Thursday, May 9, 2013

4. The Meredith Kercher Case - Back to the Drawing Board (Part Four)


The evidence gets really confusing....



Introduction

I started out this new series because the Italian Court of Cassation recently quashed the Appeal Court's acquittal of both Amanda Knox and Raffaele Sollecito.

In Parts One through Three, I talked about the Appeal Court's decisions pertaining to the break-in and the DNA evidence (the evidence with regard to the "double-DNA" knife and the bra clasp).

In this part, I'll start to wrap things up by discussing the most interesting other aspects of the Appeal Court's ruling. I will focus on the footprint on the bathroom mat - which inevitably leads one to evaluate whether or not a clean-up operation took place -  and the traces left by  Luminol found in various places in the apartment at the Via della Pergola.  

The print on the mat and the clean-up

In the smaller of the two bathrooms in the upper floor of the house at the Via della Pergola (the bathroom used by Kercher and Amanda Knox, just next to Kercher's bedroom), a mat was found on which the imprint of someone's bare right foot was found. The print was made due to the fact that the foot has been bloodied; the blood was Kercher's.

The original court, after having heard the expert testimony from both the prosecution and the defence, decided that the footprint matched the right foot of Sollecito.

The Appeal Court rejects this, coming to the conclusion that the print is "of no evidentiary value against defendant Sollecito". In fact, the court states: "It cannot (...) be ruled out that Guede (...) experienced the loss of his right shoe in the course of the violent aggressive manoeuvres to which he subjected Kercher, thus resulting in the soiling of his foot with blood, which he took it upon himself to wash in the small bathroom situated immediately to the left of the door to Meredith’s bedroom. Otherwise, his right shoe should also have left some kind of bloody trace along the corridor as he exited; he likely went through it, however, with his right foot bare, even if cleaned of blood by this point."

Now before addressing the forensic evidence regarding possible matches of the footprint, there is another issue to bear in mind. Is the fact that print existed in the first place of any relevance for the question as to whether more than one person was involved in the murder?

Well, if we consider the Appeal Court's reasoning, as quoted above, one might assume that something along the following lines must have happened: 
  •  sometime during or preceding the attack on Meredith, Guede loses his right shoe;
  • Guede kills Kercher, doing so in such a manner as to inevitably leave considerable amounts of her blood in her room and, no doubt, on his person. He also leaves evidence of his presence in her room (there is, for example, the imprint of his bloodied hand on a pillow). He then goes to the bathroom to wash his bare foot;
  • he then cleans up, thereby forgetting or not noticing the single footprint on the bathmat (and, indeed, the hand on the pillow), and possibly puts his shoe back on (or perhaps, as the Appeal Court assumes, he does not);
  • and he then departs, leaving, however, a discernible trail of shoe prints from his left shoe.

If one accepts such a sequence of events, a clear problem arises. After all, if you wish to believe that the same person (i.e. Guede) could have left both the footprint on the mat and the shoe prints, you would have to accept that that person would have had to be, at one and the same time, cautious enough to have cleaned up after the attack (and even, at least to a certain extent, to have "rearranged" the crime scene), but careless enough to ultimately walk out of the apartment with a clearly bloodied shoe (and the mark of his hand on a pillow). He would have had to spend considerable time and attention to trying removing his presence in some ways, whilst at the same time not caring at all whether his presence was discernible by other means. (I should perhaps at this point bring into recollection that the shoe prints leading from Kercher's room to the exit were visible to the naked eye; these were not traces only discernible by the use of Luminol.)

In other words, it seems fairly clear that the two sets of actions could not have been taken by one single person. They are, simply put, contradictory in nature. To a certain extent, the Appeal Court itself inadvertently strengthens this contradiction, when it assumes that Guede left the apartment wearing only his left shoe, with his right foot still bare. That, certainly, is the action of someone who wants to leave as quickly as possible, and not the action of someone who has decided, instead, to stay for a while and remove as much evidence as possible. Besides, it seems rather absurd to suppose that Guede would have cleaned up the place whilst wearing just one shoe.

The Appeal Court however, recognises the problem, and deals with it in a way that is quite crucial not just to the issue of the footprint on the mat but also to the issue of the traces revealed by Luminol (which I'll discuss later). It does so by assuming there was no clean-up.

I must admit that when I first realised this, I was momentarily at a loss for words. That the apartment had, at least partially, been cleaned after the attack had been, for me, a given (as it had been for the original court, which stated that "a cleaning activity was certainly carried out".)

The Appeal Court's reasoning on this cannot be found in its deliberations concerning the footprint; it is only when one arrives at the court's reasoning with regard to the traces left by Luminol that this aspect of the ruling becomes clear. There, the court expressly rejects the idea that any clean-up took place: "the occurrence of a clean-up is negated by the sheer number of traces found in the house", the court states.

I must admit I have no true understanding of how the Appeal Court might find this to be the case. If Kercher was murdered in the way she was murdered - that is, stabbed repeatedly with one or two knives - she would have lost a considerable amount of blood. Whoever attacked her must gotten that blood on them; the blood must have gotten on their (bare or shod) feet. Of course some sort of clean-up took place, and this follows from the simple fact that the very next day, various people (not just Knox and Sollecito, but also Romanelli, a friend of hers and both their boyfriends, along with two officers from the Postal Police) all entered the apartment and none of them, at that point in time, realised that they were walking into the scene of a violent murder. It follows from Knox's own testimony: she states she had earlier taken a shower and not realised anything was actually wrong; she noticed on or two little blood stains but assumed someone had had her period. And, certainly, it follows from the single footprint on the bath mat in the small bathroom, with the absence of any other prints leading from Kercher's room to the bathroom and the mat.  

In short, if the fact that some sort of clean-up had taken place seemed, to me, a given, that is because it is. I don't see how you can get around that at all; any other idea would appear to be totally illogical. Nevertheless, that is the assumption of the Appeal Court, and because of this, I feel that that court's reasoning is seriously suspect.

Now, I have already said that, if you do assume a clean-up took place, this also implies that the Appeal Court's reasoning with regard to the footprint on the bath mat becomes clearly illogical.

*******

It is only when the above issues has been addressed that the next question arises, which is whether or not the forensic evidence shows if the print matches Sollecito's foot. As to that question, I will try to be brief. The reason is very simple: I'm basically at a loss.

You see, the experts of the police came to the conclusion that the print was compatible with Sollecito's foot, but incompatible with Guede's.  However, the expert hired by Sollecito, professor Vinci, disagreed: he stated that the print could, indeed, have been left by Guede. To substantiate this, Vinci made one very important assumption, which is that the print of the foot's big toe is actually not a print of just that toe, but also a print of the next (or second) toe. The print is therefore merged; it's two toes. This is important because it turns out that Sollecito's big toe is actually much bigger than Guede's. In other words, to explain the size of the print when it comes to the big toe (to explain that the print could nevertheless be Guede's), the defence had to assume that the print was two toes.

