The case of the murder of Meredith Kercher has moved to a new phase with the reconviction of Amanda Knox and Raffaele Sollecito on January 30th in a court in Florence. They were originally convicted on December 4th 2009 by a Perugia court and sentenced to 26 and 25 years in prison respectively. On appeal, which in Italy is in effect a retrial carried out to higher standards with a more educated jury, they were not simply found ‘not guilty’, but were, in fact, declared to be completely ‘innocent’ of the crime. This is a much stronger statement.
But the Court of Cassation in Rome then decided that the appeal court had not taken into account ‘osmotic evidence of guilt’ and ordered a retrial in Florence, which ended with a reconviction and increased sentences for both defendants.
Key to the case are two items of DNA forensic evidence found on a large knife and a bra clasp. They have been used as the most important evidence by the prosecution to argue that Knox and Sollecito were involved in the crime. However but the accuracy of the DNA tests has been strongly disputed by many forensic experts, both in court and outside. During the Appeal Trial, two independent forensic experts, Conti and Vecchiotti, were appointed to look at the DNA evidence impartially. Their report documented in detail the procedures used for sampling and the subsequent analysis of these two items of evidence. They concluded that both were seriously flawed, and therefore, they argued, the DNA evidence could not be used as evidence of guilt.
The large knife had been taken from a drawer in Raffaele Sollecito’s apartment four days after the murder. That particular knife was chosen from a series of possible knives by a police officer, who chose it based on ‘investigative intuition.’ Subsequently it was found that that knife was too large to fit most of the cuts on the body of the victim, the cuts on Guede’s hands and the bloody imprint of a knife found on the pillow. These cuts had all been caused by a smaller knife.
The prosecution have argued that there must have been two knives involved. This large knife, which they say produced one slash on the victim, and was cleaned and returned to drawer where it came from, and another knife, which has never been found. However it is clear from testimony in court that the precautions used when sampling that large knife were seriously flawed. The police officers who took the knife from the apartment were wearing ordinary clothes and when they got to the police station the knife was not properly sealed before storage. When it was analysed by standard DNA procedures, which require 0.5-1.2 ng/μl of DNA, the key sample (B) was found to be too low.
Instead of accepting that result, Stafanoni, the police forensic geneticist, then concentrated the sample to 10 μl and carried out LCT high[MK1] sensitivity analysis. It is well known that the principle problem of LCT analysis is contamination and it requires especially careful handling and storage because it can easily result in false positives due to its extreme sensitivity. There are internationally agreed protocols to minimise the risk of false results. The Conti and Vecchiotti report lists in detail that many of these protocols were not followed. In particular, all samples must be run in 2-3 replications to confirm alleles are indeed from the victim. This is the required procedure of the Carabinieri Forensic laboratory. No replicates were run. The sample lab must run the LCN samples before reference samples, and in particular before samples which are known to contain the DNA of the victim, in order to avoid cross contamination. In this case the samples were run in the middle of a long run containing many samples with the victim’s DNA. It is essential to run both positive and negative blanks associated with LCN samples to reduce the chance of random contamination. This was not done.
The 2nd item of evidence which has been strongly disputed is the DNA on the bra clasp, which was retrieved from the floor of Meredith’s room 46 days[M2] after the murder, The bra clasp had previously been photographed in the room but not retrieved. It had been moved several times and when it was eventually retrieved it was demonstrably contaminated, being a darker colour than it had been initially.
As with the knife, the standards used to handle the clasp were seriously below those normally accepted.. The glove used to pick up the bra clasp was grimey. In this case there was sufficient DNA for a normal analysis. The major DNA signal unsurprisingly was the victim’s. In addition there were minor signals, including some that were associated with male Y chromosomes.
Stefanoni stated in court that she had identified only two persons from the DNA, Kercher and Sollecito. However in order to identify Sollecito, Stefanoni simply ignored 14 significant peaks in the analysis as ‘stutter’ peaks. This selective interpretation is against the specific recommendation of the DNA commission of the International Society of Forensic Genetics concerning the correct interpretation of mixtures. If these peaks are not ignored then there were 3-5 male minor contributors present in the trace besides the victim, of which one of the haplotypes may correspond to the Y haplotype of Sollecito. Since nobody has ever suggested that a series of men were involved in the murder, the only conclusion that can be drawn is that this is demonstrable contamination.
The only place Sollecito’s DNA was found in the apartment was on a cigarette butt, together with Knox’s. The prosecution tried to argue that ince her DNA was not found on the bra clasp, he must have been in the room. This argument is clearly flawed since many of the places where Sollecito had clearly left his DNA had not been analysed. This includes the door handle, which he had touched when trying to open the door before the postal police arrived, broke the door down and found the body.
In the Florence retrial, the serious criticisms of the procedures carried out, which were identified in the Conti and Vecchiotti report, were not addressed. They were simply ignored. The court accepted the view of a prosecution witness and the assertion by the prosecutor that the lab where these analyses had been done are world class and therefore their results must be trusted. No lab anywhere in the world has the right to be considered immune from mistakes. Indeed the best labs are especially keen to document and share with anyone relevant, the steps they take to minimize the chance of contamination.
In this case the Court of Cassation demanded that the defence prove that contamination had occurred. Furthermore, they were asked to do so without full disclosure of the Electronic Data Files. This is a reversal of the natural law of all justice systems, where the accused are considered innocent until proven guilty and have the right to examine and challenge the evidence against them.
Given the very high media attention that this case has attracted throughout western Europe, we are particularly concerned that not only has this gross misuse of DNA forensic science resulted in the wrongful conviction of two young people ersons but that these standards of proof will could be used in the future in courts throughout Europe and N.America to justify other wrongful or doubtful convictions on the grounds that such standards were good enough in the case of Knox and Sollecito.
By Professor Michael Krom from Leeds University