Continuation Part 2 - Discussion of the Amanda Knox case

[lionking]

I would caution against trying to apply percentages to determine reasonable doubt. A juror could find a piece of evidence proposed by the prosecution lacking and yet vote guilty based on the totality of the other evidence presented.

As I noted above, the questions of jury unanimity and reasonable doubt I addressed apply only to U.S. law. I do not know how they apply to Italian law.

I do agree with this. I only raised it because one poster said he would find AK guilty if he was only 90% convinced of her guilt. I challanged this.

 

[komponisto]

lionking, as I asked before, I would very much like to know what rate of wrongful convictions you would accept. For me it is between 1 in 100 and 1 in 1000; do you think that is an unreasonably high standard?

 

[RoseMontague]

I do agree with this. I only raised it because one poster said he would find AK guilty if he was only 90% convinced of her guilt. I challanged this.

I don't see any definitions putting a percentage on what is beyond a reasonable doubt. So if 90% works for you, so be it. I still would like to hear from you what the 90% is of the prosecution case that you are certain about and I am interested in also hearing the 10% you have doubt on.

 

[lionking]

You either have no understanding of probabilty theory or you are willing to accept that 10% of people convicted of murder are actually innocent. If you were wrongly accussed of murder would you be content that there was a 10% chance that you would be wrongly convicted?

Please show how 90% certainty of guilt translates to 10% chance of wrong conviction? They are two different things entirely, which was why I described this claim as a strawman. I think it is you who doesn't understand probability.

 

[LondonJohn]

The standard is beyond reasonable doubt, not beyond absolute doubt. 10% doubt is reasonable in my opinion.

I'm afraid you may not know what the word "reasonable" means in the term "beyond a reasonable doubt". It doesn't mean the same as it does in this sentence: "even though it was raining, there were still a reasonable number of people walking without umbrellas".

In fact, "beyond a reasonable doubt" could perhaps most accurately be defined as "beyond all doubt which has its basis in reason", or, more lengthily but probably more accurately still: "beyond all doubt which a reasonable person could imagine".

An example: imagine I were on trial for murdering a man with a gun in a pub fight. Imagine five separate, impartial (and sober) people saw me fire the gun, and all five identified me in an ID parade. And my fingerprints were all over the gun which was found discarded outside the pub. Now, just on that evidence, a jury should find me guilty beyond a reasonable doubt. But imagine that one juror said: "Ah yes, but what if all five of the witnesses hadn't actually seen the defendant, but they all happened to pick him out by sheer coincidence in the ID parade? And what if, in addition, the real murderer had managed to press the gun into the defendant's hand in the melee after the shooting, thereby placing the defendant's fingerprints onto the gun? Both of these things are technically possible, so I have doubts as to the defendant's guilt!"

And this is where "reasonable" doubt comes into play. In this instance, it should be apparent to anyone that the doubts expressed by this juror were not "reasonable" doubts - or doubts that a reasonable person could be expected to hold.

So, actually, "beyond a reasonable doubt" really does mean 100% certainty - to the exclusion of ALL doubts which could be held up as reasonable doubts. And incidentally, it doesn't mean that the jurors should apply this standard to each and every piece of the case separately, and indeed it's entirely possible for the jurors to reject large chunks of the prosecution's evidence and still convict beyond a reasonable doubt. But, in that case, it's necessary that the pieces of evidence against the defendant which they DO accept should add up to a total belief in the defendant's guilt in order for them to return a guilty verdict.

Again, to use my earlier example, suppose that there were those two pieces of evidence against me in the murder trial (the five eye witnesses, and my prints on the gun). Suppose that in addition to this, the prosecution introduced evidence that I had told an acquaintance of mine, earlier on the day of the murder, that "I'm going to kill that guy today - I hate him so much". Now, the jury perhaps might choose not to accept this piece of evidence (maybe this acquaintance had a grudge against me, or couldn't be certain of what I'd said) - but the first two pieces of evidence would still constitute a good enough case to return a guilty verdict beyond a reasonable doubt.

 

[bucketoftea]

Welcome to the discussion hariseldonsr. I am not familiar with any news stories that relate this particular event in the early days of this case. It is possible that I may have missed it. Regardless, my opinion is that Rudy killed Meredith and he acted alone in doing so.

Hi, hariseldonsr, Mary

Yes there was such an early story, He was actually shouting down a telephone "I'll kill her" or some such, but in any case ruled out as related to Meredith's murder in any way. Early news stories are always part inadvertent gobbledygook.

 

[RoseMontague]

Not at all. Washing with water and friction you could remove all the organic components and diminish the Fe to the point that it's indistinguishable from the ferrous content of the blade. All that scouring might in fact be what forced particulate (not 4 random cells, but a tiny piece of tissue; made up of 4 or more cells) matter into a deformity on the blade's surface.

Does that help?

Not really. If I had a bloody knife, I would use more than water and friction to clean it before I used it to slice my breakfast toast the next morning.

 

[komponisto]

Please show how 90% certainty of guilt translates to 10% chance of wrong conviction?

If you believe 90% chance of guilt, then you must also believe 10% chance of innocence. It has to add up to 100%.

