Continuation Part 14: Amanda Knox/Raffaele Sollecito

[Bill Williams]

Where did she post it? Some PMF forum that doesn't allow her logic to be challenged?

..... would she publish it in a manner that it can be peer reviewed?

 

[acbytesla]

I have to disagree. Those factors also go into the "pure Bayesian reasoning". I haven't read her post, but there's nothing wrong with the hypothetical reasoning you describe. One piece of evidence certainly can raise an assessed probability of guilt from 10% to 91%...and another piece of evidence can lower it down from 91% to 0.91%



This is extremely important, but, once again, it's part of the Bayesian math. See http://lesswrong.com/lw/ih/absence_of_evidence_is_evidence_of_absence/.

With all due respect, there are some things that cannot be explained post hoc with math. For eample Consider that you found a drop of milk at a precise location, you cannot with math come up with and accurate statistical model for that event. This is in fact Thouhhtful's reasoning on contamination.

 

[anglolawyer]

This Bayesian stuff, when applied to proof of criminal responsibility, brings me out in hives. The idea of the cops coming for advice on a case with only one piece of evidence which fit their suspect 99/1 is just laughable. Hey, why not use just two loci for DNA profiling? That would save a lot of time and money and would give even better odds of 250/1. Caso chiuso!

 

[Sonia]

I have commented about this elsewhere
A woman will murder a partner who has a history of abusing her - often she get life imprison and the state make seek the death penalty.
A man will murder his partner for nagging him - he might get manslaughter and 10 years in jail.

Or far less than 10. Just one egregious example:

In Ireland a man named Eamonn Lillis bashed his wife's brains in with a brick on the patio of their lovely home in Howth in late 2008. He lied to the police and tried to pretend that an intruder had done it.

Incredibly, he was convicted of manslaughter instead of murder, even though there was plenty of evidence to show that he planned the crime.

He has served 5 years, and now he will be released any day to take possession of the 1.3 EUR estate he inherited from the wife he murdered.

http://www.independent.ie/irish-new...thin-a-month-with-13m-in-pocket-31110254.html

His wife's family and his own daughter have campaigned to change the law in Ireland so a person who is convicted of killing another person cannot automatically inherit the victim's property. The law was recently changed, but it only affects future criminals. Not Eamonn Lillis.

 

[Grinder]

A pity you have such a poor memory. I extracted from Dan O. in cross examination some threads back that if the blood cells Stefanoni scraped from the blade were the size of tennis balls, the scratch would be about the size of the Titanic. IOW scale it up.

Sorry must have been studying football stats.

 

[komponisto]

+1

And in addition, it was easy for them to pigeonhole Knox in the "mean girls" category - that was something that was culturally familiar to many of them: the entitled, spoiled, reasonably-attractive brat who is a bully and a manipulator, and who is always dodging responsibility for her actions.

Yeah, that's a good description of how she was portrayed. Interestingly it is the polar opposite of what kind of person she is. I have tried to describe her. She is not shy, and she has a goofball sense of humor. She has a wonderful, infectious giggle. But she is overall a quiet, low-key, introspective person with good common sense. And she is conscientious to a fault, believe me. She does NOT want to make anyone feel bad or cause a problem. She is utterly lacking in malice, and she shrinks from conflict.

Around the time she went to Perugia, she was experimenting with the hedonistic party lifestyle, giving it a try. She described this in her book. I could see how she might not have made a great impression in her effort to whoop it up with people she didn't know well. It might have seemed forced, because it's not her nature to be garrulous with strangers. She is most comfortable in a private setting with the people she knows well, her family and her close friends. They love her dearly, because they know she's an absolute gem of a person.

Amanda Knox, my friend. She has always made me proud to be her supporter, and I feel privileged to have her as a friend.

 

[carbonjam72]

+1

 

[Charlie Wilkes]

Has thoughtful ever come on these boards? Think she might return, now that there's a final verdict? I'm curious if she has had her faith in her positions shaken at all.

She was on IIP for awhile. She ran into a couple of scientists who work in PCR labs, who understand the problem of contamination in any experiments that involve microscopic quantities of biological material.