Now the original court rejected Vinci's claim, using a rather simple argument: it stated that the print was clearly homogenous. If it had been the merged print of not one but two toes, there would have had to have been some incongruity, some "interruption of continuity" (as that court states) in the print. However, there is none, the court alleges.

The Appeal Court , however, states the opposite, where it confirms that "examining the black and white images [in professor Vinci's report], one remains convinced of the validity of his belief."

And that's basically it. In other words, the original court looks at the print and decides it's Sollecito's foot, whilst the Appeal Court looks at the print (the photos in Vinci's report) and decides that it could well be Guede's.

As a result, I have no clear idea of which court's opinion is the more valid. I can only turn to a few rather vague ideas and notions. Whilst these, by and large, would seem to indicate that, yes, it's rather more likely that we're talking about Sollecito's print than Guede's, that's just not enough.  

As to what the Court of Cassation made of this, once again I don't know. From a strict legal point of view, I would have to assume that that court might well have had to accept the Appeal Court's reasoning, since it doesn't seem to be overtly illogical or insufficiently substantiated. 

Please remember though, that we are only talking about possibilities (or probabilities). Neither the original court or the Appeal Court rules out in any absolute sense that the print could have been made by either Sollecito's or Guede's foot. As a result, the remarks I made earlier - regarding the relevance of the print when it comes to the question of a "lone wolf" criminal - remain quite intact, as do my comments regarding the clean-up. And these, to my mind, are the central issues.

*******


Sollecito's foot
Guede's foot



 


Click on either image for a larger picture.
As to the footprint on the mat, here are the measured sizes: Big toe height = 39. Big toe width = 30. 
Metatarsus height = 50. Metatarsus width = 99. 
(The photos are from the prosecution's report.)


*******

The Luminol prints

As the Appeal Court correctly states, during the second inspection of the house at the Via della Pergola, on the 18th of December, 2007, Luminol was applied to the floor of the corridor, to the kitchen/living-room, to the bedrooms of Knox and Filomena Romanelli, and, finally, to the larger bathroom.

Now, Luminol is a substance that can detect the (invisible) existence of blood, but it reacts equally to other substances, like fruit juice or bleach.

The use of Luminol revealed various traces (I believe there were six traces in all). These traces were further examined; that is, they were, firstly, examined as to whether it could established that they were left by blood. Secondly, DNA samples were taken and tested.  

As for the blood sample testing: all the results came up negative. It could not, in other words, be established that the traces were blood traces.

As for the DNA testing: two traces were found to contain the mixed DNA of both Knox and Kercher. One of these traces had been taken in Romanelli's room; the other had been taken in the hallway. The other four traces did not contain mixed samples, but they did contain the DNA of just Knox.

The original court, in considering this outcome, came to the conclusion that in the case of the two mixed samples, a reasonable supposition is that these traces consisted of the blood of Kercher, mixed with the DNA of Knox. In doing so, that court takes into consideration that whilst the testing done did not establish the existence of blood, there was no sufficient reason to exclude the possibility that they had, nevertheless, been blood samples.

The Appeal Court disagreed, stating the following: "First and foremost, the certain, true fact is that the generic blood test gave a negative result. According to the [first] Court this happened because of the scarcity of the available biological material, but the consultant for the defense, (...) specified (....) that the (...) test is very sensitive, so much as to give a positive result even with only five red blood cells present. Dr. Stefanoni herself, moreover, clarified (...) that, while a positive test result could be deceptive due to reactivity of the chemical with other substances, a negative result gives certainty that no blood is present."

The first thing that should be noted here is that the Appeal Court has muddled its facts. Dr Stefanoni (who is attached to the Scientific Police and therefore the prosecution) was, at the time, clearly talking about the results from the use of Luminol, not the results from specific and subsequent blood tests. And her remarks seem correct: a negative Luminol test will no doubt be conclusive (i.e. if the Luminol does not cause a reaction, well then, no blood, or any other substance that reacts to Luminol, will be present). This, however, clearly does not apply to the six traces in question, since all these traces led to positive Luminol results; there was a reaction. 

What we are left with is the true statement that the specific blood test lead to a negative result. Here, however, the same applies as when examining the "double-DNA" knife and the bra clasp: the fact that it could not be established that Kercher's DNA sample on the knife was left by blood, or Sollecito's sample on the clasp was left by his handling of the clasp, does not in itself rule out either possibility.  It simply means that the answer to that question must be sought elsewhere, by different means.

The means by which the original court reaches its answer is relatively straightforward. We have, that court considers, the positive Luminol samples. Then we have, in two cases, further DNA samples that were mixed (they contained both the DNA of Knox and Kercher).  What is the most reasonable explanation for this? Well, the original court states, it's surely that Knox, in her bare feet - feet still bloodied with Kercher's blood - left these traces.

To which it can be added that one of the traces (the one in Romanelli's room) does not seem attributable to a bare foot (it is simply amorphous). The second trace, however (the one found in the corridor), does seem to have been left by a foot (indeed, it seems to have been left by Knox's foot).*

And to which it can further be added that any other explanation seems much more improbable. In particular, the notion that the two mixed traces could just be evidence of the fact that both Knox and Kercher happened to live in the same apartment seems unlikely. After all, such an idea would necessitate that , at two very specific places and at more or less the same time, Knox and Kercher both left their DNA traces, and you would further have to assume that, again at the same place and at more or less the same time, some innocuous substance with which Luminol reacts (such as fruit juice or bleach) was used. If you assume that substance to have been incidental, such as, say, spilled fruit, common sense dictates that the odds of this having happened become very high indeed; if, alternatively, you presume a much more generic substance caused the reaction, such as bleach, the opposite apples; it becomes almost impossible to understand why only two traces were found.

In other words, the original court reached its decision on the basis of a fairly logical set of thoughts, and one that I, at least find compelling.

It is interesting to note that the Appeal Court seems to have all sorts of difficulties in trying to reach an opposing view. The court starts by assuming the Luminol traces could indeed have been caused by bleach. After all, why not? There were four young women living together in the apartment, who must have had friends coming over regularly; bleach must have been used often. In doing this, however, the Appeal Court is already starting to miss the crux of the matter; after all, it is not just the positive Luminol test that is indicative, it is the combination of that test and the DNA sampling, and the fact that so few traces were found, which is important.

