Sunday, March 21, 2010

Viewpoints on the Italian Justice System

Part XI in a series on the Knox/Sollecito case

The murder of Meredith Kercher has caused the Italian justice system to come under scrutiny. The first is a letter Count Neri Capponi wrote about the Monster of Florence case, and the second are his comments on the Knox/Sollecito case to Judy Bachrach of Vanity Fair. Count Neri Capponi is a judge and a lawyer, and his son Niccolo is an acquaintance of Douglas Preston and Mario Spezi, authors of The Atlantic article and the book The Monster of Florence.

The travesty of justice undergone by Douglas Preston and Mario Spezi ( is the tip of the iceberg. The Italian judiciary (which includes the public prosecutors) is a branch of the civil service. This particular branch chooses its members, is self-ruling, and is accountable to no one: a state within the state! This body of bureaucrats can be roughly divided into three sections: a large minority, corrupt and affiliated to the former Communist Party; a large section of honest people who are too frightened to stand up to the political minority (which controls the offices of the judiciary); and a minority of brave and honest men with little influence. Political and dishonest judges have an infallible method of silencing or discrediting opponents, political or otherwise. A bogus indictment, the tapping of telephones, the conversations (often doctored) fed to the press to start a smear campaign, a spectacular arrest, prolonged preventive detention under the worst possible conditions, third-degree interrogations, and finally a trial that lasts many years and ends in the acquittal of a ruined man. Spezi was lucky, because the powerful Florentine public prosecutor is no friend of the Perugia prosecutor’s and, I am told, “suggested” that Spezi be freed; the Perugia court, I am told, accepted the “suggestion.”
Count Neri Capponi
Florence, Italy

The Italian legal system, ecclesiastical judge Count Neri Capponi informs me, will not work in Amanda’s favor. “Our system stems from the Inquisition and also from medieval law,” he explains. What this means, in effect, he says, is that justice in Italy “is based on the supremacy of the prosecution. This nullifies the fact—written in our constitution by the way—that you’re innocent until proven guilty.

Amanda Knox was accused of slander for her allegation of police abuse ( Based on Frank Sfarzo’s report, one is skeptical about how objectively her allegations were investigated. From a report prepared by the human rights group EveryOne on police violence in Italy, “When human rights activists report episodes of violence or abuse of power being perpetrated by rogue officers to local or national institutions, a worrying phenomenon nearly always takes place. Instead of collecting precise reports of the episodes in order to investigate and identify those responsible for the abuse, the superiors shut up like a clam, denying without question that such disgraceful acts could have taken place. They assume a threatening tone with the associations and threaten to report them for slander, libel and defamation etc. This attitude, which the leaders of EveryOne themselves have witnessed on several occasions, prevents the rogue officers being isolated and their behaviour discouraged. On the contrary, it makes them feel part of an agency in which they are allowed to act above the law using violence, threats and acts of gratuitous coercion. According to the activists, after reporting misconduct by uniformed police officers towards racial minorities, it is not rare for the activists themselves to be followed by plain clothes policemen or summoned to police stations or headquarters and “advised” not to take any further action.” (thanks to Observer for pointing out this report*)

Commentator Peter Popham has written a couple of articles on the Italian system. Comparing the British and Italian systems Popham wrote (, “One of the great virtues of the British judicial system is that, whatever ideas a detective or prosecutor may have about a case, he is not allowed to voice them until the case comes to court. And a very good thing too.

They manage these things differently in Italy, where prosecutors regularly leak their theories to the newspapers, often in extraordinary detail. Reporters compete for the juiciest tit-bits. As a result, by the time the trial comes around, the public already know what they think about a case, and why. This makes miscarriages of justice horribly likely. Take the Perugia murder: Mr Mignini made up his mind about it, and got his theories splashed across the media, in early November 2007. But weeks later forensic evidence led the police to another suspect who had little or no connection to Knox, Sollecito and their African friend. Rudy Guede, unlike the original three, was tied to the crime scene by fingerprints, hand prints and DNA evidence. In a separate trial he has already been convicted of the murder.

