Continuation - Discussion of the Amanda Knox case

[LondonJohn]

Well I was looking for a clapping smiley face but I can't find one.

CLAP CLAP CLAP

(And I'm a staunch you-know-what but I differ from my party on certain things.)

I echo your sentiments - it was the use of the word "wife" which confused me. I have one gay male friend and one gay female friend who are both in civil partnerships (as they're called here), and they both use the term "partner" rather than "husband" or "wife" to refer to their other half. Even so, I guess I feel slightly sheepish at any sort of assumption that writing the word "wife" might have implied that the writer was male!

 

[shuttlt]

I don't imagine that the shape of the curve is going to be radically different, which is going to be the main thing

Hard to know though, isn't it?

However regardless it's going to remain overwhelmingly likely that Meredith died at the earliest possible time consistent with the witnesses saying she did not get home until 21:05 or something similar, and it's going to remain reasonably medically certain that she was dead by 22:00, or maybe 22:30 at the very latest.

Doesn't this depend on knowing the shape of the extreme edge of the distribution? Presumably the hard lower bound of t=0 is going to make the distribution asymmetrical. Will this have an impact on the calculations?

 

[LondonJohn]

Does anyone have an estimate for how much time it takes for the stomach to empty, from beginning to end? Thanks.

That's a more difficult question to answer, as in this case it's far more dependent upon the amount of food consumed. The general range is considered to be between 2 hours (for a small meal of easily-digestible food) and 5-6 hours (for large meals with harder-to-digest elements).

Incidentally, someone overnight (my time) - Machiavelli I believe - made an observation about casein in cheese being harder for the stomach acids and enzymes to break down. While this is true, my understanding is that it would have little or no effect upon T(lag) - unless the entire meal were casein-based. The stomach is a far more intelligent organ than most people realise. It is not only capable of compartmentalisation through muscular constriction (solving the "multiple meal" problem). It also uses careful regulation of the opening of the pyloric sphincter to "sieve" digested chyme from partially-digested food matter. It then passes the already-digested matter through to the duodenum, while retaining the partially-digested matter until it is properly broken down. It doesn't therefore need to wait until the entire meal has been broken down before it even starts to pass chyme through to the duodenum.

Many gastroenterologists and anatomists contend that there is essentially a whole second brain worth of neural network situated in the abdomen, dedicated to operating the digestive system. It's an extraordinarily complex and complicated piece of machinery, with an incredible amount of individual muscles and nerve fibres. I've personally seen more of my own GI system than I'd ever bargained upon seeing, and I'm fascinated by it.

 

[RoseMontague]

Looks like a great discussion on TOD in reference to the stomach contents yesterday. My opinion on this is that the start of the mealtime is certain in the range of 6PM to 6:30PM. The TOD put forward in the Massei report of about 11:30 or slightly after makes no sense in relation to the time of her last meal as that would be 5 hours or more past the time she ate. None of the experts quoted in the Massei report quote 5+hours on this and I have not seen this figure in the tests quoted here. I believe all the evidence we have including Meredith's cell phone records and Rudy's statements indicate a TOD shortly after 9PM. I think the computer activity around this time is very important and I believe the homeless park bench fellow's testimony is worthless.

 

[LondonJohn]

Hard to know though, isn't it?


Doesn't this depend on knowing the shape of the extreme edge of the distribution? Presumably the hard lower bound of t=0 is going to make the distribution asymmetrical. Will this have an impact on the calculations?

It is indeed an asymmetric bell curve, which is why it can't correctly be termed a normal curve. However, the curve follows a bell pattern - it's just that the left half of the curve is squeezed since - as you point out - t=0 is a solid backstop on the left edge.

Also, in response to something you said earlier about what I'll call the "Robert Wadlow conjecture", Wadlow had an easily-identifiable (even in the 1920s) pituitary problem, which resulted in a highly abnormal amount of growth hormone being released (he was still growing at the time of his death at age 22). He therefore sits outside the sample population for height analysis, since his height was the result of bodily malfunction rather than normal genetic programming. Had his pituitary problem been successfully treated in childhood, he would have attained a "normal" height consistent with his DNA (and a normal diet and environmental factors). When you remove bodily malfunction from the equation, it's vanishingly rare for an adult male to grow beyond 7ft tall.

The reason why this is important is that Meredith had no known gastro-intestinal problems. To my knowledge she'd never had any abnormalities in this area, and nothing was identified in the post-mortem analysis. We can therefore be pretty confident that she conformed to the known parameters on gastro-intestinal function.

