Continuation Part Eight: Discussion of the Amanda Knox/Raffaele Sollecito case

[Mary_H]

You are aware (I hope) that platonov knew the answer to this question before he wrote the question..............

I figured, but thought some other people might like to know the answer to the question.

(And FWIW, Nencini still faces significant disciplinary action related to his press activities and statements following the verdict).

That I did not know. Thanks.

 

[Kaosium]

Well....... you go away for a bit and you come back to find the "stomach/duodenum contents vs ToD" debate rearing its (not so) ugly head again!

And skimming through the last few pages, I subscribe almost entirely to your opinion on this issue. There's sufficient experimental data - when properly interpreted and placed in the correct context - to indicate convincingly that Meredith Kercher a) very probably died between 9pm and 9.30pm, b) almost certainly died between 9pm and 10pm, and c) without a shadow of a doubt did not die later than 10.30pm.

There is absolutely no question whatsoever that the Massei (and Mignini) ToD of 11.30+ is impossible. And it's hugely clear (from my perspective) that the defence teams dropped the ball significantly on this issue. Had they - in the Massei trial - got a competent and authoritative forensic pathologist and/or gastroenterologist to explain this issue properly, there never would have been any wiggle room for Mignini or Massei to go for an 11.30+ ToD. This in turn would have forced them to consider a pre-10.15 ToD (on account of the presence of the broken down car). And in turn that would have forced them to reconcile all the other evidence (or "evidence") with a pre-10.15 ToD.

I realise that the Nencini court re-convicted on a much earlier ToD (which in itself raises fundamental issues to me of how much appeal courts should be able/allowed to alter the inferred facts). But I think the main damage was done in the Massei court. It's also fair to say that the stomach/duodenum contents cannot - even in our special case - categorically rule in or out ToDs in the way that other physical/witness evidence might do. But it can provide compelling and important evidence - particularly in these specific and unusual circumstances - that the murder very likely (90%+) took place between 9.00pm and 9.30pm, almost certainly (99%+) took place between 9.00pm and 10.00pm, and definitely (100%) did not take place any later than 10.30pm.

Thanks, LJ. I'm also virtually certain that where you disagreed you are right and I probably agree but put it wrong. I took this stuff long ago and I forget the jargon and/or like to use words people know better. I also had some cringeworthy brainfarts, like on what would happen with the relative difference with a skewed curve in one example.

 

[Mary_H]

Really ... that is a surprise !

Does this mean that the predictions from the same quarters about the the ECHR ruling [ the dissolution of the Italian republic and Amanda being crowned Queen ] may also be off ??

Someday someone should count up all the unfulfilled predictions Peter Quennell has made. They would fill a book.

 

[Desert Fox]

The previous scenarios that have been suggested, including sexual assault, are well within the limits, or rules, of human behavior. The fact is, Rudy cut Meredith's throat and then did not immediately leave the room; he hung around and did other things. Dan wants to say that Rudy attacked and stabbed without a component of sexual arousal, but the evidence suggests otherwise.

I agree that is pretty unlikely.

 

[platonov]

Known knowns

Well....... you go away for a bit and you come back to find the "stomach/duodenum contents vs ToD" debate rearing its (not so) ugly head again!

<snip>

.

Yea, the early and precise ToD is back (again)....The old ones are the best they say.
Would you mind explaining to the class your 25% error bars that were mentioned upthread.
That should be both informative and entertaining.



Or should I post the link ......
[does platonov know the answer to this Q also ? ]

 

[Grinder]

It does, but nevermind it's not relevant.

What possible difference would it make to Introna that Lalli said 2-3 hours instead of 2-4 hours? Try being specific.

It doesn't say they didn't finish the movie, there's no reason to think they didn't, though I allow they probably skipped the credits as most people do.

It doesn't say they did finish it After we ate the apple crumble with ice cream, then We did chat and then to the nine,
have gone because it was supposed to be just a meeting place for the
the afternoon​

If they watched to the end a more normal way to express it would be to say they watched the movie and went home rather explaining why they left just before 9.

The timing of everything except Meredith's arrival home is vague.

Yes, the dinner was probably before 7:00, that's what I was trying to show the movie and their account suggested.

Using the run time of the movie doesn't add anything because they stopped on more than one occasion for eating, at other times to chat and it doesn't appear they watched to the end. We also don't know if Meredith ate at the same time. Somewhere I saw a quote of one of the girls saying Meredith didn't feel well and ate lightly. I thought it was posted here. Nope it was at the link you gave me - Sophie Purton said that Meredith ate only part of her pizza.

Oh my! To be clear I'm not saying this is true as I can't find the source because it comes from CD.

"we prepared the pizza, we have eaten, then
We looked at the pictures of Halloween to your computer and then
We started to watch the movie"

Guess it was the royal we.