When Guede exploded on the scene, the investigators should have torn up their work and started again. But by this time the "guilt" of Knox and Sollecito was so well established in the media and in the public's mind that there was no going back. The jury sitting on the case absorbed all those early reports. As a result, justice may be done in Perugia next week, but I wouldn't bank on it.”

Popham also brought up ( an aspect of the Italian system that might work in Knox and Sollecito’s favor, “The good news for them is that the appeal is essentially a re-trial: every aspect of the case will be examined afresh. It is very common for convictions at the first trial to be overturned on appeal. A conviction is not considered "definitive" until it is confirmed by the Court of Cassation.” Another commentator has indicated that Knox’s incriminating statement was thrown out before her trial in a way that might not have happened until after her trial in another country.

With respect to the broad latitude given to the prosecutor for his summation, Scott H. Greenfield ( lamented, “During the summation, the prosecutor told the jury about the things Amanda Knox might have said to Meredith Kercher before the alleged drug-induced orgy that ended with her throat being slashed.

‘You are always behaving like a little saint. Now we will show you. Now we will make you have sex.’

This would be a horrible thing to say, except that it never happened. No one says that such a statement was ever made. But summations in Perugia aren't limited to evidence, as they are here. Rather, this is a permissible indulgence into fantasy, a made up dramatization of what the prosecutors contend might have happened. It's used to inflame the jury. It's what prosecutors try to do everywhere, except that there are no restrictions on such fabrications in Italy. Still, arousing passion gets a far better visceral response that appealing to reason.”

I have a relatively limited number of facts on which to base an evaluation of the Italian justice system. Broadly speaking, it has things to admire and areas for improvements. My comments here should not be construed as a blanket indictment of the Italian justice system.

*update 3/22/10

Tuesday, March 16, 2010

Luminol and some of the footprint evidence

Part X in a series on the Knox/Sollecito case

The story surrounding the footprints in the murder of Meredith Kercher is complex, and it may take more than one post to cover it. However, the recent jury’s statement included a reference to the footprint evidence as one indicator of guilt. In this post we will summarize the analyses of both pro-prosecution and pro-defense points of view with respect to a bloody footprint found on a bathmat and the three luminol-positive footprints in the hallway.

Chemistry and forensics of luminol
Luminol was used to visualize some of the footprints. Luminol is typically used as a mixture that includes sodium carbonate and sodium perborate. Luminol reacts with the iron atom in hemoglobin, and also reacts with other substances, such as bleach and fruit juice, or with substances that contain a metal ion that can catalyze the light-producing chemistry. Fruit pulp is rich in certain peroxidase enzymes that have an iron ion that has much in common with the iron in hemoglobin. In hemoglobin and in the peroxidase family of enzymes the iron ion is bound to four nitrogen atoms of an organic molecule called protoporphyrin IX. The combination of iron and protoporphyrin IX is called heme, which is tightly bound to the protein portions of hemoglobin or catalase, respectively. A fifth nitrogen atom from a histidine residue within the protein also coordinates the iron ion (Frey and Hegedus, Enzymatic Reaction Mechanisms, pp. 203-209). Heme is called a prosthetic group, a nonprotein molecule that helps a protein to do its job.

Both hemoglobin and peroxidases are proteins, but only peroxidases are enzymes (biological catalysts). In the luminol reaction however, both hemoglobin and peroxidases are acting catalytically. A paper by Barni et al., “Forensic application of the luminol reaction as a presumptive test for latent blood detection,” Talanta 72 (2007) 896–913, is a review of luminol chemistry and forensics. Hydrogen peroxide oxidizes luminol with the production of light. The iron ion within hemoglobin or other substances is a catalyst; in other words, one molecule of hemoglobin converts many molecules of luminol and hydrogen peroxide to produce 3-aminophthalate. Thus the catalytic behavior of metal ions partially explains the sensitivity of this test. Since photons of light are emitted in this reaction, the process is classified as a chemiluminescent reaction.