Perhaps because of this, the Appeal Court goes on to make two further remarks. Firstly, it states the following: "The limited number of footprints detected can be explained by treading at different times and by the use of bleach on points specifically dirty.  After all, doubts similar to those raised in the ruling could be brought up even supposing that the traces are of blood: why only in those few points, moreover not consecutive and instead spread in various rooms?"

This quotation is, it must be said, truly odd. If the traces were a reaction to bleach, surely there must have been a lot more of them. So the Appeal Court is trying to explain why there were, in fact, so few. And in doing so, it immediately ties this issue to another one altogether, which is the question of why only two mixed DNA samples were found.

Well, that question has already been answered by the original court, and that court's answer was simple: a clean-up operation had taken place. So one would expect the Appeal Court to concur, this, after all, explained the conundrum the Appeal Court itself had just made obvious.

What, however, does the Appeal Court do? It immediately and rather astonishingly rejects the idea of a clean up. The second remark it makes is: "the occurrence of a clean-up is negated by the sheer number of traces found in the house".

It is at this point that one must seriously consider that the Appeal Court has more or less lost the ability to think straight. The court starts its arguments by asking itself why there are so few footprints. It links this to the fact that, well, there are only two mixed DNA traces, too. It then offers the answer to the question of how this can be the case: there must have been a clean-up. It then rejects that notion. And how does it do this? By pointing to "the sheer number" of traces.  I think the only correct response to all this would be: "Eh?"

Having said this, however, I must point out that the Appeal Court, after this momentary lapse of reason, comes back quite strongly. It finishes its arguments on the traces left by Luminol with a simple and possibly adequate remark, stating that the DNA in the mixed samples was once more so small that LCN testing was required. And just like the testing done with the knife, serious questions can be raised as to the reliability of the results procured. Once more, the question arises if the tests conducted by the Scientific Police were adequate since, once more (as the Appeal Court states), just a single test was carried out (and, as far as I know, no new testing - done, perhaps, on the basis of methods "still in development" - can be done). 

So, how will the Court of Cassation have thought about all this? Well, at the very least I would assume that it could not have overly pleased with the Appeal Court's considerations; these are, in many ways, clearly illogical in very substantive ways.  Of course some sort of clean up seems to have taken place; the Appeal Court's negation of this makes very little sense at all. So of course there is the very real question as to whether Guede could and would have engaged in this clean up and if not, if that fact alone does not clearly point to the guilt of Knox and Sollecito. And of course the traces left by Luminol should be taken seriously in this regard. And yet, there remains the unclear answer to the question of whether or not the footprint on the mat should or should not be attributed to Sollecito. And there remains the question of whether the results of the LCN testing on the mixed samples can or cannot be considered reliable. And I am uncertain as to how the Court of Cassation has answered either of these questions. 
 ________ 

* This paragraph was edited on the basis of the comments you can read below. Originally, I had stated that both prints were attributable to bare feet, thereby following the Appeal Court's statements. The Appeal Court seems to have been wrong, though.

Sunday, May 5, 2013

3. The Meredith Kercher Case - Back to the Drawing Board (Part Three)


Huh! There's gotta be some contamination somewhere!



Introduction

I started out this new series because the Italian Court of Cassation recently quashed the Appeal Court's acquittal of both Amanda Knox and Raffaele Sollecito.

In Part One, I discussed the break-in, the Appeal Court's findings on this and the possible way in which the Court of Cassation might have viewed that court's findings. In Part Two, I discussed the DNA evidence with regard to the so-called "double-DNA" knife, again attempting to guess what the Court of Cassation made of the Appeal Court's decision on this matter.

This time around, I'll be looking at the DNA evidence with regard to the bra clasp. After that, I'll make a few remarks concerning the possibility of the DNA results deriving from contamination.

The bra clasp

Just to re-iterate: the experts appointed by the Appeal Court maintained that (a) it is not clear that the DNA found on the clasp was attributable to any handling of the clasp by Sollecito, (b) it is not clear that it actually was Sollecito's DNA in the first place, and (c) the DNA could be the result of contamination.

As I did when discussing the knife, I'll focus on the central tenet of this argument: it isn't clear that the DNA found belonged to Sollecito. However, I'll start with the experts' first point: it isn't clear that Sollecito's DNA derived from his handling of the clasp.

My remark here is quite similar to the one I made when it came to the knife, and the notion that it cannot be established that Kercher's DNA (as found on the knife) was derived from blood. The experts' comments may well be correct, but are essentially meaningless. That it can not be established that Sollecito's DNA was due to his handling of the clasp in no way excludes the pertinence of his DNA being present in the first place.

So, on to the real question: was Sollecito's DNA actually found on the clasp? 

Well, the experts - and, subsequently, the Appeal Court - place a lot of emphasis on the fact that the DNA sample on the clasp was "mixed". That is: the sample contained the DNA of more than one person. The "major contributor" to the DNA was obviously Kercher herself, but besides hers, the DNA of at least one other person was present.

According to the original Court of Assizes, that person was Raffaele Sollecito. However, the experts dismissed this, insofar as they rejected the notion that the sample contained only the DNA of Kercher and Sollecito. Instead, they maintained, the DNA was that of Kercher and furthermore pointed to several males, one of whom could, indeed, be Sollecito.

Now, in itself, such an assertion may seem rather innocuous.  The question is not, after all, whether "other males" ever touched Kercher's bra; the question is whether Sollecito's DNA was present. And it would appear that the experts are at least partially acknowledging that it could well be Sollecito's DNA.

However, this part of the report has lead to rather surprising results. Firstly, it allowed the Appeal Court to not just repeat the same views on proper procedures and protocols that it had already made in the case of the knife, but to elaborate on them. "Therefore," the Appeal Court says at one point, "(...) it is easy to understand that when a mixture is present (a sample with several contributors) the problem of tracing a specific profile – that is, of identifying the peaks to pair in the graph, distinguishing them from those which are irrelevant and from other pairings – is particularly complex, since it is very often not possible to exclude alternative but equally plausible pairings."

In other words, the Appeal Court felt that the Scientific Police failed to act properly, not adhering to the (even stricter)  procedures and protocols applicable in the case of mixed samples. Hence, the result obtained is unreliable.

At this point it could be noted that the Appeal Court falls into the same trap as it did when it came to the knife: it does not look at this particular case and then try to determine the reliability of the test on the basis of the specifics of all the relevant factors involved; instead, it holds this test up to the general standards, as generally accepted by various experts in the field. The gap I pointed out when discussing the knife returns, perhaps even more strongly, here.

