SomeAlibi wrote:Jackie wrote:thoughtful wrote:Jackie wrote:Hi, T. - Doctrinaire considerations aside, do you believe 95% certainty is 'too little to convict'?
I'm serious - If you HAD to put a number on 'beyond a reasonable doubt', what would it be, 96% certainty? 99.999999%?
After having studied this question in some depth and thought about it a lot, I can only answer: you cannot put a number on it.
Imagine a case where the only evidence is numerical and probabilistic. There are such cases, for example cold hits, in which a DNA sample from a crime scene of many years ago is run through a database and a match is found.
If the match is not excluded for obvious reasons - say the match is of the right age and lived in the right area at the time of the crime - then he's called a "plausible suspect".
Now, if the DNA match is perfect, it's more than a plausible suspect, you've found the man. But if the old DNA is degraded, there can be more than one person in the world who could match the crime DNA.
Say there are X people in the world who could match that DNA. You calculate the intersection of that group with people of the right age who lived in the area at the time of the crime (under the assumption that those are completely independent properties). You end up with a probability that your match is the criminal which is less than 100%; according to how big X is, the probability could be 97%, 95% etc. Now you have to ask yourself: if it's 97%, would I convict? 95%? 90%?
In reality, for most jury members, I believe, the answer is going to be...and I'm afraid that Fiona will be very upset by this...that in the absence of any other information or proof whatsoever, the jury is not going to go on the number only (unless it's on the order of 99% or more); they are going to take into account factors such as the character of the accused and his past criminal record. And I believe acquittal in the face of a 95% or 97% probability would be absolutely possible, if the man was someone whose personality impressed the jury in a very positive way, just as conviction on much lower percentages would be possible for someone who had committed other crimes similar to the crime in question. Even though, as Fiona points out, these facts have no actual bearing on the evidence at all.
It just seems to me impossible to ask a jury to either acquit or convict on the purest evidence "95%". It simply is not a quantity which can be used to define "reasonable doubt".
Were you 95% certain that you didn't overpay for your home(s)?
Are you 95% certain that the returns on your retirement portfolio investments are going to beat the S&P 500, or even just bank interest rates?
When you or your loved ones need medical care, are you 95% certain that the doctor in Q knows what s/he is doing? Are either you or the doc 95% certain that the drugs and procedures recommended will be effective?
I think you're rejecting the notion that RD is based upon reason and COMMON SENSE and have equated RD with a level of 'scientific certainty' that comes perilously close to the (unattainable) 'absolute certainty' standard.
I'd hate to have my children walk the streets of a jurisdiction that gave violent rapists and killers the benefit of your particular formulation of the RD standard, T..
Somehow I get the feeling that you've never spent any time inside criminal courts, up close and personal with the 'criminal element' of our society. Am I wrong?
I think you've misread Thoughtful's post. She says clearly you can't put a number on guilt more than once??
Read back - T. wrote:
"...stating that if guilt appears 95% certain then one should vote for conviction can be equated with saying that our system is expected to convict 5% of innocent people, which is unacceptable..."
How is this not a rejection of '95% certainty'?