They tried to clue her in, and she flatly rejected their expertise.

I also called her on an egregious error in her book, where she conflated the account Raffaele gave to police with a spurious account from a UK tabloid. I showed her court records that proved what she had written was false. She simply insisted she was right without attempting to explain the discrepancy.

She gives the appearance of being rational and fair-minded, but she's not. I doubt she will ever change her thinking one iota.

 

[lonepinealex]

I maintain that the woman would not pass a first year statistics course. However, that's probably not it - it is amazing how even the most reputable scientist can get locked into a confirmation bias....

..... it's why one of the bedrocks of the discipline is peer review.

It would be one thing for a professional mathematician to say what she said, it would be quite another to have colleagues agree.

Do mathematicians do peer review? I've always mentally separated it from the hard sciences.

 

[lonepinealex]

Yeah, that's a good description of how she was portrayed. Interestingly it is the polar opposite of what kind of person she is. I have tried to describe her. She is not shy, and she has a goofball sense of humor. She has a wonderful, infectious giggle. But she is overall a quiet, low-key, introspective person with good common sense. And she is conscientious to a fault, believe me. She does NOT want to make anyone feel bad or cause a problem. She is utterly lacking in malice, and she shrinks from conflict.

Around the time she went to Perugia, she was experimenting with the hedonistic party lifestyle, giving it a try. She described this in her book. I could see how she might not have made a great impression in her effort to whoop it up with people she didn't know well. It might have seemed forced, because it's not her nature to be garrulous with strangers. She is most comfortable in a private setting with the people she knows well, her family and her close friends. They love her dearly, because they know she's an absolute gem of a person.

Amanda Knox, my friend. She has always made me proud to be her supporter, and I feel privileged to have her as a friend.

That's really nice to know. Of course, it wouldn't matter if she was a heinous bitch, she'd still be innocent. But it's nice to know she isn't.

 

[carbonjam72]

She was on IIP for awhile. She ran into a couple of scientists who work in PCR labs, who understand the problem of contamination in any experiments that involve microscopic quantities of biological material.

They tried to clue her in, and she flatly rejected their expertise.

I also called her on an egregious error in her book, where she conflated the account Raffaele gave to police with a spurious account from a UK tabloid. I showed her court records that proved what she had written was false. She simply insisted she was right without attempting to explain the discrepancy.

She gives the appearance of being rational and fair-minded, but she's not. I doubt she will ever change her thinking one iota.

There's been discussion, and in fairness I don't know if its true, but that her daughter is married to an actual friend of Rudy Guede?

If so, might she have a "rooting interest" in lessening the guilt of Guede by blaming others, since her own daughter is far closer to Rudy's actual social circle than Amanda ever was?

She seems to display some sort of mental block that prevents her from thinking things through, and leads her unerringly to unfortunate conclusions.

I'm fascinated by the thought process. Mach too.

 

[Kauffer]

She was on IIP for awhile. She ran into a couple of scientists who work in PCR labs, who understand the problem of contamination in any experiments that involve microscopic quantities of biological material.

They tried to clue her in, and she flatly rejected their expertise.

I also called her on an egregious error in her book, where she conflated the account Raffaele gave to police with a spurious account from a UK tabloid. I showed her court records that proved what she had written was false. She simply insisted she was right without attempting to explain the discrepancy.

She gives the appearance of being rational and fair-minded, but she's not. I doubt she will ever change her thinking one iota.

This is odd isn't it? Someone who lives by proof at work, rejects it in this case.

Isn't there an even odder poster at PMF - Greggy, I think - says he's a scientist who has a professional understanding of forensic evidence, actually rejects the probative value of the DNA evidence against Amanda and Raffaele, yet still thinks they are guilty!? Apologies to Greggy, if I've misrepresented him.

 

[NancyS]

Yeah, that's a good description of how she was portrayed. Interestingly it is the polar opposite of what kind of person she is. I have tried to describe her. She is not shy, and she has a goofball sense of humor. She has a wonderful, infectious giggle. But she is overall a quiet, low-key, introspective person with good common sense. And she is conscientious to a fault, believe me. She does NOT want to make anyone feel bad or cause a problem. She is utterly lacking in malice, and she shrinks from conflict.