 

[LondonJohn]

Not at all. Washing with water and friction you could remove all the organic components and diminish the Fe to the point that it's indistinguishable from the ferrous content of the blade. All that scouring might in fact be what forced particulate (not 4 random cells, but a tiny piece of tissue; made up of 4 or more cells) matter into a deformity on the blade's surface.

Does that help?

Did I really just see someone write "....diminish the Fe to the point that it's indistinguishable from the ferrous content of the blade"?

Does this person actually believe that a ferrous (Fe+2) cation (or the oxidised ferric (Fe+3) cation in an iron compound) is the same thing as the fixed, crystalline structure of the iron molecules in steel? Does this person therefore believe that spraying Luminol onto a steel knife blade would make it light up like a Christmas tree?

Oh dear.

 

[lionking]

If you believe 90% chance of guilt, then you must also believe 10% chance of innocence. It has to add up to 100%.

Wrong. 10% doubt does not equate to 10% chance of innocence.

 

[LondonJohn]

Wrong. 10% doubt does not equate to 10% chance of innocence.

See my previous-but-one post............

 

[bucketoftea]

Not really. If I had a bloody knife, I would use more than water and friction to clean it before I used it to slice my breakfast toast the next morning.

Quite so; so would I, and soak it in a solution of so-called "bio" laundry detergent or meat tenderizer @ 40, followed by a 30% hypochlorite solution.

Actually, if I had a tainted knife eg I had seen my child using it to cut up dog poop, I'd chuck it.

 

[RoseMontague]

Wrong. 10% doubt does not equate to 10% chance of innocence.

Yes, I made an earlier point on this one. What I what like to know is what you consider the certain parts of the prosecution's case that add up to 90% to be?

 

[Danceme]

Do you always go through such contortions when trying to figure out simple, straightforward events?

Do you have a point or are you once again just being rude? Somehow I think your intent was definitely the latter.

 

[Chris C]

Prediction: your fellow jurors will balk if you try to argue that there is indeed clear and convincing evidence against the defendant, but that they should be acquitted anyway.

What in this case that the prosecution put forth is clear and convincing?

 

[Chris C]

In Doug Preston’s book “The Monster of Florence” which he co-authored with Mario Spezi, he mentions in the afterword that the night of the murder (November 1st/2nd 2007) a male individual was seen by several witnesses washing himself of blood in a public fountain and screaming “I killed her…I killed her” over and over. Police and an ambulance were called, statements were taken, one article appeared the next day or the day after in the local newspaper. The person was held in a drug rehab hospital for apparently several weeks during which the Knox ‘story’ was developing. Then, nothing – nothing ever appeared about it again. I can’t believe Preston and Spezi just dreamed this up, so what happed…?

The defense and prosecution both investigated this and both sides to the best of my knowledge decided it was unrelated to Meredith's murder.

 

[komponisto]

Wrong. 10% doubt does not equate to 10% chance of innocence.

If it doesn't, then that means you're not well-calibrated. See here, particularly the following passage:

You should only assign a calibrated confidence of 98% if you're confident enough that you think you could answer a hundred similar questions, of equal difficulty, one after the other, each independent from the others, and be wrong, on average, about twice. We'll keep track of how often you're right, over time, and if it turns out that when you say "90% sure" you're right about 7 times out of 10, then we'll say you're poorly calibrated.

Remember Spock from Star Trek? Spock often says something along the lines of, "Captain, if you steer the Enterprise directly into a black hole, our probability of survival is only 2.837%." Yet nine times out of ten the Enterprise is not destroyed. What kind of tragic fool gives a figure with four significant digits of precision that is wrong by two orders of magnitude?

The people who write this stuff have no idea what scientists mean by "probability". They suppose that a probability of 99.9% is something like feeling really sure. They suppose that Spock's statement expresses the challenge of successfully steering the Enterprise through a black hole, like a video game rated five stars for difficulty. What we mean by "probability" is that if you utter the words "two percent probability" on fifty independent occasions, it better not happen more than once.

If you say "98% probable" a thousand times, and you are surprised only five times, we still ding you for poor calibration. You're allocating too much probability mass to the possibility that you're wrong. You should say "99.5% probable" to maximize your score. The scoring rule rewards accurate calibration, encouraging neither humility nor arrogance.

 

[komponisto]

What in this case that the prosecution put forth is clear and convincing?

Nothing! That's why from my point of view this discussion is somewhat tangential: I don't argue that the prosecution merely "failed to prove AK and RS are guilty beyond a reasonable doubt"; I believe they're innocent beyond a reasonable doubt!

 

[Chris C]

Not at all. Washing with water and friction you could remove all the organic components and diminish the Fe to the point that it's indistinguishable from the ferrous content of the blade. All that scouring might in fact be what forced particulate (not 4 random cells, but a tiny piece of tissue; made up of 4 or more cells) matter into a deformity on the blade's surface.

Does that help?

Seriously? Blood isn't that easy to wash off. It sticks. It would stick in every deformity on the blades surface. It would get in the smallest crack on the handle of the blade and would be unable to wash out without the use of chemicals that would destroy dna. There is a reason they take knives apart to check for blood. Though it seems this is one thing not done in the knox/sollecito case.

 

[lionking]

If it doesn't, then that means you're not well-calibrated. See here, particularly the following passage:

It's not about calibration, it's about two different issues. Doubt of guilt is not the same as certainty of innocence. And another slur is noted.