This paper discusses whether one can tell whether or not the substance reacting with luminol is really blood:

“Due to the possible presence of these substances at the crime scene, the luminol test must not be considered sufficiently specific to permit an unequivocal identification of blood [15,18,51,88,89]….

Generally visual examination is used when the luminol test is employed in a forensic situation, rather than instrumental detection of the luminescence. An experienced practitioner may distinguish the true blood-catalyzed chemiluminescence from that produced by other substances by the evaluation of parameters observable to the naked eye such as emission intensity, duration and spatial distribution. However this approach may also lead to misinterpretation, due to a subjective, informal and non-quantitative evaluation, for example, because its intensity is qualitatively much weaker than that expected for blood. In other circumstances an emission of similar intensity may be thought to derive from diluted bloodstains and is accepted. Therefore, caution should be exercised when using the test.”

Thus the Luminol test is a presumptive test and should be confirmed by one that is more specific for blood. However (, “The prosecution's key forensic witness told the court she can tell by looking at a luminol reaction whether it involves blood or something else, but she did not perform any scientific tests to validate this claim.”

The prosecution attributed one hallway footprint to Raffaele Sollecito. Francesco Vinci testified that the luminol footprint in the hallway is a human footprint but one that cannot be attributed to anyone in particular ( The shape of the big toe and the print of the putative second toe were key elements to his conclusions. Frank Sfarzo wrote, “With such limited elements Vinci is only able to rule out Raffaele (or the girls) and to say that the print is compatible with Rudi. And we can only admire his honesty and consistency.”

Additional posts on the footprints by Frank Sfarzo

Darkness Descending
Now let us turn to the analysis of the footprints that Colonel Luciano Garofano gave to the author of the book Darkness Descending, by Paul Russell and Graham Johnson (Pocket Books, 2009). Colonel Garofano is a well-known forensic scientist who recently retired from the Carabinieri. He was much friendlier to the prosecution’s case than to the defense’s case, and I do not agree with everything that he said in this book, particularly with respect to the footprint on the blue bathmat discussed below.

“The other problem I have is the way the Luminol was applied. The size of the blobs shows that it was not carefully vaporized but squirted. That creates two problems. It dilutes the sample and it dilates the print. We have a print attributed to Sollecito, which matches his foot in the size of the big toe, the width of the metatarsus and the width of the heel, but does not present the characteristic details each of our feet present. The print can be said to be compatible, but not 100 per cent.”

“Now let’s have a look at the prints attributed to Amanda Knox. There’s one in her bedroom facing the exit to the room, and there are two right feet in the corridor walking in the direction of the victim’s room. The same goes here. The method of application of the Luminol is insufficiently subtle to positively identify a foot, but the result can be said to be generally compatible with Amanda Knox’s.

“But I didn’t see who else they compared the prints with. Just Rudy, Amanda, and Raffaele? So we only have a choice between those? We don’t have the footprint of other women or men, as a comparison? Pity.”

Colonel Garofano believed that it is likely that Raffale’s foot made the print on the blue bathmat, but he did not sound certain. However, he did not discuss the fatal criticism that the prosecution’s expert witness made a serious error in the measurement of this print, discussed below. Colonel Garofano did not address the fact that Rudy’s big toe looks nothing like Raffaele’s, and the bathmat print. Nor did he seriously entertain the possibility that the prints are not even blood. He also did not explain why the print attributed to Ms. Knox points toward Meredith’s room and why there is not a full set of prints. These are serious omissions in that it is difficult to picture how the prints could have been made during the commission of the murder.

Science Spheres
Mark Waterbury discussed some of the problems in the footprint evidence in the hallway and elsewhere (
“Luminol glowing footprints were found in a hallway, and some may have been Amanda's, it is hard to know for sure because they were only compared with her feet, and found to be ‘compatible.’ Again, no controls. Meredith, Laura, Filomena, none of the other resident's feet were compared to these footprints. The footprints were tested for blood, and it came out negative. No blood. So, why are they important? Amanda lived there, after all.”

