Continuation Part 11: Amanda Knox/Raffaele Sollecito

[anglolawyer]

LOL - look. I give up. I will try and think of an example that doesn't involve blasted luminol at all!

 

[Dan O.]

Aaarrgghh!!! It doesn't matter DF! IT's only a freakin' thought experiment (obviously a very badly designed one).

I disagree. It's not a badly designed thought experiment. It is indicative of how many prosecutors aparently think. It may be good enough to sway the jury. But until it is nailed down with the math it won't reveal the truth.

 

[anglolawyer]

I disagree. It's not a badly designed thought experiment. It is indicative of how many prosecutors aparently think. It may be good enough to sway the jury. But until it is nailed down with the math it won't reveal the truth.

With great respect, my esteemed and learned friend, codswallop!

 

[Numbers]

I disagree. It's not a badly designed thought experiment. It is indicative of how many prosecutors aparently think. It may be good enough to sway the jury. But until it is nailed down with the math it won't reveal the truth.

I would put this a bit differently. The nailing down is with some additional chemistry and biochemistry (such as a second presumptive test and an antigen-antibody test for human blood). There is little or no need for math to be used by the actual testers. The tests are colorimetric.

 

[christianahannah]

I can even give you an example from the current case. The Luminol tract in the hall showing the characteristic rings of the shoe print. If this tract had not been previously analyzed and even if there had been no TMB and DNA tests from this Luminol tract, because this tract is part of a continuous trail from the murder room where the substance creating the shoe print was in fact identified as Meredith's blood, it can be stated as fact that this tract is also blood BARD.

If the Luminol spray had been continued into the living room and kitchen and the continuation of the shoe print marks were discovered that were not directly visible I would also concur that they were blood.

Why was Luminol not sprayed in the living room and kitchen? The clear answer is that they were only looking for evidence that would incriminate Amanda so they were only looking where she said she walked barefoot after stepping out of the shower onto the bloody bathmat. And they were only recording tracts that looked like bare footprints and ignoring the tracts that looked like Rudy's bloody shoe prints. This bias in the evidence collection should result in the evidence and those collecting it being thrown out.

Luminol was sprayed into the living room/kitchen area, in addition to being sprayed in Filomena's room and the large bathroom (I think page 83 of Stefanoni's testimony - I have not checked this is so). I am not sure how much of an area was sprayed but there were no results results yielded in the large bathroom or living room/ kitchen.

Good to know I can be gone a day or two and luminol is still being discussed. When I have time I may go back and try to answer some of Chris_Halkides questions on that very subject.

 

[Numbers]

LOL - look. I give up. I will try and think of an example that doesn't involve blasted luminol at all!

The problem goes beyond luminol.
It's the multistep issue.

Here's an analogy:

To go to work in the morning, I: 1) enter my garage, 2) get in my car, 3) turn it on, 4) press a button to open the garage door, 5) press slightly on the accelerator pedal, and 6) drive the car through the open garage exit.

The above description is an analogy to a "protocol" for a scientific or forensic test.

Now, lets partially complete (truncate) the protocol. Let's stop at (3) for example.
You now tell me that I am at work. Others may disagree.

If we skip a step, the result may also be undesirable. Suppose we complete all steps but (4), although the garage door is in the closed position.

This silly analogy may help explain what the technically-trained feel when a multistep process blood identification process is characterized as completed by a truncated process.

 

[Kaosium]

Someone on another thread introduced me to the Dunning Kruger effect recently. IMO Ms Schaps, whom I regard as an honest but misguided poster on this case, greatly over reaches when she gets into the DNA evidence.

They missed one in the historical precedents section at the bottom:

"The best lack all conviction, while the worst
Are full of passionate intensity."

 

[Dan O.]

Luminol was sprayed into the living room/kitchen area, in addition to being sprayed in Filomena's room and the large bathroom (I think page 83 of Stefanoni's testimony - I have not checked this is so). I am not sure how much of an area was sprayed but there were no results results yielded in the large bathroom or living room/ kitchen.

Good to know I can be gone a day or two and luminol is still being discussed. When I have time I may go back and try to answer some of Chris_Halkides questions on that very subject.

Page 200 (original) of Massei reports: "The living room-kitchen corner gave a negative result on the Luminol test, as did the larger bathroom."

This is revealing because we know that there were visible blood tracts in the livingroom and there must be another step from the last marker H. Has anybody checked the label od the product the Italian police were using? Does it perhaps say: "Lumin-some-big-bare-footprints".

 

[anglolawyer]

The problem goes beyond luminol.
It's the multistep issue.