Rather astonishingly, though, the Appeal Court's reasoning does not end here. It furthermore takes a number of remarks of one the experts (as made before the Appeal Court orally) into consideration. After the deliberation quoted just above, it goes on to say: "And this is why Prof. Vecchiotti [one of the experts] was able to say that it was fundamentally possible to identify anyone’s profile in the graph obtained from the Scientific Police, even her own."

At first glance, I must admit, this view would seem to be quite nonsensical. I readily admit I have little specific knowledge of DNA or the way in which it is or should be tested, but the idea that the testing done by the Scientific Police was so shoddy (or, perhaps, that the DNA sample was just so contaminated) that basically anyone's DNA could have been obtained seems to me to be quite incredible.

What the expert, in her oral testimony, probably meant to say is that, since this is a mixed sample, various bits and pieces were found in the results that could, in theory, be combined in a great many different ways, thereby procuring the DNA profile of a variety of possible (that is: theoretical) people. However, such a remark, if made, would seem to be somewhat misleading. Ultimately, it will of course be up to the person interpreting the results to decide if a person's DNA is or is not actually present, but the interpreter will certainly, first of all, need to find a full profile (bits and pieces are insufficient). Sollecito's full DNA was found in the result (as the experts expressly acknowledge); besides Kercher's, no-one else's DNA (that is: a full profile of DNA) was found. In theory further DNA profiles could also be put together that would lead to "someone else", but that person would be purely hypothetical: he or she would simply not exist. The only way to get around this - to nevertheless find the DNA of an actual person - would seem to be to accept a less than full profile; to just use, in other words, some bits here and some pieces there. This, of course, the Scientific Police did not do, and neither would anyone else, including the experts.

All this is, I would guess, what the expert was referring to. However, the Appeal Court seems to have misunderstood her, and thereby quite possibly have totally misrepresented the experts' actual findings.

Contamination

After having remarked on the reliability of the DNA test result, the Appeal Court quickly moves to the issue of the possible contamination of the sample. The court deliberated similarly with regard to the knife, so the issues regarding contamination can best be discussed together.

There is a preliminary remark to make here, though, and it's an important one.

When I talk about contamination, I am not talking about the reliability of the DNA sample itself. What I am talking about is the question of whether the sample, though reliably attributable to a single person, should nevertheless be disregarded due to the fact that it existed where it did due to contamination.

An example may explain this. A sample may well be severely impaired due to, say, bacteriological damage. It may then become difficult to ascertain whether or not a match can definitively be made. You could, in such a case, say that the sample had been "contaminated" by bacteria. However, that is not what I'd understand the specific term of "contamination" to mean. Alternatively, you could have a perfectly fine sample, and yet have to dismiss it because, simply put, the DNA could very well have been inadvertently placed there by a police officer wearing dirty gloves. Now that, in my book, would be contamination.

I mention this because the distinction I make does not seem to be the same distinction made by the experts appointed by the Appeal Court. Indeed, I am not quite sure what definition of the term the experts are actually using; theirs seem to be a somewhat imprecise term. (To be fair, though, my own interpretation may seem practical, but could well be a little dodgy scientifically!)

Having said this, though, I must immediately remark on a clear flaw in the experts' report. The report extensively deals with contamination in general, and the possibility of contamination having happened in the laboratory of the Scientific Police in particular. However, it seems the experts were unaware of the fact that the Scientific Police employed certain "negative controls" during their handling of the samples in the laboratory, and the use of these controls greatly decreased the risk of any contamination having taken place there. To put it somewhat more simply: it seems as if one may reasonably disregard the remarks made by the experts concerning the serious possibility of contamination having taken place inside the laboratory.

What we are then dealing with is the question of whether anything went wrong during the period before or during the collection of either the knife or the bra clasp, and up to the moment the two exhibits entered the lab. What we are talking about, therefore, comes much closer to my own definition of contamination.

So, did anything go wrong? Well, yes. Does this mean that the knife and the clasp probably got contaminated? Err, perhaps. Perhaps not.

Let's start with the knife. The knife was found by a police officer in Sollecito's apartment. The officer looked in a drawer in Sollecito's kitchen, saw this shiny knife, and apparently thought "Hey, this could be the murder weapon". He popped it into a bag and took it away. It was tested and, hey presto, it was the murder weapon. All this seems, to be frank, rather too good to be true (and, hence, one cannot help but be a bit suspicious), but it does not in any way indicate that the knife was actually contaminated.

After all, how could it have been? If the contamination didn't happen in the laboratory (and, as I have said, it seems that we must accept that it didn't) then how could Kercher's DNA have found its way onto the knife? The only answer would seem to be that, somewhere during the process of collecting the knife as evidence, someone who handled it had Kercher's DNA on his or her hands. However, there just doesn't seem to be any real possibility of that having happened (at least, not inadvertently, and therefore in good faith). Some possibility of contamination still remains, of course - you cannot, after all, rule out all theoretical possibilities - but it becomes quite implausible.

As for the bra clasp, the story is rather different. In my earlier series describing the original Court Of Assizes ruling (that is, The Verdict in Perugia series), I stated the now almost infamous fact that the bra clasp was not secured as evidence for some 47 days after the murder. Instead, it seems to have remained somewhere in Kercher's room. When it finally was procured, it was, perhaps, mishandled (picked up from the ground, dropped back down, and then picked up again, by officers who perhaps were wearing gloves already worn in a different room). Even worse, it was found in a different place than where it had been directly after the murder; somehow, it had been moved. Besides this, the experts, as stated above, found that the clasp contained not just the Kercher's DNA and that of Sollecito, but also the DNA of "other males". This certainly seems to be an indication of sorts that the clasp may well have been contaminated.

The question, of course, is how the Court of Cassation looked at this matter. 

To understand the this a bit better, though, you should first be aware of a legal issue that arises here.  That issue is simple: who has to prove what? The prosecution, perhaps not surprisingly, has argued that whilst it might need to prove the fact that the samples found on the knife and on the clasp were, indeed, those of Kercher and Sollecito, it need not prove the absence of possible contamination. The defence, on the other hand, maintained that discounting the real possibility of contamination was simply part of the prosecution's task of proving "without a reasonable doubt" that Knox and Sollecito are guilty.

This is, it should be said, very much the sort of issue the Court of Cassation is equipped to deal with, and, I would surmise, very much the sort of issue the that court has dealt with many times in the past. I would guess that its ruling in this case will not favour either the prosecution or the defence unequivocally. Instead, I suspect the Court of Cassation will have thought that it's a bit of both.

I would, more specifically, suspect that the Court of Cassation may state that it is the prosecution's task to full disclose any and all information with regard to any possible contamination (i.e., to state clearly and simply what the police did or did not do with regard to the knife and the clasp), and that it is then the task of the defence to establish whether, given the information provided, any real chance of contamination existed.