 

[Chris_Halkides]

ask your surgeon or anesthesiologist

That's a more difficult question to answer, as in this case it's far more dependent upon the amount of food consumed. The general range is considered to be between 2 hours (for a small meal of easily-digestible food) and 5-6 hours (for large meals with harder-to-digest elements).

LondonJohn,

Yesterday I cited an anesthesiology group's patient information that said, “By not eating solid food for at least 6 hours or drinking clear liquids for 2 hours prior to surgery, you decrease the risk of aspiration pneumonia and increase the safety of your anesthetic.”

The 6-hour period suggests to me that most of the contents of the stomach will be gone by this time. This makes sense if t(lag) were 2-3 hours but not if t(lag) were 5 hours.

 

[LondonJohn]

Looks like a great discussion on TOD in reference to the stomach contents yesterday. My opinion on this is that the start of the mealtime is certain in the range of 6PM to 6:30PM. The TOD put forward in the Massei report of about 11:30 or slightly after makes no sense in relation to the time of her last meal as that would be 5 hours or more past the time she ate. None of the experts quoted in the Massei report quote 5+hours on this and I have not seen this figure in the tests quoted here. I believe all the evidence we have including Meredith's cell phone records and Rudy's statements indicate a TOD shortly after 9PM. I think the computer activity around this time is very important and I believe the homeless park bench fellow's testimony is worthless.

In addition to all the stomach contents debate, there's another point which by itself nails the prosecution/court theory of a ToD at 11.30-11.45pm: Guede was already in the company of other people across town by 11.30pm. Now Guede was undoubtedly in the murder cottage after Meredith was stabbed - he left a fingerprint in Meredith's blood, and shoeprints in her blood. And even by his own account, he had blood on his clothing, meaning that he changed clothes once he got back to his apartment.

All of this means that the murder can't have happened any later than about 11.15pm (being generous). And, in that case, it can't have happened later than 10.30pm, owing to the witnesses in the broken-down car opposite the cottage. They firmly testified that they saw or heard no activity inside the dark cottage - and no comings or goings - during all the time they were there between 10.30pm and 11.40pm.

So Guede's appearance across town at 11.30pm actually pushes the time of death back to before 10.30pm, in and of itself. I wonder if the people who met with Guede at 11.30pm will be brought in by the defence to testify in the appeal?

 

[LondonJohn]

LondonJohn,

Yesterday I cited an anesthesiology group's patient information that said, “By not eating solid food for at least 6 hours or drinking clear liquids for 2 hours prior to surgery, you decrease the risk of aspiration pneumonia and increase the safety of your anesthetic.”

The 6-hour period suggests to me that most of the contents of the stomach will be gone by this time. This makes sense if t(lag) were 2-3 hours but not if t(lag) were 5 hours.

Exactly. I don't think there's ever been a recorded example of a T(lag) anywhere near 5 hours (or even anywhere near 4 hours) in an otherwise healthy adult. As I pointed out a couple of posts ago, total stomach emptying can be retarded by the presence of certain hard-to-break-down proteins or fibres in the meal. But even these items are totally reduced to chyme within 5 hours or so - by which time pretty much all of the rest of the meal will already have passed through to the duodenum. The stomach is intelligent enough to pass through the meal in the order in which it's broken down into chyme.

 

[shuttlt]

It is indeed an asymmetric bell curve, which is why it can't correctly be termed a normal curve. However, the curve follows a bell pattern - it's just that the left half of the curve is squeezed since - as you point out - t=0 is a solid backstop on the left edge.

Also, in response to something you said earlier about what I'll call the "Robert Wadlow conjecture", Wadlow had an easily-identifiable (even in the 1920s) pituitary problem, which resulted in a highly abnormal amount of growth hormone being released (he was still growing at the time of his death at age 22). He therefore sits outside the sample population for height analysis, since his height was the result of bodily malfunction rather than normal genetic programming. Had his pituitary problem been successfully treated in childhood, he would have attained a "normal" height consistent with his DNA (and a normal diet and environmental factors). When you remove bodily malfunction from the equation, it's vanishingly rare for an adult male to grow beyond 7ft tall.

The reason why this is important is that Meredith had no known gastro-intestinal problems. To my knowledge she'd never had any abnormalities in this area, and nothing was identified in the post-mortem analysis. We can therefore be pretty confident that she conformed to the known parameters on gastro-intestinal function.