“Amanda's DNA was said to be found in one of these footprints. Did they also test a meter away from the footprints, to see if her DNA was all over the apartment where she lived? No. That would have been another control experiment. Was the DNA actually associated with the footprint, or did it just happen to be there, because the resident's DNA was all over their apartment, as people's DNA usually is? We will never know. They skipped the control experiments, and presented results without any reference.”

With respect to Meredith’s DNA and the luminol-enhanced footprints, I would argue that not finding her DNA is evidence against the proposition that the substance is her blood. However, finding her DNA would be not conclusive evidence that the substance is her blood. That is because there could be non-blood derived biological material from Meredith mixed into the footprint. In other words finding her DNA would have been consistent with blood being responsible for the luminol reaction but would not demand that conclusion.

Perugia Murder File
A powerpoint, Dear-Mr-Marriott-I-Shrunk-the-Black-Kid.pps, authored by Kermit ( attempted to rebut arguments presented at the Friends of Amanda website. The first of these is that the prosecution’s expert witness Mr. Rinaldi, made a serious error in his size measurements. Kermit’s presentation also addresses two of the problems in ascribing the bloody footprint to Mr. Sollecito, the narrowness of the big toe ( and the appearance of a mark near the big toe. Mr. Guede’s big toe is shorter than his second toe, but Mr. Sollecito’s second toe does not even show up in his footprint. Mr. Guede’s big toe is narrower than Mr. Sollecito’s. According to Kermit, the reason that the bloody imprint of the big toe seems narrower than Mr. Sollecito’s is that the raised tufts of the blue bathmat have picked up more blood than the unraised portions. Likewise, the mark Sollecito’s expert witness Francesco Vinci would ascribe to a second toe (, Kermit claims is Mr. Sollecito’s big toe. Kermit said elsewhere that his reason for writing his powerpoint presentations were to refute the lone wolf theory of the crime, because the prints were of varying sizes.

I do not find Kermit’s arguments about the raised tufts to be convincing. There is a raised tuft of lighter color in between the big toe and the second mark, possibly from Guede’s second toe. If Kermit were correct, that portion of the bathmat should be as dark as the rest. The validity of Kermit’s measurements has also been questioned by two commenters at a forum at James Randi Educational Foundation (

Defense-friendly blogs
Charlie Wilkes of the website Friends of Amanda presented the argument that the bathmat footprint measurements made by the prosecution’s witness, Lorenzo Rinaldi, contained a critical error. ( Mr. Wilkes also discussed the footprints elsewhere ( The site Injustice in Perugia gives a detailed account with images of the footprints ( These two sites make several telling points, especially with respect to the luminol-positive footprints in the hallway. The footprint L9 (in the knoxarchive numbering scheme) is attributed to Amanda by Rinaldi. Yet is alone, and it points toward, not away from Meredith’s room. L6 and L7 are two right feet, one of which is attributed to Raffaele. The footprint in the hallway attributed to Raffaele look quite indistinct to me. The other one is attributed to no one at all, and one wonders to whom it belongs. None of the footprints tested positive for blood and none were positive for Meredith’s DNA.

None of the three footprints were part of a trail. It is difficult to see how Amanda Knox and Raffaele Sollecito could have removed other footprints (as part of an alleged cleanup) without also cleaning up Rudy Guede’s nearby bloody shoeprints.

Prosecutorial tunnel vision
The failure to obtain reference footprints is disturbingly reminiscent of the lack of reference DNA from Laura or Filomena, the other two flatmates. It suggests that once the investigators locked onto Amanda and Raffaele, they did not reopen their field of vision. Crime journalist Mario Spezi said about the word “compatible,” in Douglas Preston’s book, The Monster of Florence, which covers a serial murderer in that city.