If that is just about right, I would expect the Court of Cassation may well have decided broadly along the following lines: the bra clasp was correctly disregarded by the Appeal Court;  the defence has adequately established that a very real danger of contamination existed. The knife, however, cannot be considered contaminated; as such, the Appeal Court's decision cannot stand, and this will be a matter that the new appeal court will have to take up again. 

Again though, there are all sorts of other permutations possible. And once more, we'll just have to wait and see.


Thursday, May 2, 2013

2. The Meredith Kercher Case - Back to the Drawing Board (Part Two)



Italy's Court of Cassation 



Introduction

I started out this new series for the very simple reason that the Italian Court of Cassation recently quashed the Appeal Court's acquittal of both Amanda Knox and Raffaele Sollecito.

I then went on to point out a number of issues with regard to one of the most intriguing aspects of the case: the question of whether or not the break-in at the house at the Via della Pergola was, or was not, staged.

I will now turn to the most complex aspect of the trial: the DNA evidence. By this I mean both the DNA samples found on the so-called "double-DNA" knife and the sample found on Kercher's bra clasp.

In this second part,  I will be looking in greater detail at the Appeal Court's decision to appoint new experts, and will discuss that court's reasoning with regard to the DNA found on the knife. In Part Three, I will be focusing on the sample found on the bra clasp, and I'll make a few remarks about the idea of any of the samples being attributable to contamination in general. As always in this series, I will try to give some sort of indication of what the Court of Cassation  may have felt on the various issues raised.

A summary of the Courts' rulings

To put it very simply, the (original) Court of Assizes heard the arguments made by both the prosecution and the defence lawyers with regard to the DNA samples, the way those samples were tested and the reliability of the results procured; it then made its decision. Its decision was that the results obtained by the Scientific Police were reliable. That is: DNA found on a knife were attributable to both Knox and Kercher, whilst DNA found on a bra clasp discovered in Meredith's room was attributable to Sollecito. In other words, the knife must have been used to attack Meredith, whilst Sollecito must have been present when Meredith was attacked.

The Appeal Court, however, did not wish to make any similar assessment. Instead, it decided to appoint two new experts. It asked these experts a number of questions. Firstly, could the testing be done again? Secondly (if a new test was impossible), how reliable are the results from the original tests?

The experts answered as follows: no new testing can be done, whilst both the results with regard to the knife (in particular, the assertion that Kercher's DNA was present on the knife) and the bra clasp (that is, that Sollecito's DNA was present on the clasp) are to be considered unreliable.

In the case of the knife, the experts reached their conclusion because (a) it cannot be established that Kercher's DNA derived from blood; (b) the testing was done using a method known as Low Number Copying (LNC), whilst the internationally accepted precautions applicable to such a method of testing were not followed, and (c) it cannot be ruled out that the DNA found was the result of some form of contamination.

In the case of the bra clasp, the experts found that (a) it is not clear that the DNA found was attributable to any handling of the clasp by Sollecito, (b) it is not clear that it actually was Sollecito's DNA in the first place, and (c) this DNA, too, could be the result of contamination.

The Appeal court thereupon accepted the experts' conclusions and, in doing so, effectively disregarded the original court's findings.

Why appoint experts?

There are a number of things to be said with regard to the Appeal Court's approach to this matter.

The first of these is simply the question why the Appeal Court decided to pose these questions to new experts. After all, the experts for both the prosecution and the defence had argued their respective positions extensively before the original court. It is because of this that the original court decided to ignore a similar request made by the defence at that time. So why did the Appeal Court think otherwise?

To be frank, the Appeal Court's reasoning in this regard escapes me. In its ruling, that court attempts to set out that it's not just appointing new experts to answer questions already answered by the original experts (a situation which could obviously be repeated ad infinitum until the desired outcome, whatever that may be, would be achieved). It states that its decision is grounded in "legal reasoning" and that it's a question of "compensating for understandable gaps in knowledge in a particularly complex field".  

However, the questions the Appeal Court thereupon asks the experts are difficult to see in any other light than as a repetition of the work carried out by the experts before the Court of Assizes. Simply put, the Appeal Court requests the new experts to choose between the opposing views of the  prosecution and the defence. In doing so, it may well have gone further than is admissible under Italian law, and it may well have handed the decision with regard to the DNA evidence (a decision which the Appeal Court itself is required to make) into the hands of experts.

This, then, is the first issue with regard to the Appeal Court's ruling. As serious as it might seem in theory, though, it need not be all too important in practice. Firstly, the Court of Cassation could have felt that the decision to appoint new experts was ultimately justified, even if the way the Appeal Court went about it was not. And it might have thought that even if the appointment were not justified, the Appeal Court itself remedied the situation by sufficiently motivating why it found the findings of the experts correct.  

There is another aspect here, which is simply practical in nature. The fact of the matter is that, justified or not, new experts were appointed and they did answer the questions submitted to them. That's not something you can readily undo. It seems more or less impossible, for example, to take the experts' report totally out of the equation; any new appeal court, in handling the matter, will know of its existence.  

The Appeal Court's reasoning with regard to the knife

The second issue with regard to the Appeal Court's handling of this matter deals directly with the Appeal Court's (and the experts') views on the reliability of the DNA samples. Were the original results of the DNA testing accurate or weren't they?

Well, let's start with the samples found on the knife. No problem arises with regard to the Knox sample; all experts agree that the sample was accurate. Grave problems arise, however, with regard to the sample attributed to Kercher.

The main problem is that the sample was very, very small, and, as a result, LCN testing was done. Now, LCN testing is, or certainly was back in 2007 (when it was originally carried out in this case), somewhat controversial. It is only done is certain countries (Italy, obviously, but also countries like the UK). It is not entirely clear (that is, it is not generally accepted by the scientific community at large) what precautions should be taken - what protocols should be followed - in carrying out this type of testing.

In particular, the Appeal Court (and the experts) felt that reliable results from any LCN testing can only be obtained through multiple tests (or "runs"). The Scientific Police however, only conducted a single test. A single test that, according to the court and the experts, could not be repeated, since the DNA sample was used up in the testing.

As a result, court and experts felt that the results obtained from the test are unreliable.

There are, it seems to me, a few things to be said about this reasoning.

Firstly, I am not immediately convinced of the absolute necessity of conducting multiple tests. In general, I can well understand that, yes, multiple runs should be done, but I am not convinced the result from a single run should always be considered invalid. And I find the Appeal Court's reasoning on this matter - which effectively only points to views held by various experts in the field in general - rather lacklustre. The Appeal Court could, for example, have pointed out how high the chance is that the result from a single run could erroneously point to the wrong person in this particular case.

