Professor Peter Gill Interview
Transcript of Professor Peter Gill’s interview on Italian documentary, Delitti & Misteri, January 29, 2015. The documentary was produced by Francesco Mura and aired on Retecapri.
Q. Professor Gill, thank you very much for meeting us and agreeing to this interview, and for clarifying for the italian TV audience how a science such as that of DNA can be used and misused. But first please describe for our audience what method you and Sir Alec Jeffreys developed around 1985.
A. I joined the Forensic Science Service in 1982 and at that time DNA profiling didn’t exist. It wasn’t till two years later that Alec Jeffreys announced his pioneering work on DNA profiling. The greatest difference between his method and the methods that we were using at that time was the incredible discriminating power of the system. So I began working with him in 1985. Now at that time it was just a concept; no one had proved in practice that DNA could be used in forensic casework. And so what we did was we got together. I designed some extraction methods to remove DNA from crime stain materials – materials like blood stains, semen stains, hair shafts etc, and I managed to extract this DNA and then took the DNA to Alec Jeffreys lab, and we processed the DNA using his new technique. And towards the end of 1985 we were able to demonstrate that we could extract DNA from old blood stains – many of these stains were many years old, and not only that but we showed that we could extract DNA from semen from vaginal swabs as well. This was a very important discovery, especially for the semen-contaminated vaginal swab, because prior to that it had been impossible to carry out this kind of analysis using conventional methods. So that by the end of 1985 we had developed all of the methods necessary to carry out forensic DNA profiling, and that’s how the revolution in this area began.
Q. I believe you hit the headlines early on when you applied these methods to solving a mystery over the deaths of the last Tsar of Russia and his family? What were you able to establish there just for the interest of the audience?
A. This was approximately five or six years after the collaborative work with Alec Jeffreys. I was approached by the Russian authorities to carry out the analysis on the remains of the Tsar’s family – they didn’t know whether they were the remains of the Tsar’s family then, and they wanted confirmation. So this was particularly challenging because these remains had been in the ground for at least 80 years, and no one had done this kind of work before. Our lab at that time was probably the most advanced in the world. We were still developing these techniques, and we were able to extract DNA from the Tsar’s remains. So what we demonstrated by comparison with Prince Philip who is a living descendent was we showed that the DNA profiles were the same, so we managed to establish a link between the Tsar’s remains and modern day descendants and we also demonstrated the link with other descendants as well. So following our work we were able to definitively demonstrate that these were the remains of the Romanovs.
Q. Professor, is it possible in your opinion, turning to forensic uses, to condemn somebody, to find someone guilty, only on the basis of DNA evidence?
A. When a case goes to court, it is very very rarely solely on the basis of DNA evidence alone. In a typical case there are several strands of evidence, and all of them would be considered by the court. It is extremely dangerous if there is just one strand of evidence, and that strand is DNA profiling, then it may be difficult to put it into the context of the case by itself.
Q. Nowadays many prosecutors in Italy tend to be too quick to use and abuse DNA evidence, using it as a key evidence on which to condemn, to find guilty. This is helped, we find, by experts whose opinion seems to vary according to who asks them. Professor, could three different DNA analyses, made by prosecution, defence and civil parties really come up with three different answers, or does everything depend in such a case on who is paying for the expert? And what can be done to prevent this happening?
A. Different experts will have different opinions. The question is often not so much whether the DNA profile has originated from one particular individual. Rather, it is more to do with ‘how did it get there, when did it get there?’ The problem with a DNA profile is that there is absolutely no information in it whatsoever that tells you when or how it was deposited, or how it became evidence. DNA doesn’t come with a date stamp attached to it. And therefore the job of the expert is always to be very clear in the evidence, and not to give so much his personal opinion, but rather to explain all of the possibilities whereby evidence may be transferred. Essentially for any case there are three possibilities. It is possible a DNA profile may be deposited before a crime event; it may be deposited during the crime event, as a result of contact between perpetrator and victim, for example; or the third possibility is that the DNA profile may be deposited on the evidence after the crime event has occurred. And the latter would be typically associated with a contamination event. The problem that we have is that recently DNA profiling has become so sensitive that we were able to detect (it in) just a handful of cells. Back in 1985 this was not so much of a problem, because we needed an area of blood stain approximately 1 or 2 centimetres in diameter before we could even get a result. But now we can get results from invisible areas, invisible stains. If I touch a table, I will leave a full DNA profile on that table. The other issue is that we have to be extremely cautious when we are collecting evidence, because we have demonstrated quite clearly that DNA can easily be transmitted from one area of a crime scene to another, for example; and this can be achieved simply by touching an area with latex gloves. And I think people in general don’t realise how easy it is that you could actually move DNA from one area to another. So the sensitivity of the technique is a double-edged sword. We detect more DNA profiles, but at the same time it increases the uncertainty about how a DNA profile was actually transmitted to become evidence.
Q. Professor Gill, you have recently written a book on the abuses of DNA in which you have written a whole chapter on the murder of Meredith Kercher. This has gone through four processes; first they were found guilty; then absolved; then found guilty again, or sent back and found guilty again. And this whole process has gone on for seven years, and still preoccupies the Italian public and many others as well. Can you tell us what you feel about this case?