“Compatible, not compatible, and incompatible are the baroque inventions of Italian experts who don't want to take responsibility. Using 'compatible' is a way to avoid admitting they haven't understood anything. Was the bullet in Pacciani's garden inserted into the monster's pistol? 'It is compatible.' Was that laryngeal break inflicted by someone who intended to kill? 'It is compatible.' Was that painting done by a monstrous psychopath? 'It is compatible.'”

“Perhaps yes, perhaps no--in short, we don't know! If the experts are chosen by the investigators, they say their results are 'compatible' with the theories of the prosecution; if they are chosen by the defendants they say that their results are 'compatible' with the theories of the defense. That adjective should be outlawed!”

Luminol was overapplied to the footprints in the hallway, and a full set of reference footprints was not taken. These two points alone make the prosecution’s attribution of two the footprints to Ms. Knox and Mr. Sollecito, respectively, impossible to accept as solid conclusions. The unattributed footprint forces one to ask who else was in the hallway and when. Moreover, the luminol-visualized footprints did not contain Meredith’s DNA and cannot be said to be blood. The failure to obtain a positive test for blood is possibly the most serious problem for the prosecution’s case. Although I do not claim to be an expert in the analysis of footprints, the bathmat footprint looks like a better match to Rudy Guede than to Raffaele Sollecito. The footprints in the hallway may date from a time prior to the murder, or on the following morning. The lack of reference footprints is one more indication that the forensics in this case was not pursued in an entirely objective manner.

Monday, March 1, 2010

An interim evaluation of the evidence against Knox and Sollecito

Part IX in a series on the Knox/Sollecito case

Here is an interim overview of some of the evidence in the murder of Meredith Kercher; individual issues may be covered in greater depth once the reasons for the verdict have been announced. We have already examined some of the problems with the DNA of the knife and bra clasp. Problematic as the LCN DNA of the knife and the weak, mixed DNA of the clasp are, they are the only things that tie Mr. Sollecito to the room and Ms. Knox to a supposed murder weapon. Without them the whole case looks insubstantial.

Forensic evidence
The investigators failed to perform and especially to report the DNA forensics to the usual degree of completeness, and this leaves many questions unanswered. The investigators did not take Filomena’s or Laura’s reference DNA samples, even though they were Meredith’s roommates. Chris Mellas, Amanda’s stepfather, said that despite a court order in the summer of 2009, the investigators failed to disclose the electronic data files of the DNA evidence (called .fsa files) and logs to the defense. The lack of disclosure of evidence concerning the DNA forensics was one of the reasons that the defense essentially moved for a mistrial in the fall, a motion which was denied. According to Barbie Nadeau (, “Just as court adjourned in July, it was revealed that the prosecution had held back key evidence from the defense and civil attorneys. On Monday, clutching the Italian constitution, Sollecito’s lead defense attorney, Giulia Bongiorno, told the judge: ‘My client has been denied his right to adequate representation.’”

Although Ms. Nadeau does not indicate the nature of the evidence withheld, the subsequent Johnson/Hampikian open letter of 19 November 2009 noted specifically lack of disclosure of the .fsa files. Because it is difficult to see why the defense would not share such files with Dr. Johnson and Professor Hampikian (or at least make them aware that the files had been finally disclosed), the open letter tends to corroborate Mr. Mellas’ account. In addition, Bob Graham wrote (, “Other forensic experts in several countries – including two from Britain – have started to study the DNA results but have delayed giving a verdict until they have received precise details on the methodology used by their Italian counterparts. The failure of the prosecution to provide these details to Knox and Sollecito’s defence teams is one of their central complaints.”

