"Everything Hans said reinforced the image I already had of him. He wasn't interested in what was true, only in whether or not he had been treated fairly."
I think that being treated fairly is a pretty important point. It's pretty obvious now that he was guilty; but, from what I've read, at the time of the trial the prosecution didn't come close to meeting their burden.
I would prefer to have a system where everyone is treated fairly, and some guilty men go free.
Robert Nozick raised an interesting question about this problem. Any system of determining guilt has some false positive rate (let's call it X). You can trivially cut X in half by flipping a coin after determining a person's guilt and only actually considering them guilty if the coin is heads.
I've always heard that it's better to let 10 guilty men go than let one innocent man go to jail. What about 11? 15? 100? 1000? Assuming we can accurately gauge our current jury system's effectiveness (and we can at least come up with a reasonable lower bound using appeal data), it would be trivial to achieve whatever ratio we desired using a simple RNG. So, what's the magic number (1:10, etc.)? And should we do so?
Correct me if I'm wrong, but the major evidence was:
1. He had taken out the front seat of his car
2. He had hosed down the interior
3. He had his passport and several thousand dollars in cash
on him.
4. He kept removing the battery from his cellphone, after he was interrogated by police.
5. He bought a few books on homicide investigation, after his wifes homicide investigation began.
6. He lawyered up quickly.
Numbers 1-3 seem very reasonable for someone living in their car. 4-5 make sense for someone who has just been interrogated by the police. And 6 is always a good idea - for anybody - before talking to the police.
Taken with the fact that his wife was dating (and recently broke up with) a confessed serial killer I'd say we have reasonable doubt.
He was, of course, guilty - I just don't think you could reasonably conclude it from the evidence presented at the trial. I think Plato said that knowledge must be "justified true belief." I don't think the jury could have known that he was guilty at the time they convicted him, because such a belief wasn't justified.
It's kind of like a broken clock that just blinks '12:00.' By looking at that clock you can't know that it's noon, even if it incidentally is at that moment.
I just don't think you could reasonably conclude it from the evidence presented at the trial.
smanek, I'm guessing you weren't in the courtroom to observe Hans during the proceedings. Judging from what I've read I think Hans's courtroom behavior and crazy explanations led the jury to find him guilty. I'm glad he didn't take his lawyers' advice to keep his mouth shut.
FWIW, I would lawyer up the second I was accused of committing a crime, regardless of guilt. If you're innocent, saying nothing is better than saying you're innocent. You want the charges dismissed as quickly as possible.
If you're guilty, the same is true. Don't confess unless you want to go to jail!
Regarding the 'random save coin toss' question: I suspect juries would adjust, and start convicting with less certainty, if they thought there was a random chance those they found guilty would still go free.
I think that being treated fairly is a pretty important point. It's pretty obvious now that he was guilty; but, from what I've read, at the time of the trial the prosecution didn't come close to meeting their burden.
I would prefer to have a system where everyone is treated fairly, and some guilty men go free.
Robert Nozick raised an interesting question about this problem. Any system of determining guilt has some false positive rate (let's call it X). You can trivially cut X in half by flipping a coin after determining a person's guilt and only actually considering them guilty if the coin is heads.
I've always heard that it's better to let 10 guilty men go than let one innocent man go to jail. What about 11? 15? 100? 1000? Assuming we can accurately gauge our current jury system's effectiveness (and we can at least come up with a reasonable lower bound using appeal data), it would be trivial to achieve whatever ratio we desired using a simple RNG. So, what's the magic number (1:10, etc.)? And should we do so?
And yes, the reporter was being a total asshole.