Well, we know they all do soon enough. This would be what we're arguing, and I'm saying the fact so few make it to three or more hours is a huge indication that last half hour weeds them out, which is something that is also suggested by the curve and the fact we know it must hit zero soon enough. This was the point I was trying to make with the actuarial tables analogy, making it to ninety is a lot different than making it to 110. That last twenty years weeds almost all of them out. This is a curve that has to hit zero, not one that can extend forever without doing so.

Yes at some point everyone will have GE. The study you sent me to at Chris' says the longest time in that study was 200 min which puts a 6:30 meal GE at 9:50 BUT that was for a tiny meal without the fat content of the pizza meal.

These workers described their test meal: “After addition of 50 mL of low-fat milk, the egg was scrambled and fried in a pan. The solid test meal was completed by a piece of brown bread (50 g) and butter (20 g).” They showed that t(lag) for a solid meal did not follow a normal distribution. The median time was 82 minutes, with the 25% percentile at 66 min. and the 75% percentile at 102 min. Out of 82 subjects (Figure 1D), the longest value was 200 minutes, and the next longest was 170 minutes (each value corresponded to a single individual).​

One indication might be to look at the comparable tests and see of those who made it to 150 minutes how many made it to 200.

Well that's kind of what I've been asking for. But I'd like to know what the results would be for a normal meal more like a pizza and apple pie. Also I'd like the numbers on 180 and 150 minutes without GE as clearly the shorter time would have more additional minutes for the remaining subjects.

Was that the test with egg whites, no milk, margarine instead of butter on the toast and nothing about it being fried? They bake the food for some of these tests and since everything else you describe indicates they want it to go through the subject like grease through a goose, I rather imagine that's the case with this one.

They want the test to be shorter in time for obvious reasons. Don't remember about the margarine.

I imagine it will be part of the appeal, just like the last one, however appealing a 10:00 ToD on this basis will be more difficult than a 11:45 ToD like last time.

That's for sure

That there's data missing from the computer is not Raffaele's fault, it's that of those whom wiped the last access data from the other programs he was using that night late on the sixth after they collected his computer (along with the knife and comic book etc) when he was in prison. I suspect that might have been inadvertent, by Raffaele's own account of his travails the night of the 5th he told them to check his computer for what he was doing, and somebody obviously did--perhaps to see if he had those programs installed. Therefore the last access data was changed to late on the sixth when the computer was in police hands and Raffaele was in solitary.

I know they screwed the pooch but if he had been online there would have been a record. It's just too bad there wasn't, but I'm sure there would be another scenario

 

[LondonJohn]

Yea, the early and precise ToD is back (again)....The old ones are the best they say.
Would you mind explaining to the class your 25% error bars that were mentioned upthread.
That should be both informative and entertaining.



Or should I post the link ......
[does platonov know the answer to this Q also ? ]

Explain to me what you mean, and I will see if I can help you out.

 

[Grinder]

The only thing wrong with the research posted is comparing the ones with easily digested substances with the ones most people eat. No one is arguing this outside Grinder, the bunnies even gave up on this one long ago, the prosecution never dared. It's like comparing the time it starts digesting a marshmallow and a marble, exaggerating for effect of course. Throw out all the ones used to attempt to detect GI maladies, (~30 min T-Lag medians) they want it to go down easy, it makes it easier on everyone doing those tests. The ones with more fibrous foods, sugar, dairy, starch, alcohol etc are the ones normal people eat and comparable to what Meredith ate that night, the ones with liquids and only egg whites with white bread toast and margarine will only lead you to the wrong conclusion (there must be something wrong with the data).

The tests with smaller easier to digest foods would have a quicker GE. The longest times would also be greater for bigger more difficult to digest food.

Yes, there they allow a ToD of up to 10:00 based upon the meal being completed by 6:30 to 7:00. It's when the meal was started that actually starts the clock and there's no evidence given by anyone that the meal started sooner than 6:30. They're also arguing against a 11:45 ToD and expect the court to come up with a scenario more reasonable than Raffaele and Amanda starting Naruto, running out the door and murdering within moments of the arrival. You'll see they still contend it's an alibi as they know the court has to account for the fiction they met up with Rudy, let him in and they have to allow time for it to develop and Rudy to have his juice and take the dump.

Two of the girls had the meal starting earlier but hey if it doesn't fit your theory...

I keep saying we don't have the "facts" needed for this to provide an alibi. As reported above one of the girls said Meredith was under the weather and didn't eat as much etc.

 

[CoulsdonUK]

You are aware (I hope) that platonov knew the answer to this question before he wrote the question..............


(And FWIW, Nencini still faces significant disciplinary action related to his press activities and statements following the verdict).

I am certain the Florence motivations will be published and the defence teams will appeal based on its contents as they really don’t have a choice.

 

[Grinder]

Explain to me what you mean, and I will see if I can help you out.