The Appeal Court, however, did not do so, leaving, to my mind, a serious gap in its reasoning (a gap which, it could be added, also appears in the experts' report). The problem here is simply that the Appeal Court seems to disregard the actual testing result itself; it does not ascertain to what extent that result (bearing in mind the way the testing was conductive) is indicative or perhaps even decisive. I find this rather surprising, given my understanding that the result was actually fairly clear cut: the Scientific Police's evaluation was simply that the DNA tested belonged to Kercher, and not, for example, that "it might belong" to Kercher, or some other equally vague attribution. (By itself this is not all that strange. If conducted incorrectly, or simply with too little material, LCN testing will, by and large, not lead to any significant result at all. You can test the DNA sample, but no match will be found. If, alternatively, a clear match is found, it would seem to me that the test was successful, and the attribution might well be correct. It would certainly be best to repeat the testing if possible, but such a repetition may not always be strictly necessary.)

As aside, I point out the Scientific Police, prior to conducting the test, invited the defence experts to be present. They did not, however, accept that invitation.

Secondly, the Appeal Court's (and the experts') assertion that the test cannot be replicated is almost certainly false. Technology progresses, and what it possible today is not what was possible back in 2007. Certainly the prosecution has alleged that a new test could be run, using today's methods. However, the Appeal Court has rejected the idea of this, and it is difficult to see this rejection in any other light than as a general suspicion of LCN testing as a whole. The Appeal Court clearly feels that any new test can only be conducted by a method still "in development", but given the court's reasoning with regard to the single test done (effectively: it's just a single test, so we'll toss it out), why on earth not conduct it?

All in all, I feel there are serious issues to be raised with regard to the Appeal Court's decision concerning the DNA on the knife. I think the chance that the Court of Cassation found the decision  unacceptable is fairly high, and I suspect that, in one way or another, it quashed the ruling on this matter. I do not, however, know what the actual decision will turn out to be. From a formal point of view, the best thing for the new appellant court to do might well be to take up the appeal and disregard the experts' report (and the Appeal Court's reasoning, insofar as it is based on the report) altogether, but as I have said, this leads to practical issues. Conversely, the Court of Cassation might rule only that a new test should be done, and leave it at that. Between these two options there is room for various alternatives.We'll have to wait and see.

A few closing remarks can be made here. Firstly, the Appeal Court did not base its decision with regard to the knife solely on the DNA testing. It also pointed out that issues can be raised with regard to the way the knife was procured (from a drawer in Sollecito's kitchen), with regard to whether it had or had not been washed (starch granules found on the knife seem to indicate it had not been), and with regard to the simple but perplexing question of how the knife (assuming it was used in attacking Kercher) came to be at the Via della Pergola in the first place (after all, it was the original court's assertion that this was not e per-meditated murder, but rather some sort of "game" gone utterly stray). 

Secondly, as set out above, the experts didn't just concern themselves with the actual DNA testing itself. They also pointed out that it could not be established that the sample derived from Kercher's blood. This is true: a specific test carried out to try and establish this turned out negative. This does not in itself mean, however, that it wasn't a blood sample; it simply means that it could not be established as being a blood sample (and the sample may also have originated from Kercher's skin or tissue, which would have been equally meaningful). Finally, the experts pointed to the possibilities of contamination, which is precisely what I'll be discussing in Part Three of this series.


Tuesday, March 26, 2013

1. The Meredith Kercher Case: Back to the Drawing Board (Part One)




The house at the Via della Pergola


Introduction

On October 3rd, 2011, the Appeal Court of  Perugia acquitted both Amanda Knox and Raffaele Sollecito, thereby overturning the convictions handed out by the original court in 2009.

The prosecution thereupon lodged an appeal against these acquittals with Italy's highest court, the Court of Cassation.

Today, that court quashed the Appeal Court's ruling, referring the case to another appellant court - one in Florence -  to be handled further.

The Court of Cassation has not yet substantiated today's ruling. It will do so in the next 90 days. Until the ruling has been substantiated, it is impossible to know why the court felt that the appellant court had erred. It is also impossible to know whether such errors can reasonably be expected to result in a different ruling to be given by the Florence court.

In a series of articles posted here in 2011 (and called, collectively, The Trial In Perugia), I have attempted to offer some insight into the case and its legal setting. I've discussed the original court's ruling (leading to the convictions) at some length, but I have not yet spent all that much attention to the Appeal Court's findings.

Given today's ruling, it's clear that this case is far from over. It is also clear that, for reasons as yet unknown, the Appeal Court's decision cannot stand and a new verdict will have to be given.

As a result, it's extremely tempting to re-examine the case and give some very preliminary thoughts on what the Court of Cassation's reasoning might have been. That's what I'll be doing in this new series of articles. 

To start out, I'll be discussing one of the most intriguing aspects of the entire case: the break-in. Did the murderer (or murderers)  of Meredith Kercher gain entrance to the apartment by breaking into an upstairs window? Or was the break-in staged?

To my mind, the answer to these questions constitute the heart of the case. After all, if one assumes an actual burglary took place, there can be no compelling reason to assume either Amanda's or Raffaele's guilt. Conversely, if one assumes the break-in was staged, it seems very difficult to assume their innocence.

From a more legal point of view, the question arises whether the Court of Cassation overturned the acquittal (in part) because of the Appeal Court's reasoning with regard to the break-in, or whether it based its decision on different grounds. 

A final introductory word here: I'll be repeating a number of things already raised during the earlier series. I apologise; it's not to be helped.

The Red Flag Paradigm

There is a book called The Crime Classification Manual. I understand it to be a more or less standard textbook on various aspects of the classification of violent crimes.

The following is a quote from this manual: 

An offender who stages a crime scene usually makes mistakes because he stages it the way he thinks a crime scene should look. While doing this, the offender experiences a great deal of stress and does not have time to fit all the pieces together logically. Inconsistencies thus begin appearing at the crime scene, the forensics and with the overall picture of the offence. These inconsistencies often serve as red flags for staging and prevent misguidance of the investigation (...)

Several important questions need to be asked during crime scene analysis. First, did the subject take inappropriate items from the crime scene if burglary appears to be the motive? (...)

Second, did the point of entry make sense? For example, an offender enters a house by a second-story window, despite the presence of easier, less conspicuous entry points.

Third, did the perpetration of this crime pose a high risk to the offender? In other words, did it happen during daylight hours, in a populated area, with obvious signs of occupation in the house (lights on, vehicles in the driveway), or involving highly visible entry points?

There is, in other words, a red flag paradigm. If one were to apply this paradigm to the case of Meredith Kercher, what happens?

