do it.
Asuras: doesnt that shift the burden
Loc Grim: Abydos that's not how burdens of proof work
Jupiter: The question is whether he can be convicted.
Abydos: No. The burden is met by there being a 96% chance he was guilty.
Loc Grim: where are you getting 96 out of? it's 95
Loc Grim: but also isn't 5% a reasonable doubt?
Iapetus: We can't put a percentage to it being beyond a reasonable doubt. We need evidence to convict.
Oberon: @abydos if he is innocent until proven guilty, then why would we subject him to a trial where he will not likely have a good opportunity to prove his innocence. the only witnesses will be 24 other people who don't want to be convicted and prison guards who want a conviction regardless of who it is.
Oberon: I understand that probabilities are used in other civil cases but the stakes are too high in the situation. This isn't about fines and fees. It's someone's life.
Jupiter: This also elides into a problem of bias, I think. Suppose we don't literally have a prison yard problem but rather a situation with a large number of a certain type. For hypothetical's sake let's say MEN, and we know for a fact it's mostly men who commit a certain type of violent crime. This type of analysis gives credence to the probabilistic rather than the evidentiary, which maybe would tell us to just prosecute all the men, or all of the ___ group, unfairly and based on "empiricism" alone
Triton: Oberon makes a great point. Also, it is 96% friends, but I had to make sure with a calculator
Abydos: Let me remind all of you that would let the one defendant go that, unless you want to violate equal protection, all 25 defendants––24 of whom killed someone––would all need to be acquitted.
Pinwheel: The probabilistic nature of our criminal system seems like a strong argument for increased government surveillance. A security camera on every street would provide concrete evidence of guilt.
Oberon: @umbriel why would they all need to be acquitted? Interesting point but I'd love to hear more.
Upper Posada: If 96% of the people committed the crime, that isn't the same as an individual meeting the 95% burden of proof, right? Because the person you are prosecuting could be 0% guilty. Is there not a question of framing and which level you define the burden?