Kaosium used an old post of yours where you estimated probabilities of TOD based on a bell curve of GE. At the time you were arguing that the 11:30 TOD was impossible which btw I agree with as stated many times.

You had a 5% chance of TOD being at 9:20.

The current discussion is different because some here are claiming that the Naruto computer interaction at 9:26 provides proof positive that kids couldn't be involved.

It has come to light that the time of eating is not certain. Not only when the girls brought out the pizza but when Meredith started eating. Combining that with the vagaries of GE I don't believe a proof of innocence results. While it is clear that a defendant need not prove innocence if one claims this does it I disagree.

The ILE did not prove guilt beyond a reasonable doubt but the digestion scenario doesn't prove innocence for many reasons.

 

[Dan O.]

You may feel that your theory is still on the table but it's not on anyone's table that has the ability to think clearly.

My theory is 100% alive and well. Why can you not even advance a rough draft of what you think happened that night? Make a counter argument to mine so we can argue about the differences. Are you afraid because any statement you might advance could be proven wrong? Are you afraid that others will attack you as you are attacking me? Being wrong is the way progress is made. You put up theories and find out which ones don't hold up then you use what you have learned and construct a new theory. If you do this honestly, each iteration brings you closer to understanding the truth.

So far, Randy has done a much better job of questioning my position. He hit on a couple of key points that I had to review. I wish more of the followers of this case were able to make such constructive critical analysis.

 

[Mary_H]

I agree that is pretty unlikely.

You know, I think Dan actually could make a case that, without the testing of the semen, there is no rock solid evidence that Rudy did assault Meredith sexually. It is doubtful he touched her genitals accidentally, but it's possible. To me, though, there would be no reason for him to move her or position her without a sexual motive. He would have just killed her and left.

When Luca Lalli first examined the body, he said he could not be certain of rape. It's funny how it went from that point to being all about rape in Mignini's mind.

 

[LondonJohn]

Kaosium used an old post of yours where you estimated probabilities of TOD based on a bell curve of GE. At the time you were arguing that the 11:30 TOD was impossible which btw I agree with as stated many times.

You had a 5% chance of TOD being at 9:20.

The current discussion is different because some here are claiming that the Naruto computer interaction at 9:26 provides proof positive that kids couldn't be involved.

It has come to light that the time of eating is not certain. Not only when the girls brought out the pizza but when Meredith started eating. Combining that with the vagaries of GE I don't believe a proof of innocence results. While it is clear that a defendant need not prove innocence if one claims this does it I disagree.

The ILE did not prove guilt beyond a reasonable doubt but the digestion scenario doesn't prove innocence for many reasons.

Ah, but for our purposes (and the court's purposes), the statistical analysis that matters is Bayesian conditional analysis.

That's to say, we must pose the following question: given that we know that the victim was still alive at 9.00pm, what are the probabilities of her dying at 9-9,30, 9.30-10, 10-10.30, 10.30+?

So even though it's statistically unlikely that Meredith died later than 9pm - the important thing is that we know this to be the case. The correct evaluation in the light of this conditional knowledge is to look at the shape of the probability distribution curve to the right of the time we know to be the effective "start" time on death (corresponding to 9pm in our case). If the "curve" were a horizontal flat line beyond this point, then effectively one could say that there's an equal probability of death occurring at any time later than 9pm (regardless of whether death later than 9pm is improbable in itself).

But the curve is not a horizontal flat line in this case. It's the end of a bell curve that slopes down to zero by around 10.30pm. And this effectively means that given we know that Meredith was still alive at 9pm, it's possible to state that it's still far more probable that she died between 9 and 9.30 than between 9.30 and 10, and that it's impossible that she died later than 10.30.


In fact, we actually have nested conditional probabilities to look at in this case, since (as you point out) there is also some ambiguity as to when the meal started. But that's OK. What we can do to deal with this ambiguity is run a series of analyses based on various different meal start times within the reasonable range.

So we could do:

Analysis 1: If the meal started at 5.30pm, then given that the victim was still alive at 9.00pm, what are the probabilities of her dying at 9-9,30, 9.30-10, 10-10.30, 10.30+?

Analysis 2: If the meal started at 6.00pm, then given that the victim was still alive at 9.00pm, what are the probabilities of her dying at 9-9,30, 9.30-10, 10-10.30, 10.30+?

Analysis 3: If the meal started at 6.30pm, then given that the victim was still alive at 9.00pm, what are the probabilities of her dying at 9-9,30, 9.30-10, 10-10.30, 10.30+?

Analysis 4: If the meal started at 7.00pm, then given that the victim was still alive at 9.00pm, what are the probabilities of her dying at 9-9,30, 9.30-10, 10-10.30, 10.30+?