A. The reason I wrote the book was not just because of this case. This was something I had a growing concern about – how DNA profiling is used throughout Europe and beyond. I work very closely with my European colleagues, and I have to say that the view of forensic science from the public, the politicians, and also I believe judges and lawyers probably comes predominantly from TV programmes which talk about CSI (Crime Scene Investigation). So in the book I talk about the CSI effect. There is an important psychological influence at play. Whenever people hear about DNA profiling, they automatically associate it with a crime event, and then again this is all part of the public psyche built up over the years by the TV shows which mis-portray what DNA profiling is, and mis-portray the limitations of the process. And this is a big problem. The problem I think is illustrated very well by the Kercher case where there is some DNA evidence, but it appears to me that the interpretation from the evidence is confused. There have been two or three judgements. I thought the Hellmann judgement was a good judgement because what it did was it accepted the possibility of contamination, and it said that the onus was on the prosecution to show that contamination had not occurred. In my opinion that is the correct way to think about the evidence. The latest judgement, however, has reversed the burden of proof which means that they have asked for the defence to prove that contamination did (not) occur. Unfortunately this is absolutely impossible for the defence to do; they are asking the defence to do something which is quite impossible, and also it is bizarre that the reverse burden of proof should be used in this case. Usually it is for the prosecution to demonstrate beyond reasonable doubt that something hasn’t occurred. The onus of proof is not usually put on the defence. So I am particularly concerned by this reverse burden of proof. It does actually put the defence in an impossible position; you cannot prove that something hasn’t happened.
Q. In the Kercher case they brought the presumptive evidence of DNA from Raffaele Sollecito recovered from Meredith Kercher’s bra clasp which had been collected 46 days later from the crime scene. A crime scene that had been repeatedly contaminated and re-investigated, and after the crime scene had been closed as far as I recall. In such a case is it actually possible to extract DNA without errors and in a way that’s meaningful?
A. In the case of the bra clasp, there is certainly (Sollecito’s) DNA on the bra clasp, but your question is very pertinent and it comes back to what I said previously; we can detect a DNA profile, but what does it mean? I can take a DNA profile from anywhere in this room, and I can find DNA profiles everywhere, but it doesn’t mean I have participated in a murder. Now with the specific example of DNA on the bra clasp, it is a complex mixture, with at least 2 males and possibly more male components in this mixture. Is it possible that Sollecito’s DNA has been transmitted from one area of the crime scene to another? This of course is a difficult question; and let me come back to… if we’re going to answer these kinds of questions, then the crime scene needs to be managed in a very strictly scientific fashion. The investigators themselves actually become vectors of DNA. For example, if an investigator touches a door handle and then proceeds into the crime scene, and then touches the bra clasp, without changing gloves in between, then I would actually expect DNA transfer to occur. And I think I would be surprised if it did not occur, actually. So this is the crucial question, and the difficulty with a crime scene is you can’t get in a little time machine and go backwards to find out what actually happened, but if there is sufficient doubt in a case, then the possibility of contamination by an investigator is something we should consider seriously, and not dismiss out of hand. My understanding is that there were no DNA swabs taken from the door handle, from the outside, for example, and therefore Sollecito’s DNA profile was not actually recovered outside the room. But my point is, there is considerable uncertainty in the case, because it is not denied that Sollecito had access to the flat, and therefore his DNA will be present in the flat. Then the question is ‘what is the significance of finding DNA on the bra clasp? And what are the possible methods of transfer?’
Q. Turning to the knife, which was taken from Raffaele Sollecito’s drawer. They found – there is no doubt that it had Amanda’s DNA on it, but none of the victim, and it is claimed by the prosecution that this is the knife that was used to stab Meredith Kercher. It was obviously washed, because it was found to be very clean. Is it possible for them to have washed off the DNA of the victim, and presumably the blood as well, and to have left the DNA of Amanda?
A. My understanding is that there was DNA from Amanda Knox on the handle, and trace DNA evidence of Meredith Kercher on the blade. However the DNA profile on the blade, purported to be from Meredith Kercher was extremely weak, and furthermore there was no evidence whatsoever of any blood. But there was evidence of starch grains which was evidence that the knife was used to cut food. So the main question is ‘was the knife used to stab Meredith Kercher?’ And obviously part of the evidence is the DNA profile on the knife handle, which matched Amanda Knox. But there are obviously two possibilities; either the knife was used to stab Kercher, or else it was used to prepare food and it was not used to stab Kercher. And the question is ‘can DNA profiles distinguish between the two events?’ And it clearly cannot, because the DNA profile gives us no information about how it was transferred. The DNA profile which I would observe on the handle of a knife would be the same regardless of whether the knife had been used to cut food or to stab a victim. I have seen some evidence from the judgement where it has been proposed that the distribution of the DNA on the knife handle is evidence that it was used to stab rather than to cut food, but there is absolutely no scientific evidence, there are no scientific papers whatsoever that would support this kind of conclusion. So again the DNA profile takes us no further forward to decide whether it was transferred as the result of a cutting action as opposed to a stabbing action. The DNA profile purported to come from Meredith Kercher is extremely low level, and again we have to consider the same possibilities of how it got there. Either it was used as a murder weapon, or else the DNA could have been transferred as a result of a contamination event. … It’s said that the knife has been cleaned with bleach, for example, which is why there is a small amount of material purported to come from Meredith Kercher, but I actually find it quite difficult to believe that bleach would selectively remove blood and leave some DNA there. And it’s also another example where no experiments have been carried out to verify whether these conclusions are feasible. I would actually carry out some experiments to take some knives; I would do careful controls – I would put blood on the knife blade, I would clean it with bleach for example, to see whether the DNA was selectively removed or stayed intact. But these types of experiments have not been carried out, and therefore all I see with this is quite a lot of speculation.