Around the time she went to Perugia, she was experimenting with the hedonistic party lifestyle, giving it a try. She described this in her book. I could see how she might not have made a great impression in her effort to whoop it up with people she didn't know well. It might have seemed forced, because it's not her nature to be garrulous with strangers. She is most comfortable in a private setting with the people she knows well, her family and her close friends. They love her dearly, because they know she's an absolute gem of a person.

Amanda Knox, my friend. She has always made me proud to be her supporter, and I feel privileged to have her as a friend.

That's interesting to know, although I would disagree that she even experimented with a hedonistic lifestyle. A few joints does not make you a hedonist - and it was hardly as if she was spending her year out in Ibiza and up dancing for days spangled on those E pills!

 

[anglolawyer]

Thoughtful/Leila gets by by just rejecting things out of hand. She can't believe anybody would ever fix any evidence, so that's it! No need to bother with all the tedium of looking at it. Brilliant!

 

[Bill Williams]

Do mathematicians do peer review? I've always mentally separated it from the hard sciences.

When Fermats last theorem was proven in 1994, the mathematician who solved the 350 year old puzzle was expected to publish his work. Which he did, and you can bet it was then peer reviewed.

Before the advent of electronic computers a guy hand calculated 100s of digits of pi. Took him years. When computers tackled this, it was proven that the last 1/3 of his digits were wrong; he'd wasted years of his life.

The mockery he received was aka peer review!

 

[acbytesla]

Yeah, that's a good description of how she was portrayed. Interestingly it is the polar opposite of what kind of person she is. I have tried to describe her. She is not shy, and she has a goofball sense of humor. She has a wonderful, infectious giggle. But she is overall a quiet, low-key, introspective person with good common sense. And she is conscientious to a fault, believe me. She does NOT want to make anyone feel bad or cause a problem. She is utterly lacking in malice, and she shrinks from conflict.

Around the time she went to Perugia, she was experimenting with the hedonistic party lifestyle, giving it a try. She described this in her book. I could see how she might not have made a great impression in her effort to whoop it up with people she didn't know well. It might have seemed forced, because it's not her nature to be garrulous with strangers. She is most comfortable in a private setting with the people she knows well, her family and her close friends. They love her dearly, because they know she's an absolute gem of a person.

Amanda Knox, my friend. She has always made me proud to be her supporter, and I feel privileged to have her as a friend.

I am really glad to hear that about her. I have never imagined her to be anything but a nice person. There really has never been any real credible reports of her being anything else. I think it telling how close she became to the priest at the prison despite being an atheist.

 

[LondonJohn]

I have to disagree. Those factors also go into the "pure Bayesian reasoning". I haven't read her post, but there's nothing wrong with the hypothetical reasoning you describe. One piece of evidence certainly can raise an assessed probability of guilt from 10% to 91%...and another piece of evidence can lower it down from 91% to 0.91%



This is extremely important, but, once again, it's part of the Bayesian math. See http://lesswrong.com/lw/ih/absence_of_evidence_is_evidence_of_absence/.

The problem is this: it's impossible to built a perfect Bayesian model that matches reality. What you get in practice - and it's perfectly illustrated in Thoughtful's example - is an incomplete model, which gives incorrect outputs.

So this particular piece of evidence - the "1 in 100" hand print that matches the suspect.

Do you think it makes any qualitative or quantitative difference to the value of this evidence whether the murder took place on a) a remote island of 11 adult inhabitants (one of whom was the victim), or b) a dense city of 5 million inhabitants?

To take the example (a), suppose that there was zero evidence against the suspect before the hand print evidence emerged. So it's entirely reasonable to suppose that a juror's a priori assessment could very well be 10% (one in ten people on the island). And suppose that in the latter example, (b), a juror for some reason assigned a 10% a priori probability to guilt.

Is it reasonable to suggest that the discovery of a hand print that matched 1 in 100 people, and which also matched the suspect, would have the same statistical impact on the mathematical probability of guilt in example (a) as in example (b)?