Certainly Wadlow had a medical problem and Meredith had no known gastro-intestinal problems, but are any of us expert enough to know that gastro-intestinal problems will necessarily be so obvious as to be diagnosed? Wadlow's condition is so obvious as to be unmissable, even if the cause requires expertise to identify. Having a lag time of 4 or 5 hours, say, strikes me as the kind of abnormality that could easily go unnoticed. I'm speculating of course, but I don't see how one can assume that we are talking about a sample with no weird outliers?

I hope it doesn't come across that I am pushing a Meredith-had-a-weird-gut theory of the crime. It just seems to me that this is a flaw in the current argument, or at least the calculated probabilities.

 

[RoseMontague]

Certainly Wadlow had a medical problem and Meredith had no known gastro-intestinal problems, but are any of us expert enough to know that gastro-intestinal problems will necessarily be so obvious as to be diagnosed? Wadlow's condition is so obvious as to be unmissable, even if the cause requires expertise to identify. Having a lag time of 4 or 5 hours, say, strikes me as the kind of abnormality that could easily go unnoticed. I'm speculating of course, but I don't see how one can assume that we are talking about a sample with no weird outliers?

I hope it doesn't come across that I am pushing a Meredith-had-a-weird-gut theory of the crime. It just seems to me that this is a flaw in the current argument, or at least the calculated probabilities.

There is no indication that Meredith could be an outlier on this issue but it makes more sense than Micheli, Matteini, and even Rudy's appeal judge. These reports all quote the 2 to 3 hour range and then proceed to ignore testimony and vaguely claim the the meal was done by 9PM because Meredith was on her way or home at that time. This despite even listing testimony from Sophie and others of a meal eaten closer to 6PM. Their reasoning also seems to ignore the start of the meal time and instead uses the end of the meal time. Massei seemed to recognize the problems with this and buys into what appears to be an invented error in the autopsy scenario to justify the later TOD.

 

[RoseMontague]

BTW,
I believe platonov was asking for any human translations of relevant appeal sections on the TOD issue. katy_did had provided one such translation here:

http://www.internationalskeptics.com/forums/showpost.php?p=6220709&postcount=4335

 

[shuttlt]

The question came up before as to why it took so long to take the temperature of the body, and I think there was some talk of incompetence.

Some difference of opinion exists over the use of a thermometer at the scene of a suspicious death. Considerable caution must be employed when considering the taking of a rectal temperature with the body in situ. If there is any possibility at all of some sexual interference, whether homosexual or heterosexual, no intereference with the clothing or perineum must be made until all forensic examinations have been completed. Certainly, no instrument should be inserted into the rectum before trace evidence has been sought

http://www.dundee.ac.uk/forensicmedicine/notes/timedeath.pdf

 

[Kevin_Lowe]

Certainly Wadlow had a medical problem and Meredith had no known gastro-intestinal problems, but are any of us expert enough to know that gastro-intestinal problems will necessarily be so obvious as to be diagnosed? Wadlow's condition is so obvious as to be unmissable, even if the cause requires expertise to identify. Having a lag time of 4 or 5 hours, say, strikes me as the kind of abnormality that could easily go unnoticed. I'm speculating of course, but I don't see how one can assume that we are talking about a sample with no weird outliers?

I hope it doesn't come across that I am pushing a Meredith-had-a-weird-gut theory of the crime. It just seems to me that this is a flaw in the current argument, or at least the calculated probabilities.

Anything's possible. Pathologists use constructions like "reasonable medical certainty" to describe the claim that food will have begin to leave the stomach within three, or just maybe four, hours of ingestion. Almost all of the time this will be the case, but people can be born with two heads so being born with a t(lag) of five hours doesn't on the face of it violate any laws of physics or anything as far as I know.

However from the studies I have read three to four hours already covers the outliers. The longest t(lag) I've seen in any study was someone who had a t(lag) of 180 minutes or so after eating quite a heavy meal, and they were a clear outlier in their sample group. I suspect that those are the people with undiagnosed gastro problems who push the envelope out to four hours when the vast majority of people will have t(lag) times which are substantially shorter. So while I can't point to a citation to back it up, I think it's likely that the outliers are already mostly taken care of by the numbers you have already seen.

