What Chance Of Innocence Can Be Tolerated For Conviction?

[Chimp]

[ QUOTE ]

Maybe. I think it's a judgement call. I'm disposed toward the current unanimous system for criminal cases. But you have a point.


[/ QUOTE ]

Unanimity is not always required in the US. Many states allow majority or super-majority verdicts in civil trials. A much smaller number allow them in criminal trials. At least, this was the situation when I was in law school about 10 years ago.

As for the original question, I remember the prof. discussing this issue in my crim law class. Most students thought the acceptable probability was between 75-90%. I can't remember what I thought at the time, but I am probably closer to the high end of that range now. The consensus was that the criminal system was intentionally set up to be a PTB-style squishy definition rather than a defined probability. This contrasts with the civil verdict system, which everyone agrees requires nothing more than >.50.

I don't know if someone is actually suggesting that we ask each juror to assign his own estimated probability to his individual verdict, but I can assure you that most American jurors would not be able to grasp this concept. The real-world application of such a system would be laughably unworkable (e.g., jurors voting to convict while at the same time assigning their probability of guilt at 10%)