Ironically (and perhaps counter-intuitively) a later meal start time actually makes it more statistically likely that Meredith died shortly after 9pm. The reason for this is the shape of the bell curve at and after (say) 2 hours after ingestion (corresponding in our case to a 7pm meal start time), compared to the shape of the bell curve at and after 3 hours after ingestion (corresponding to a 6pm meal start time). If this requires further explanation and clarification, I'll be happy to do so but will probably need to construct a number of graphs to illustrate the principles.

 

[LondonJohn]

I am certain the Florence motivations will be published and the defence teams will appeal based on its contents as they really don’t have a choice.

Uh. So am I. But that's not what this series of posts was about.

 

[Samson]

Your issue is with the principle of prosecution appeals, not DJ. So far as Italy goes anyway. The US entered into its treaty with Italy knowing their system and Diocletus has posted stuff to the effect that the ECHR does not regard prosecution appeals as intrinsically unfair.

We presently only have prosecution appeals against sentence although the way things are going I'm sure we will fix that soon. We already abolished (the rule against) DJ a while back. I don't know why we bother with trials TBH. The police should just be able to deem people guilty. If you have one, tear up your extradition treaty with us.

In fact the fast track trials were introduced long after the treaty was signed. The fast track trial of Rudy meant that he was effectively silenced, despite being the only witness that could describe how the crime he accused Amanda and Raffaele of committing, occurred. Is it possible anywhere else to accuse a person of murder, as an eyewitness, and avoid being called to testify?

 

[Desert Fox]

You know, I think Dan actually could make a case that, without the testing of the semen, there is no rock solid evidence that Rudy did assault Meredith sexually. It is doubtful he touched her genitals accidentally, but it's possible. To me, though, there would be no reason for him to move her or position her without a sexual motive. He would have just killed her and left.

When Luca Lalli first examined the body, he said he could not be certain of rape. It's funny how it went from that point to being all about rape in Mignini's mind.

It is possible that there was some failed sexual activity.
I have to be honest, his defense that they had consensual sex indicates that he did rape her.

 

[Grinder]

Lj,

There was one study that indicated that once a certain longer time for GE is reached ( 3 hours IIRC) the curve actually does form an upwards tail.

As much as it may trouble you i think that you will in the end agree that from digestion it can't be proven, when combined with the 9:26 Naruto interface, that the kids could not possibly have been involved.

While everyone agrees that the longer from dinner the greater likelihood GE would have commenced, with the odd exception mentioned above. Kaosium maintains that the chance of Meredith living until 9:45 or later is so small it is definitive proof the kids are innocent.

Clearly the later she ate the more of the curve that lies beyond any time picked. If as one of the girls apparently indicated Meredith didn't eat right away and the dinner started at 6:30 perhaps Meredith didn't eat until 6:45 or 7. At that time the outlier GE subject of 200 minutes puts the TOD well past 10. That study used a small meal designed to digest more quickly than a "regular" meal in order to condense the time needed for each subject.

To be clear I am not saying that TOD couldn't have been as early as minutes after she entered the cottage.

I believe that the phone activity is more powerful at putting an end on the time possible of about 9:45-55.

 

[Samson]

If she began eating at 8 it would not be far-fetched to say that she must have been murdered by 9:45. The outlier times for T(lag) with a very small and easy to digest meal is 200 minutes according to one study and 120 minutes with another.

Bottom line is that there is not prcise enough information nor is digestion a precise enough indicator to place TOD within a 1/2 hour period in this case.

Obviously I agree that the meal start time is unknown in general and more so as far as Meredith. The vagaries of digestion add to the impossibility of such a narrow window.

The ILE made so many errors I find it hard to rule out one in this element. Lalli or someone mishandled one of the alcohol tests or so they say. He didn't weigh the body which apparently is forensics 101. Lalli misstates digestion time and in fact uses the wrong category T(1/2) rather than T(lag). At least my research indicates that 2 hours for GE initiation is too long for low end.

As always I would love to see a group of gastric specialists write an analysis of this case.

I would suggest Perumal as an independent from South Africa, involved in the Pistorius case. The media would report on anything that could link Knox and Pistorius. He jeopardised Pistorius' defence by linking stomach contents to a short time between eating and death.

 

[Bill Williams]

Someday someone should count up all the unfulfilled predictions Peter Quennell has made. They would fill a book.

Given that Peter Quennell likes to brag that he has 100s of lawyers behind his website, it is telling that one of the only ones who offers a "legal opinion" that PQ acknowledges is Machiavelli/Yummi.

And Yummi/Machiavelli, whoever the heck he is, has admitted to being not a lawyer.

 

[platonov]

I don’t know about that but you’d definitely save a quarter.

Explain to me what you mean, and I will see if I can help you out.

Kaosium – you brought them up [the 25% error bars].
Can you help out LJ here / refresh his memory

LJ – I hope you are not just making numbers up and then forgetting about them