Here's an analogy:

To go to work in the morning, I: 1) enter my garage, 2) get in my car, 3) turn it on, 4) press a button to open the garage door, 5) press slightly on the accelerator pedal, and 6) drive the car through the open garage exit.

The above description is an analogy to a "protocol" for a scientific or forensic test.

Now, lets partially complete (truncate) the protocol. Let's stop at (3) for example.
You now tell me that I am at work. Others may disagree.

If we skip a step, the result may also be undesirable. Suppose we complete all steps but (4), although the garage door is in the closed position.

This silly analogy may help explain what the technically-trained feel when a multistep process blood identification process is characterized as completed by a truncated process.

Are you saying no usable result can ever be obtained from any scientific test unless conducted strictly in accordance with protocol?

 

[Bill Williams]

The point is, unless the kitchen knife matched the bedsheet outline, then with no DNA tying the kitchen knife to the crime, then there is nothing tying that knife to the crime.

Crini had to come up with that knowing that the RIS Carabinieri report to the Nencini court added nothing to the understanding of the crime - in relation to sample 36i.

Go bad to the threads circa Nov 2013 where you vigourously argue that the kitchen knife is a match for the bedsheet outline.

Then turn to the Nencini report for a discussion of it.

Bill, when arguments are imported to this forum, progress can be made, because this is the default forum. Leila's last contribution on IIP on the subject was this: on February 4 2014:

Leila Shneps says:

Indeed, I wished for nothing so much as to discuss all the DNA evidence in more detail and above all the new test on the knife, but all the most interesting parts as far as I am concerned had to go due to the stringent word limit. Don't worry, further articles are in the making.

I thoroughly disagree that the new test on the knife, which for my part I would always have strongly recommended, is exculpatory. To me it proves exactly one thing: perfectly valid tests with clear results can be performed on extremely small samples, indeed on a sample so small that Vecchiotti wrongly claimed it was too small to test. If the result of 36I is accepted, then I can see no reason to exclude the one on 36B; certainly not the smallness of the sample.
Of course, it does not weigh as it would have had it showed Meredith's DNA. But it lends weight to the previous result by removing one of the major objections raised by Vecchiotti, that very small samples cannot give reliable results. (Recall that the "too low" result obtained by Stefanoni from the Qubit Fluorometer meant only that the sample was below 200 pg, which is really quite ample for testing purposes; I believe that 36I was much smaller.) Her other major objection, that the DNA result could have been a trace from a previous run of Meredith's DNA through the machine was removed in court, where she herself admitted that the six day gap between testing of any item containing Meredith's DNA and the knife was too much to allow of this type of contamination.

Thus the result of the testing on 36I as I see it, and as I believe the court saw it, was essentially to invalidate Vecchiotti's assertion, and to validate the result of 36B.

I have seen it expressed on here (by Chris Halkides, I believe) that one possible source of the DNA on the knife was innocent transfer before the crime, or possibly transfer during the transport of the knife (although this would have to be tertiary or quaternary transfer which is very unlikely). Under these hypotheses, the laboratory work on the knife is not at fault. If these arguments had been presented in court, it would have been up to the jury to deliberate whether the DNA came on the knife by innocent transfer (namely Amanda touching Meredith and then touching nothing else on her way back to Raffaele's until she directly handled a knife blade), or tertiary transfer during transport, or during the crime. Even if I don't find the innocent transfer argument convincing - it would be more so if Meredith's DNA had been found on Amanda's clothing, for instance - it is certainly more realistic than random accusations of contamination flying around the lab. However, none of these possibilities were raised in court. I can't deny that Amanda's defense appears to me to have been sadly lacking in imagination all through.

Leila Schnepps does not understand the problem with 36B, even as described by Judge Massei in his 2010 motivations.

The problem with 36 B is that it was such a small example, Stefanoni could only do one test on it, since the test was a destructive one. To do a full examination of 36B required two tests:

1) the first to determine what it was
2) the second to determine who it belonged to​

Stefanoni chose - rightly so - to find out who it belonged to. There was no point in doing the lone test to find out what the sample was, without knowing who it belonged to. (But then again, please consider a prior issue. Stefanoni said she found the sample 36B in a striation on the blade that no one else can either find or see.)

Stefanoni said that the same belonged to Meredith. Yet unless it is Meredith's **blood**, knowing who it belongs to is relatively meaningless.

So, Leila Schnepps first. She is misrepresenting the issue with regard to the smallness of the sample 36B. Her analogy between it and 36I is meaningless.

And Machiavelli second. Crini had to argue for the bedsheet stain being a match for Raffaele's kitchen knife to keep the kitchen knife as part of this crime. Please note how Machiavelli de facto agrees with this, since he tries to argue for a second knife on other grounds, grounds other than the bedsheet stain.