I say no. I say that in example (a), the hand print evidence might well raise the probability of guilt from 10% to something above 91%. And in example (b), it would raise the probability of guilt to something well, well below 91%. To be slightly simplistic in the assessment, there's actually a statistical unlikelihood that someone with this hand print existed among the 10 remaining people on the island, so the fact that it matched the suspect would be highly significant. Whereas in a city of 5 million people, there would (statistically) be 50,000 people with that hand print.

It seems to me that factors such as this simply aren't addressed by this type of use of Bayesian theory. I take your point that it should be in theory possible to incorporate all of this (and much more besides) into a Bayesian model. But the problem is that it would end up being a totally unwieldy and unworkable model, because there would have to be so many nested boolean conditionals. So what you end up with is a simplistic single-variable (or, at most, double-variable) example such as the one given by Thoughtful - and that simplistic example gives a wrong and misleading output.

 

[Kauffer]

Yeah, that's a good description of how she was portrayed. Interestingly it is the polar opposite of what kind of person she is. I have tried to describe her. She is not shy, and she has a goofball sense of humor. She has a wonderful, infectious giggle. But she is overall a quiet, low-key, introspective person with good common sense. And she is conscientious to a fault, believe me. She does NOT want to make anyone feel bad or cause a problem. She is utterly lacking in malice, and she shrinks from conflict.

Around the time she went to Perugia, she was experimenting with the hedonistic party lifestyle, giving it a try. She described this in her book. I could see how she might not have made a great impression in her effort to whoop it up with people she didn't know well. It might have seemed forced, because it's not her nature to be garrulous with strangers. She is most comfortable in a private setting with the people she knows well, her family and her close friends. They love her dearly, because they know she's an absolute gem of a person.

Amanda Knox, my friend. She has always made me proud to be her supporter, and I feel privileged to have her as a friend.

Everyone, aged 20 tries on some new clothes to see how they fit - there's no problem with that.

It sounds like most of us are more angry about what has happened than she is capable of feeling. I would really like Mignini to go to prison for a time. I very rarely feel like that about anyone. I mostly think that people should not be there. But, I suspect he has ruined many lives.

 

[LondonJohn]

This Bayesian stuff, when applied to proof of criminal responsibility, brings me out in hives. The idea of the cops coming for advice on a case with only one piece of evidence which fit their suspect 99/1 is just laughable. Hey, why not use just two loci for DNA profiling? That would save a lot of time and money and would give even better odds of 250/1. Caso chiuso!

And that cuts to the heart of the problem with Thoughtful's use of Bayesian analysis. You know full well that it's preposterous to suggest that this one item of evidence which is shared by 1 in 100 of the population could, singlehandedly, raise a juror's probability of guilt from 10% to 91% (or 50% to 99%). It simply couldn't happen in the real world.

And Thoughtful is not wrong because her maths are wrong. Her maths are right. But her model is wrong. Her model is wrong because it does not - and cannot - model the extremely complex set of variables and nested conditional probabilities that are, in effect, used by actual people to calculate their assessment of guilt. If such a complex model replicating the real world could be built, it could only live inside a computer and would be essentially impossible to explain to the average person. People make these complex multi-variable assessments in their heads without even realising how they're doing it.

 

[carbonjam72]

And that cuts to the heart of the problem with Thoughtful's use of Bayesian analysis. You know full well that it's preposterous to suggest that this one item of evidence which is shared by 1 in 100 of the population could, singlehandedly, raise a juror's probability of guilt from 10% to 91% (or 50% to 99%). It simply couldn't happen in the real world.

And Thoughtful is not wrong because her maths are wrong. Her maths are right. But her model is wrong. Her model is wrong because it does not - and cannot - model the extremely complex set of variables and nested conditional probabilities that are, in effect, used by actual people to calculate their assessment of guilt. If such a complex model replicating the real world could be built, it could only live inside a computer and would be essentially impossible to explain to the average person. People make these complex multi-variable assessments in their heads without even realising how they're doing it.

Can anyone actually explain her argument for guilt?