Why are the .fsa files important? Technical bulletin 40-035 from Chromosomal Laboratories, Inc., is a checklist of what the laboratory is expected to provide, and it states, “Copies of all data files used and created in the course of performing tests and analyzing data in this case, including .fsa files, if applicable. These files should include all data necessary to independently reanalyze the raw data.” In response to a question on this subject, Professor Dan Krane wrote, “The biggest concern that I personally have regarding this case is the refusal of the prosecution to provide the defense with a copy of the electronic data that underlies the DNA test results -- that is virtually unheard of world-wide today and it would be especially important to review that data in a case such as this which seems to involve such low level samples.” According to a knowledgeable source, “Each tested sample has its own file. The file contains the full electropherogram trace information along with other information about the testing conditions (e.g., date, time, injection time, voltage, temperature, current, the RFU threshold used by the analyst). If you have the electronic data, you can use the DNA analysis software (GeneScan & Genotyper or GeneMapper ID) to independently analyze the electronic data. That allows you to examine the results as closely as possible (zoom in on the electropherogram to evaluate low-level results) and establish the RFU threshold of your choosing.”

One leaves DNA by many mechanisms including shedding hair and skin. Finding Ms. Knox’s DNA in her own flat is a very unsurprising result, whether or not it happened to be mixed with Meredith’s blood. Her DNA was found in Mr. Sollecito’s flat as well. The lack of their DNA in Meredith’s room on all but the bra clasp is a silence that shouts. In a pretrial hearing, Mr. Sollecito’s lawyers argued that it would have been very unlikely for him to be able to leave DNA on the bra clasp without also leaving it on the bra as well, yet none was found there (

The other biological forensics was also problematic. Luminol was overapplied to one of the footprints, according to Colonel Garofano in the book Darkness Descending, leading to a loss of detail in the image. Moreover, luminol is a presumptive test for blood, one that should be confirmed by further testing. Although some claim to have the ability to tell blood from other substances via luminol alone, this claim is controversial in the forensic literature. Moreover, defense expert witness Sara Gino stated "We were not told that, first of all, the prints were treated with a substance which should have indicated whether they were blood, and the result was very uncertain." In addition, the investigators did not obtain reference footprints from all of the roommates. The temperature of Meredith’s body was taken about two days after death had occurred, making it harder to estimate the time of death.

The investigators could have performed the electronic forensics much more professionally. For example, the investigators damaged three hard drives, Meredith’s, Amanda’s, and one of Raffale’s, ( Amanda’s hard drive is said to have pictures of her and Meredith, evidence that would belie the impression that the two roommates did not get along, but her drive has not yet been salvaged. Her defense team has offered to pay for a different group of experts to attempt to retrieve the information, but this offer was rebuffed. The investigators also inadvertently erased a piece of potential alibi information (, the time that the file Stardust was accessed (

Non-forensic evidence
The prosecution would have us believe that a gang rape occurred, despite the fact that this is an unusual occurrence. Ms. Knox and Mr. Guede had met only twice, the two men had never met, and neither Ms. Knox nor Mr. Sollecito communicated with Mr. Guede in any documented way. It strikes me as extraordinarily risky to commit a crime with someone one does not know. Then Ms. Knox and Mr. Sollecito are imagined to have removed evidence linking them to the crime but not evidence linking Mr. Guede to the crime. In addition, the prosecution changed the motive during the trial but did not have evidence to support either motive.

Raffaele’s kitchen knife is too large to have made the bloody outline of the knife on the sheet. It is too large to make two of the three wounds, and any knife could have made the third. There is no reason for Raffaele and Amanda to have transported the knife in the first place. On top of that we are asked to believe that Raffaele and Amanda would not dispose of this knife, even though Rudy would have seen them use it and could rat them out, and even though the other knife was disposed of. We are also asked to believe that they would clean the blade of the knife but not the handle. This whole scenario would be risible if it were in a third-rate mystery novel.

Raffaele and Amanda are said to have faked evidence of the flat’s being burglarized. I do not find the argument that the break-in was staged to be convincing. There is little evidence pointing that way. Filomena’s room was tidy when she left it, but she recalled that when she returned, there was glass on her clothes. This suggests that someone put the clothes on the floor to simulate a break-in, then broke the window in her room. However, the photographic evidence is not definitive, and it is possible to imagine a number of explanations. Suppose the window were broken, and glass ended up on the laptop. Then suppose that the laptop were moved from the desk to the floor, where some of the glass slid off to some clothing and some remained on the computer. Also, Filomena was allowed back into her room to retrieve some items, and it is possible that she inadvertently did something to put glass on the strewn clothes and remembers the room as though she found it that way.