Obviously you can't exactly convict someone on the basis that the alleged victim could in theory have had an incredibly unusual and undiagnosed gastrointestinal anomaly which was not detected in the autopsy or in any other medical procedure ever performed on that person, and which resulted in a t(lag) longer than any I've seen in the literature for a healthy person. It's philosophically possible but for legal purposes I think it's as good as impossible, and as explanations for the facts go it's a huge stretch compared to to just figuring she died around 21:10.

 

[shuttlt]

Kevin,

I am mainly arguing with your calculation of the percentages.

Also, I came across this quote:

Results:  The mean ± SD of half gastric emptying time (T1/2) of a fluid test meal was determined to be 80.5 ± 22.1 min and for Tlag to be 40.3 ± 10.2 min. However, the T1/2 and Tlag of solid meals did not fit to normal distribution and thus median and percentiles were determined. The median time of T1/2 for solids was 127 min (25–75% percentiles: 112.0–168.3 min) and 81.5 min for Tlag (25–75% percentiles: 65.5–102.0 min).

from back in September (http://www.internationalskeptics.com/forums/showpost.php?p=6389510&postcount=8215). That seems to be saying that a calculation like the one you are doing, can't be done. Again, I'm mainly arguing with your calculations of percentages here. I don't see how one can really say more than that in the experience of gastroenterologists, you don't see lag times of more than four hours (of whatever it is). The numbers just give a false sense of precision.

 

[Kevin_Lowe]

Kevin,

I am mainly arguing with your calculation of the percentages.

Those are LondonJohn's calculations, not mine. I know of no reason to think they are not accurate but they aren't my work.

Also, I came across this quote:

from back in September (http://www.internationalskeptics.com/forums/showpost.php?p=6389510&postcount=8215). That seems to be saying that a calculation like the one you are doing, can't be done. Again, I'm mainly arguing with your calculations of percentages here. I don't see how one can really say more than that in the experience of gastroenterologists, you don't see lag times of more than four hours (of whatever it is). The numbers just give a false sense of precision.

If you'd read a bit further, you'd have seen (as LondonJohn recently explained again just recently for you) that t(lag) doesn't follow a normal distribution because the left hand side of the curve has a hard limit at t=0. However the right hand side of the curve is the same shape as a normal distribution. That is all the text you quoted is saying.

 

[Fine]

Anything's possible. Pathologists use constructions like "reasonable medical certainty" to describe the claim that food will have begin to leave the stomach within three, or just maybe four, hours of ingestion. Almost all of the time this will be the case, but people can be born with two heads so being born with a t(lag) of five hours doesn't on the face of it violate any laws of physics or anything as far as I know.

However from the studies I have read three to four hours already covers the outliers. The longest t(lag) I've seen in any study was someone who had a t(lag) of 180 minutes or so after eating quite a heavy meal, and they were a clear outlier in their sample group. I suspect that those are the people with undiagnosed gastro problems who push the envelope out to four hours when the vast majority of people will have t(lag) times which are substantially shorter. So while I can't point to a citation to back it up, I think it's likely that the outliers are already mostly taken care of by the numbers you have already seen.

Obviously you can't exactly convict someone on the basis that the alleged victim could in theory have had an incredibly unusual and undiagnosed gastrointestinal anomaly which was not detected in the autopsy or in any other medical procedure ever performed on that person, and which resulted in a t(lag) longer than any I've seen in the literature for a healthy person. It's philosophically possible but for legal purposes I think it's as good as impossible, and as explanations for the facts go it's a huge stretch compared to to just figuring she died around 21:10.

________________________

Kevin,

In the studies you've read, what were the people doing while they digested their meals? Subjected to any stress, or ---as I suspect---sitting around a hospital cafeteria chit-chatting with one another and thumbing through back-issue copies of PEOPLE magazine?

For all we know Meredith may have had a disturbing ---"gut-wrenching"?---argument with Amanda before she left the cottage on the afternoon of November 1st. (Or would Amanda, or Raffaele, have mentioned that argument?)

///

 

[shuttlt]

Those are LondonJohn's calculations, not mine. I know of no reason to think they are not accurate but they aren't my work.

Quite so. Sorry for my confusion.

If you'd read a bit further, you'd have seen (as LondonJohn recently explained again just recently for you) that t(lag) doesn't follow a normal distribution because the left hand side of the curve has a hard limit at t=0. However the right hand side of the curve is the same shape as a normal distribution. That is all the text you quoted is saying.