Well, one might firstly remark that, in our case, nothing was actually stolen. A window was broken in one of the girl's rooms on the second floor of the house. It was Romanelli's room, and that room had clearly been ransacked. Nothing, however, had been taken. No jewellery or other small valuables (which could easily have been slipped into a pocket or bag) were missing; there was a laptop, but it wasn't taken either.

That, it seems to me, is the first flag waving.

Secondly, there is the point of entry. It seems to make little sense to enter the apartment via Romanelli's window, since doing so would have been cumbersome and such an entry could easily have been seen by anyone passing on the street outside. It should be remembered that a much less conspicuous entry point was available via the balcony at the rear of the house, a fact that Rudy Guede (assuming he was the burglar) would have been aware of. More important still, however, is the question of why Guede would have wanted to break into the second-story apartment in the first place. If he was going to burgle the house, why not try the ground floor first?

Another red flag, surely.

Thirdly, the high-risk factor. A great deal has already been said about the risks - and difficulties - of gaining entry through Romanelli's window, but I'll nevertheless repeat some of the more interesting points here. First, however, another observation.

If Guede broke into the apartment when Meredith was already home, it seems likely that he must have noticed that the lights were on in the apartment. This alone would seem to make any burglary attempt prohibitively risky. After all, breaking into a second story apartment when one might more easily and safely break in on the ground floor seems curious enough, but doing so when the lights are on upstairs but not downstairs is not realistically conceivable. In other words, this alone might well cause the third red flag to be hoisted.

There is, however, more to consider. It should be remembered that the break-in would have required Guede to climb up to Romanelli's window to open the shutters, to then climb back down to throw a rock through that window, and finally to climb back up to enter the apartment. (It could be argued that this "double-climb" may not have been necessary if the shutters were open to begin with, or if they could have been opened from the terrace on the second floor, but these are notions the appellant court does not seriously entertain.) Given the location of the window and the time of the break-in (roughly nine o'clock, or perhaps a little later, at night), there were obvious risks involved. The endeavour becomes even more fraught with danger if one considers that the burglar, when first attempting the climb, has no way of knowing whether he will actually be able to open the window at all, since he wouldn't know if the shutters would open (and, even if he thought they would, he wouldn't know whether the window might have the added protection of additional inside shutters, or "scuri".)  As for the climb itself, the burglar probably was helped by the presence of a nail attached to the wall which he used as a support. Again, though, he probably would not have been able to see the nail from the ground in the dark, let alone know that it would hold his weight.

In other words, such a break-in seems very risky indeed. This warrants the third red flag, I'd say.

So, all told, there seem to be clear initial signals that the break-in was staged.

Timing

A major problem that furthermore arises with regard to the burglary is one of timing. It is an issue that, to my mind, has not been dealt with conclusively by the Appeal Court, but that nevertheless seems vital. The question is: when could an actual burglary have taken place? And to phrase this a little more relevantly: did Guede enter the apartment before or after Meredith Kercher had returned home?

If Kercher was already in the apartment when Guede came in through the window, it seems impossible that she would not have been aware of this. She would have heard the window breaking and Guede climbing inside; she would have heard him ransacking the place. And if she were aware of the break-in, she must have reacted to it in some way. She might, for example, have locked herself in her bedroom and called the police. Alternatively, she might have left her bedroom to see what on earth was going on. None of these things happened, however; Kercher was attacked in her own bedroom (the door had not been forced open) and she put up no struggle to speak of. No call was made to the police.

Then there is the problem of the bathroom. We know that Guede used the larger of the two bathrooms (situated fairly close to the front door), since he didn't flush and his faeces were found in the toilet. If he entered the apartment when Kercher was already home, one would  assume that he went to the bathroom after having attacked and killed Kercher. That doesn't make sense, though, given the fact that, at that point, Guede would have fled the apartment as quickly as possible (to use the appeal court's own words). He certainly wouldn't have taken the time to use the bathroom, but not the time to steal whatever he could.  There is, besides,  the fact that traces of shoeprints were found, leading away from Kercher's room and towards the front door; as far as I know, this trail of prints seems to exclude the possibility that the person walking made a detour towards the bathroom.

If, alternatively, Guede broke in when Kercher was not yet home, one must assume that, for whatever reason, she did not at first realise that a burglary was taking place. This is again hard to imagine, but it might possibly be explained by assuming that Guede, having entered the apartment and starting to ransack Romanellli's room, decided to use the bathroom, in which case Kercher may not initially have been aware of his presence. However, in such a scenario, one would assume that Guede would in all likelihood have turned the lights on in the apartment, which in itself might have been enough to warn Kercher. Besides this, there is the question of Romanelli's door. Was it closed or open when Kercher returned? Assuming it was closed could explain why Kercher didn't realise the room had been ransacked, but why on earth would Guede have closed it when he went to the bathroom? If it were open, and lights were on inside, how is it possible that Kercher passed right by it and didn't realise it had been ransacked?  (As an aside, I mention that, according to Amanda Knox's testimony, Romanelli's door was closed when Kercher's body was found the next day.)

Furthermore, one would have to assume that Guede decided, once Kercher returned, to confront her. Given the location of both Romanelli's room and the bathroom, he could (if in either of these rooms) have opted to flee out of the front door relatively easily. Instead, he must have walked down to corridor to Kercher's bedroom, where the attack took place. In this scenario, therefore, one must also assume that Guede turned from a small-time burglar into a sexual abuser and murderer, and one must assume that this transformation occurred almost instantly and by happenstance (the trigger, so to speak, being Kercher's return). I'm no psychologist, but I do find this unlikely.

Then, too, there is the notion entertained by the Appeal Court court that Kercher was already partially undressed when she was killed. How does that fit in with the notion that Guede was already in the apartment when Kercher returned? Are we to assume that he waited for her to start undressing before confronting her? But why on earth would he do that?  

Finally, the timing is important for a reason already mentioned. Kercher returned to the apartment round about 9 o'clock at night. If Guede were already inside the apartment, it could be pointed out that the break-in took place relatively early, at a time when people would generally speaking still be up. Given the method of breaking in and the visibility from the street, this, too, seems odd.

Traces outside

The Appeal Court court looks at the fact that no traces were found on the wall that indicated a climb. It dismisses the absence of such traces as insignificant, however, pointing to the fact that at one point during the investigation, an attempt was made to re-enact the climb. The court notes that the person acting out this reconstruction (who was, by the way, taller than Guede) "easily succeeded in his attempt at reaching the window for demonstrative purposes, without bending the nail or leaving other traces on the wall".