Q. The knife and bra clasp have been much discussed, but there is this other element which has been completely ignored by prosecutors and by the Court, and that is that there are presumptive traces of semen shown up by the crimescope on the cushion on which Meredith’s body lay. Do you think that examination of this would be crucial under normal circumstances to reveal the identity of the real murderer or murderers?
A. If semen stains have been discovered in the vicinity of the victim’s body, then I am extremely surprised that no efforts have been taken to analyse the material. It doesn’t mean to say that it is associated with the crime event, – it might have been deposited some time previous to the crime, for example – but nevertheless if there is potentially important evidence like that which has been discovered, then it is difficult to understand why it has not been analysed, as it may take us further forward in understanding how this crime was committed.
Q. What similar cases to the murder of Meredith Kercher have you been involved with and are there other cases which would help inform Italians who are obsessed with what seems to be a use and abuse of evidence in the justice system?
A. My role throughout most of my career has been to carry out research. I am involved with case work, but usually I am helping other scientists to understand their cases and to report them. There is quite a big forensic science community within Europe, and we work under the auspices of the European Network of Forensic Science Institutes (ENFSI). The big problem that we have is two-fold. There is a lack of funding, and there is also a lack of education. And these are the two biggest issues that face us. Forensic Science is extremely disparate. It needs much, much better coordination. It needs to be taken much more seriously. Forensic Science needs to evolve into a much more mature science, and we can only do this if there are new initiatives within the EU that allow us to carry out this training that is so essential.
Q. What other cases have you had experience with elsewhere in the world, which would help us to put his Italian case into context?
A. There is one case I can allude to, which is the case of Adam Scott, which is quite a notorious case here in the UK. A man was accused of a rape, because his DNA profile was obtained from the swabs – the vaginal swabs, and it matched a man who was some hundreds of kilometres away from the crime scene. And he denied ever having been in the place in his life. The evidence was the DNA profile. He was arrested and incarcerated for about six months, and all the time he was protesting his innocence. Luckily for him it came to light that there had been a contamination event in the actual laboratory. His DNA profile had actually been submitted to the same laboratory a couple of weeks previously, and his DNA profile from that particular event from saliva, was transmitted into the casework analysis for the second system. So of course he came up positive for this particular case, but his DNA had actually been transferred from a previous event. So this is a very good illustration of DNA actually moving within the laboratory, and not only that, it can move from one case to another case. Fortunately for him there was sufficient evidence to show that contamination had actually occurred, and he was eventually released, although he was in prison for about six months before he was set free. So the fact that this kind of event can happen should make us all very concerned, and it should make us all very very aware that this is probably not a one off example. The problem is that if you observe a contamination event, and you prove a contamination event has occurred, then there are probably a large number of cases where (it has) occurred and you have not verified that a contamination event has actually happened; and therefore it is missed, and that is my biggest concern I think. For every mistake that you find, there are many which go undiscovered, and that means that there are probably innocent people in prison.
Q. Professor Gill, before taking our leave, could you give a final message to the Italian Public and its justice system on this complicated and poorly understood theme, with a view to pointing the way to how things might be improved in the future.
A. The first thing to do is to forget everything that you think you know about forensic science that you have gathered from TV shows. This is completely wrong, and really we have to start again. Forensic science is the same as any other science. That means that we need to make sure that we have very good evidence, and critically we always need to explore all possibilities, not just one or two obvious ones, but particularly how did the DNA get there, when did it get there? Is there evidence for contamination? And even when there isn’t evidence for contamination, are we sure that we are looking for it? Are laboratories themselves actually in a fit position to carry out this kind of profiling? Do they carry out laboratory controls to look for evidence of contamination within the laboratory, for example? How often do they clean the laboratory? Are they changing their latex gloves in between handling evidence for example? There are many, many important things to consider. The problem is that once a DNA profile is obtained, then the (DNA) evidence seems to counter evidence that points away from the suspect. And I call this the ‘swamping effect’. Everyone is so blinded by the huge power of DNA profiling that the obvious falls by the wayside. So DNA always has to be considered in the context of the other evidence, and should never be considered by itself.
Videos of Professor Peter Gill on Porta a Porta and BBC Radio 4 discussing the case can be found here: http://www.amandaknoxcase.com/videos/