I read that, I don't see that that is necessarily what the other quote is saying though. The Gaussian distribution may be the normal assumption, but it isn't always the right one.

 

[Kevin_Lowe]

Kevin,

In the studies you've read, what were the people doing while they digested their meals? Subjected to any stress, or ---as I suspect---sitting around a hospital cafeteria chit-chatting with one another and thumbing through back-issue copies of PEOPLE magazine?

For all we know Meredith may have had a disturbing ---"gut-wrenching"?---argument with Amanda before she left the cottage on the afternoon of November 1st. Or would Amanda, or Raffaele, have mentioned that argument?

Interesting idea. Let's pretend for a minute that Meredith would not have mentioned such a huge argument to her friends while spending time with them, although it seems like an unusual thing not to mention if it's such a huge argument that it ruins your digestion hours later even after you have relaxed with your friends, watched a DVD and eaten some pizza and apple crumble.

Pretending this, can you cite any studies that indicate that the stress of an argument can lead to a t(lag) of five hours or more in a normal, healthy young woman who has eaten a small-to-moderate sized meal of pizza with no alcohol in relaxed surroundings?

It seems possible that you are interpreting the word "stress" to mean the kind of stress refer to when we say something like "I had a really stressful day at the office". In medicine it's a very broad term that refers to anything that strains or exceeds a person's physical and psychological coping mechanisms. Very serious stress can lad to total stoppage of the digestive process, but we're talking about fairly serious insults to the body in such cases, not merely worrying about whether you left the iron on.

I'm aware of no evidence that mild psychological stress like an argument with a housemate can lead to t(lag) times of five hours or more but as always, show me the evidence and I will be convinced. I think it's very highly likely however that this question is akin to asking "I heard some people wear hats that make them taller - is it possible the killer was wearing a hat that made them four meters tall"? The sheer size of the effect you need makes me think it's unlikely to come from an argument substantially earlier.

 

[Machiavelli]

Since you cannot cite any scientific papers to back these claims up, and we have repeatedly cited scientific papers showing these claims to be completely false, you cannot expect anyone to take you seriously.

I should take seriously a guy who is unable to read a single written phrase on a diary, calculates a probability of 21,3% from three events that imply a 80% probability, thinks that people are members of self selected communities, thinks that logic is psychology and never sits on an armchair when he thinks rationally.
Who cares of what you take seriously. I repeat what I want.

This is incorrect. Her blood alcohol levels were compatible with either possibility and her stomach alcohol levels were unknown due to fermentation.

Now make a brief calculation how much alcohol she should have drunk the night before to reach that level 20 hours later. Then confront with testimonies. The decreas of alcholemy follows an exponential curve.
If alcholemy 30 minutes after assumption is 2.0 (often this amount causes ethilic pre-coma), after 18 hours will be 0,005. Which is 100 times lower than Meredith's alcoholemy.

Nonsense snipped: You make these claims but you cannot back them up and you are flatly contradicted by the scientific literature. What you are saying is factually false. Stop repeating it.

Yes, sure.

You seem to be in denial about the likely evidentiary value of the error log files, which seem highly likely to confirm Raffaele and Amanda's mutual alibi well beyond reasonable doubt. But in any case, yes, that Naruto file didn't open itself for its own entertainment.

One serious thing. I think you can understand that the logical starting point of my reasoning is: there is evidence. The evidence of ther involvement is already there. It is not something I am uncertain about.
An element like the Naruto file combined with an early time of death is merely suggestive. It is elements suggesting the idea of an alibi. They do not consist in evidence of an alibi. On the other plate of the scale there is evidence of involvement, that cannot be overweighted by suggestive elements. You should not think as if I just feel suspicion on the two. An alibi to counter this evidence would only be a proof that the two of the defendants stayed away from the cottage continuously over a period of time that covers all possible times of death.
So the wrong choice, a mistake from the roots, is to go betting on the time of death and on the Naruto file. The game can only be played on the evidence accusing them.

 

[Machiavelli]

Appealing to stress to solve the problem gets you precisely nowhere, because even if she was first attacked long before she was murdered, which is not what the forensic evidence indicates anyway, Amanda and Raffaele could not possibly have been present when she was first attacked.

My comment was:

"No? And why not? Because of Naruto episode 101?"

Yet, I didn't understand why you think Amanda and Raffaele could not possibly have been present when she started to have problems or suffering harassment at home, just based on the log of Naruto episode 101 file (if there is).