I am getting so tired of reading and rereading the Massei Motivations report.... but the whole reason a second knife had to be brought to the prosecution (alleged and not proven) was because of the obvious mismatch of the kitchen knife and the bedsheet stain - a stain of a blade outline in Meredith's blood.

Once again - asserted and not proven.....

Yet when the DNA collapses on the kitchen knife - Crini knows full well that he's going to have to argue some other evidence to keep that knife in play...

...... hence - go back to those threads circa Nov 2013. You'll find no less than Machiavelli himself (or one of them, anyway) arguing for an identity of the kitchen knife with the bedsheet stain.....

...... all a de facto admission that the DNA evidence linking the kitchen knife to the crime otherwise, is bogus.

As someone else said in commenting on this - Machiavelli (one of them at least) is virtually arguing for reasonable doubt for AK and RS.)

It is also of note that this is the firs time in a long while that Machiavelli has responded to something Bill Williams has posted here.

Which is also a de facto admission that Machiavellis is going to give a wide berth to his former claim of conspiracy within the Perugian Court of Appeal....

.... namely that he (or one of them at least) believes that De Nunzio, the Court of Appeal president, conspired criminally with Hellmann to pervert justice in Italy.

 

[Chris_Halkides]

addendum to luminol list

Good to know I can be gone a day or two and luminol is still being discussed. When I have time I may go back and try to answer some of Chris_Halkides questions on that very subject.

I thought of one more question to add to my list for the PG-commenters. 11. What is your explanation for the luminol-positive reactions in Raffaele's flat?

 

[Numbers]

Are you saying no usable result can ever be obtained from any scientific test unless conducted strictly in accordance with protocol?

No. But the result of not following protocol, which may or may not be "useable", must be evaluated or investigated more than if one had followed protocol (in a test of known reliability) and the "use" may be for some purpose other than intended for the original protocol.

In the silly car analogy, if I stop at (3), I have verified that my car battery and starter can start my car's engine (for the typical gasoline-powered vehicle). I have verified that, if someone had planted a bomb in my car, it has not yet gone off even though I turned the ignition. But I have not given myself the ability to drive to work with the car unless I follow through with step (4) and subsequent steps.

In science, useful results sometimes occur from not following protocols or making mistakes. Sometimes new materials are invented or new discoveries are made. Sometimes people blow themselves up instead. It all depends; it is contingent on the details. But if one is doing forensic work, the investigation of a crime, one should follow the protocols established to give reliable results useful for a FAIR and UNBIASED criminal investigation. Not a FALSE or MISLEADING investigation.

 

[Chris_Halkides]

Gill interview

Peter Gill describes framework or hierarchy of levels (p. 19 of his book) at which DNA evidence can be evaluated:
1. The sub-source level refers to the strength of evidence of the DNA profile itself.
2. The source level is an evaluation of the strength of the DNA profile if it can be associated with a particular body fluid, such as semen or blood
3. The activity level associates the DNA with the crime itself, e.g. sexual assault.
4. The highest level deals with the ultimate issue of guilt/innocence.

A link to a radio interview on 4 November is here. The interview with Dr. Gill starts around 20 minutes into the broadcast. He mentions the issue of not knowing the source of the DNA with respect to low template work.

 

[Numbers]

Are you saying no usable result can ever be obtained from any scientific test unless conducted strictly in accordance with protocol?

I like your question so much that I will offer, if you don't mind, another analogy.

Suppose we change our frame of reference to medical diagnostic tests.

An individual, X, goes to a physician for a diagnosis; X's symptoms suggest either a very serious and often debilitating or fatal disease, or a very mild condition and short-term condition that does not at all affect one's life and health over-all. The physician orders a diagnostic test.

The results come back. X is very seriously ill. But then the physician receives an urgent call from the diagnostic lab. It seems that the technician doing the test either failed to complete certain steps in the protocol or perhaps mixed up samples; they're not sure. However, to make amends, they will discount the cost of the test substantially.

What useful result has come from this particular diagnostic test regarding X's state of health?

Besides, perhaps especially in the US, a possible malpractice law suit?

 

[codyjuneau]

After detecting the body-shaped luminol CL the next step is to do additional tests, such as an antibody-antigen test for blood.

And one should not have over-applied the luminol if high concentrations can produce self-CL. Look for swirl or wipe marks or similar indications of a clean-up. Look for other evidence. You don't mention if it's known that a criminal act involving blood is known or suspected in your hypothetical.