It has been argued that Raffaele called the Carabinieri after the unexpected arrival of the Postal Police, in an attempt to portray himself as a concerned but innocent citizen, then lied and said that his call to the Carabinieri preceded the appearance of the Postal Police (who were returning missing cell phones). The prosecution wants us to believe that Mr. Sollecito lied to the police about when he placed the call, before he was a suspect. The explanation for what probably did happen is complex, but unless one believes that Mr. Sollecito is very stupid, the prosecution’s theory does not make much sense. Attempting to lie about when the call to the Carabinieri would have little upside (he could have said to the postal police that he was just about to call them or the Carabinieri), and has a huge downside. One would generally believe that the police would be meticulous record keepers about the times of calls and other matters. If Mr. Sollecito were tempted to concoct this story, a moment’s thought should have been enough to convince him of the likelihood of being found out. The two defendants did do some dumb things, but I do not believe that they are that stupid.

Ms. Knox’s incriminating statements were thrown out as evidence for the murder trial; therefore, they cannot be used to assess her legal guilt or innocence. However, suppose we consider her incriminating statements against Lumumba in assessing factual innocence and ask why she made them. Her subsequent comments about the interrogation are similar to those made by people who have been pressured into making false confessions. Francesca Bene’s interrogation provides a modest amount of support for what Ms. Knox said, as discussed in a previous post. In addition, we have Douglas Preston’s account of his being interrogated in a foreign language, and his evaluation that his halting answers made him sound like a liar. It is quite possible that Ms. Knox’s interrogators misperceived similar behaviors of hers as evidence that she had something to hide and therefore pressed harder. Moreover, her later statement, “But I've said this many times so as to make myself clear: these things seem unreal to me, like a dream,” implicitly disavows Lumumba’s involvement.

The investigators claim that no undue coercion was used, but Dr. Giobbi said that he heard Amanda screaming. The investigators claim that they did not record the key interrogations with Amanda that night, despite having previously recorded her interviews and phone calls. Furthermore, they lied when they said that Amanda came in to the police headquarters of her own volition ( Therefore, Ms. Knox’s account of her interrogation is more credible than theirs is.

Ms. Knox’s allegedly performing a cartwheel at the police station was the fodder for a great deal of negative publicity. Her own explanation of that episode, as relayed through Chris Mellas to Frank Sfarzo of Perugia-Shock, is that she was stretching to relieve stress when a policeman commented on her flexibility. They fell into a conversation, and he asked her what other gymnastics or exercises she knew. Ms. Knox’s testimony at her trial mentioned meeting the policeman but did not say whether or not he asked her to demonstrate other gymnastic moves. Elsewhere, she implied that she did at least some exercises or stretches of her own accord. Therefore, her testimony neither contradicts nor confirms the version Mr. Sfarzo reported. In any case, her yoga or gymnastics have little or no bearing on her guilt or innocence.

What about Raffaele’s probable lie about his cooking with Meredith and his pricking her finger? Does this not generate suspicion? His statement was made only after he had heard that her DNA was found on the knife blade, not before, and it was made without knowledge of the quality of the data. His statement was stupid, but understandable. Therefore, I would say that there is reason to be suspicious of Amanda and Raffaele, probably not enough to indict, and certainly not enough to convict.

The defense have argued that this tragic crime was the work of a lone assailant, Rudy Guede. The evidence linking Mr. Guede to the crime is much stronger; there is a bloody handprint, the bowel movement in the toilet, and there are several pieces of DNA evidence. I am puzzled by the sexual aspect of the crime, but despite this reservation, Mr. Guede acting alone in the rape and murder is the most logical explanation to me. If information about Rudy’s involvement had come back earlier, I doubt that the investigators would have ever moved for an indictment against Raffaele or Amanda. In other words, this is an instance of garden-variety tunnel-vision on the part of the investigators.