Now, there are various things which could be said about this reconstruction. Perhaps the most important one would be that it's just too vague to be of any real importance. "Reaching the window", for example, is not the same thing as "reaching the window in such a way as to be  able to climb into the apartment". The aspect that interests me here, though, is the fact that the reconstruction left "no traces", as the Appeal Court notes.

The reconstruction took place during the day, and there were obviously a number of people present. Clearly the ground was dry, or some traces of earth or mud would have been left on the wall.

The actual climb, though, would have taken place early at night, when it was dark. Furthermore, it had rained the day before, and the ground would have been "very wet" (as both Romanelli and her boyfriend attested to during the original trial). In other words, there are reasons to assume that whilst the reconstruction may have been "traceless", the actual climb would not have been (and this is compounded by the fact that the climb would have taken place not once, but twice).

It should also be noted that the police did not find any traces on the ground. There were, for example, no shoeprints underneath the window. Neither was any glass found. The latter is, by the way, addressed by the Appeal Court. It states that "the process of throwing the rock (...) did not make it necessary for some glass to end up on the outside as well as the inside of the room", and leaves it at that. This does not appear convincing, especially when one considers that the burglar, having tossed the rock and climbed back up the window, would certainly have had to remove bits and pieces of glass in order not to hurt himself when climbing inside. He would have had no reason whatsoever to make sure that each and every piece ended up inside the room instead of outside, on the ground. So even if "the process of throwing" would not in itself have left glass shards outside on the ground, the entry into the apartment, in all likelihood, would have.

Traces inside

"The smoking gun", the Appeal Court finds, for the original court's ruling regarding the staging of the break-in is the position of the glass inside Romanelli's room. The glass was on top of all sorts of objects which had been moved during the ransacking, and that meant, according to the original court, that the window had been broken after the ransacking took place.

The Appeal Court rejects this on the basis of a very simple supposition: the room was disturbed by a number of people after the burglary was discovered, and the "jumble" recorded - including the presence of glass on top of objects - could well be the result of that disturbance. It does not, according to the Appeal Court, constitute convincing evidence that the window was broken after the room had been ransacked. This is, to be frank, an extremely summary bit of reasoning, especially give the original court's rather extensive thoughts on the matter. 

The appeal court's further substantiation

When it comes to the break-in, there are a few further aspects of the Appeal Court's verdict which should be mentioned, if only to complete the picture.

There is, for example, the simple assertion that Guede could have broken into the apartment because he was "agile and athletic enough to play basketball". Again, this seems to be an overtly summary observation. There is also the reference to the fact that Guede had wounds on his right hand when he was arrested. This is also curious, because if he was wounded while breaking into the apartment, traces of blood would have been left on or near the window. No such traces were found, though.

Then there is the reference to Partick Lumumba. The appeal court notes that Amanda Knox's (false) implication of Lumumba in the crime does not fit with a staging of the break-in, but its only argument is that Lumumba was "not at all suited" to be cast in the role of a burglar. That may well be true, but if the only alternative may have been Guede himself, it's easy to see why Knox would have chosen Lumumba. It could be pointed out that all implications regarding Lumumba in the crime were clearly hazardous (after all, Lumumba was innocent, and it turned out his innocence could be proven).

Grounds for the Court of Cassation?

On the basis of the above one might well consider the Appeal Court's reasoning with regard to the break-in to be suspect. All in all, it seems much more likely that the break-in was staged than that Guede actually committed the burglary in the manner the appellant court finds feasible.

That does not, however, mean that the Court of Cassation will have quashed the appellant court's decision for this reason. In fact, this seems, at first glance, to be quite unlikely. 

The reason for this lies in the Court of Cassation's legal remit. In short, that court has to decide on matters of law, and not on factual matters. It is, in other words, not up to the court to weigh the facts and then decide whether it's more likely that the break-in actually occurred or whether is was staged; its only task here is to consider whether or not the appellant court's reasoning stands the test of being sufficiently substantiated.  

And generally speaking, that does seem to be the case. The Appeal Court's decision in this regard might not be particularly strong - it could be considered downright weak, even - but that doesn't mean it's necessarily wrong. It doesn't mean its reasoning is incomprehensible, let alone nonsensical.

Nevertheless, there are two aspects of the Appeal Court's decision which could well have led the Court of Cassation to quash its ruling.

The first and perhaps most important of these deals with the way the Appeal Court has handled the break-in in general. Effectively, it looks at the original court's findings and rejects them. It does not, however, offer its own exposition of events. It does not, for example, make any clear observations with regard to the timing of the burglary (was it before or after Kercher came home?). It does not take into account the events that occurred right after the burglary. It does not deal in any detail on the (compounded) unlikelihood of the burglary having happened as it must have (the double-climb theory). As such, the Appeal Court's decision contains rather obvious gaps; the Court of Cassation could well have found that these gaps are too severe for the acquittals to stand.  

The second possibility is a little more technical. It deals with the broken glass inside the room. The Appeal Court finds that the room was a "jumble" and that there was glass both under and above items of clothing and other objects. From this, it deduces that an actual break-in may well have happened. However, in doing so, it glosses over several pertinent parts of the original court's deliberations. For example, the original court based its opinions, in part, on Romanelli's testimony. Romanelli testified that, when she left the apartment, the laptop was  "in its case, standing up". When she returned - after the burglary - the laptop was "underneath" and when she picked it up, there was glass on top of it. In other words: the laptop had been moved, and afterwards glass had gotten on top of it. This means exactly what it seems to mean: the glass was distributed after the moving of the laptop. There are only two ways this could have happened: either the window was broken after the room was ransacked, or, somehow, the glass was already on the floor but was "re-distributed" at some later moment.  It would seem, however, that the latter option should be excluded, at least in the case of the laptop, since I understand Romanelli was the first to enter her room.      

Given this, it might be argued that the Appeal Court ignored evidence available to it, evidence that, it should be added, directly refutes that court's conclusions. Again, it is possible the Court of Cassation found this a ground for the overturning of the Appeal Court's decision.

There is one final remark to be made here. It does not concern this particular trial, but rather the trial against Rudy Guede. That trial, too, went all the way up to the Court of Cassation. In its ruling, the court confirmed the guilty verdicts already handed down by the two lower courts, and in doing so, it makes a very interesting remark concerning the break-in:

"And it should also be noted, as the judges of the lower courts have correctly held, that following the murder an activity occurred intended to simulate an attempted theft, which the judges of lower courts and the defence of the same appellant agree was an operation done by others and not by the defendant (...)".

The key word here is "correctly". If the Court of Cassation felt at the time that the lower courts had "correctly" established that the break-in was staged (and even that it was staged "by others" than Guede), then one wonders what its current thoughts might have been.