Could be that someone rolled in dirt or mud that had iron compounds or other material that would produce luminol CL and then laid down on the floor. No crime in that scenario.

Turnip juice or blood. You decide.

Look, I can keep tweaking the thought experiment further if you want. You are already moving into the realm of fanciful explanation which lies beyond the zone of reasonable doubt and becomes too much to swallow. I mean, when was the last time you rolled in dirt and then came home and rolled around on the floor? Not impossible, just fanciful. And yes, I was posting in the context of a murder enquiry so it went without saying, I thought, that there had been a murder.

We agree the luminol evidence in this case is garbage. It doesn't even have zero value since the negative effect of the associated lies, certainly using Nencini's test (the one he sarcastically employed when rejected the defence request for re-testing of the clasp) obliges us to discount or discard all of Stefanoni's evidence. We don't agree whether luminol-only evidence can ever have probative value, even weak, non-zero value in the right context.

.
Not that fanciful if the someone is a dog that lives there.

Cody
.

 

[Numbers]

.
Not that fanciful if the someone is a dog that lives there.

Cody
.

Yes.

 

[codyjuneau]

.
FWIW, I once wrote my last will and testament using the 'C' programming language. One could enter the relevant conditions at the time of my death and the program would spit out a result indicating how to divide up my estate. The source code was available obviously, it was easily readable, and it had the advantage of being able to supply testable results while I was still alive. I later reconsidered because I was not sure it would be legally acceptable.

I have often thought that a programming language specifically designed for legal issues might be useful. For example, I think it would be fairly easy to write code that would spit out the duration and conditions of a prison sentence based on whatever rules Machi was trying to explain the other day. It would certainly be easier to understand. As with any new endeavor, some of the solutions might not be obvious but would evolve with useage and development. In fact just implementing such a concept would probably inspire new ways of looking at old problems, such as the current topic of when evidence is deemed reliable and precise. Think of how computers themselves have changed how we look at the world now. Who in 1980 could have envisioned the solutions that computers and the Internet have evolved to provide.

Cody
.

 

[anglolawyer]

I like your question so much that I will offer, if you don't mind, another analogy.

Suppose we change our frame of reference to medical diagnostic tests.

An individual, X, goes to a physician for a diagnosis; X's symptoms suggest either a very serious and often debilitating or fatal disease, or a very mild condition and short-term condition that does not at all affect one's life and health over-all. The physician orders a diagnostic test.

The results come back. X is very seriously ill. But then the physician receives an urgent call from the diagnostic lab. It seems that the technician doing the test either failed to complete certain steps in the protocol or perhaps mixed up samples; they're not sure. However, to make amends, they will discount the cost of the test substantially.

What useful result has come from this particular diagnostic test regarding X's state of health?
Besides, perhaps especially in the US, a possible malpractice law suit?

None. What if the technician took all the steps but because he was distracted and didn't get round to labelling the specimen, he is not certain but is nevertheless pretty sure of the result and there is no time to conduct a fresh test?

Complete certainty is nice to have but not always (or ever?) attainable. If we insist on scientific certainty and nothing less then we may end up with no scientific evidence at all. IOW even with properly validated procedures we are only approaching certainty, not attaining it. Not always anyway.

The degree of certainty required will depend on the task. If we are sending a guy to Mars we have almost no margin for error but certainty is not required to prove guilt. Are we certain this fingerprint belongs to the defendant? Certain?
Not everyone thinks so.

 

[Numbers]

None. What if the technician took all the steps but because he was distracted and didn't get round to labelling the specimen, he is not certain but is nevertheless pretty sure of the result and there is no time to conduct a fresh test?

Complete certainty is nice to have but not always (or ever?) attainable. If we insist on scientific certainty and nothing less then we may end up with no scientific evidence at all. IOW even with properly validated procedures we are only approaching certainty, not attaining it. Not always anyway.

The degree of certainty required will depend on the task. If we are sending a guy to Mars we have almost no margin for error but certainty is not required to prove guilt. Are we certain this fingerprint belongs to the defendant? Certain?
Not everyone thinks so.

Again, we need to consider as you indicate the consequences of error.

But why bother to do any test at all? First the sentence, then the trial. (Paraphrasing Lewis Carrol, maybe.)

And in Italy, since you can be in preventive detention for a year or so until your trial, detained on bogus evidence that the prosecutor knows is false, that must be justice.

 

[Diocletus]

Are you saying no usable result can ever be obtained from any scientific test unless conducted strictly in accordance with protocol?

Usable for what? The sole piece of evidence justifying the imprisonment of someone for life? Yeah, protocol-violating evidence shouldn't be used for that.