Episode 7
If you enjoyed this episode or at least learned something new, please consider supporting the podcast by buying me a coffee.
The Death of Socrates - Jacques-Louis David
In this episode I spoke with Professors Dr. Matthias Wienroth and Dr. Carol McCartney, two of the authors of "Ethics as Lived Practice. Anticipatory Capacity and Ethical Decision-Making in Forensic Genetics". It's a fascinating discussion about forensic genetics and the rapidly evolving technologies from the perspective of a social scientist and law professor.
Read the full text article: Ethics as Lived Practice. Anticipatory Capacity and Ethical Decision-Making in Forensic Genetics
Works/News mentioned in the episode:
Nuffield Report. The forensic use of bioinformation: ethical issues
Marper Case - European court rules DNA database breaches human rights
The effectiveness of the UK national DNA database
The murder of Rachel Nickell
The Phantom of Heilbron (This case spawned and entire new ISO standard!)
Dr Carole McCartney is a Professor in the School of Law, Northumbria University. Previously senior lecturer in criminal law and criminal justice at the University of Leeds, and Bond University, Queensland, Australia. Carole has written on Australian justice, Innocence Projects, miscarriages of justice, policing cooperation, and DNA, forensic science and criminal justice more widely. She established an Innocence Project at the University of Leeds in 2005...Learn More about Dr. McCartney
Dr. Matthias Wienroth has been published in numerous peer-reviewed journals, he is a founding member of the of the interdisciplinary scientific initiative on new and emerging forensic genetics technologies WIE-DNA (Germany) and of the global network STS MIGTEC (Science and Technology Studies of Migration and Technologies). He is a member of the Society for Social Studies of Science (4S) and the European Association for Studies of Science and Technology (EASST). Since 2015, he has been building an interdisciplinary and international network for the Social Studies of Forensic Science..."How can technology contribute to the 'good society', and what does 'good society’ even mean? I address these questions in my work on studying science and technology as social practices. I am particularly interested in how technologies shape the ways we know and (un)know human beings and relationships, individuals, communities and society. The fields I primarily study this in are security & justice (forensics and biometrics), and human health..." Learn More about Dr. Weinroth
Additional works by Dr. Wienroth:
EUROFORGEN and Sense About Science booklet on forensic genetics
https://senseaboutscience.org/wp-content/uploads/2017/01/making-sense-of-forensic-genetics.pdf
Others articles:
Value beyond scientific validity: let’s RULE (Reliability, Utility, LEgitimacy)
Misuse of Romani genetic data:
· Europe’s Roma people are vulnerable to poor practice in genetics
· The DNA of Roma People Has Long Been Misused, Scientists Reveal
Join our Facebook Page!https://www.facebook.com/SecretsfromtheCrimeLab
Comments? Feedback? Suggestions? Or if you would like to be a guest?
Get in touch on the Contact Page
Ep 7 Transcript
[00:00:00] Angela: Hello and good morning or good afternoon as it were, depending on where you're geographically located, and welcome to Secrets from the Crime Lab, an educational podcast where we discuss forensic science and related scientific fields and topics. I'm your host, Angela Swarts. And joining me today, I have two professors from the University of Northumbria. Dr. Carol McCartney and Dr. Matthias Weinroth. And I'm expecting, you're going to have to correct me on your name pronunciation yet again. Would you like to go first? Introduce yourself?
[00:00:33] Dr. Matthias Weinroth: Sure thing. Thank you very much, Angela yeah. My name is Matthias Weinroth and I'm a senior research fellow at the University of Northumbria at Newcastle in the UK. Um, I'm particularly interested in, the study of sciences and technologies biosciences and biotechnologies from a social scientific perspective. So I'm interested in the social and ethical aspects of how we create and produce knowledge using particular types of biotechnologies. One of the areas that I'm particularly interested in is forensic genetics. And that's the reason why I'm on your podcast. And, another is healthcare. So where a lot of the very similar issues come together around knowledge about the human body and about,about the way that, the technologies are used to create and recreate social order. And that's of course, particularly interesting when it comes to criminal justice and the investigation of crimes, just how technologies there contribute to social order. For example, through the production of new types of suspect identities, or the way that these technologies can produce forms of intelligence or evidence. I'm interested in just the way that these kinds of various different types of biotechnologies, particularly in forensic genetics contribute to this, the social order, that,that is there, beginning and the end of the criminal justice system.
[00:01:54] Angela: Yes. thank you for coming on. Carol, would you like to introduce yourself?
[00:01:58] Dr. Carol McCartney: Hi, I'm pleasure to be here. So I'm Dr. Carol McCartney. And I would say I've probably almost 20 years now studying criminal justice systems, but with a particular focus on miscarriages of justice. And of course that has then led to a focus on forensic science and particularly, biometrics, DNA, fingerprints, and so forth. But really looking at the whole sort of gamut of forensic science from, we call it from cradle to grave. Although I think the Americans say from soup to nuts. that I liked. So yeah, interested really in all aspects of forensic science, but particularly of course its impact on justice.
[00:02:40] Angela: So it turns out that I thought my first introduction to both of you was this paper that prompted me to contact you, to begin with. And it's in, in Gene's Ethics as a Lived Practice, Anticipatory Capacity and Ethical Decision-Making in Forensic Genetics. And from my experience as a forensic genetics professional, we're really not focusing as much as we need to on the ethical side of things. I think in your paper, you commented that it does seem like in a perfunctory administrative step that you go through rather than something that we should be keeping in mind at all times when we're doing our work. And then I hadn't realized when I started to look further into both of your bodies of research I'd actually read your work before Carol. The work that you did with the, Nuffield Council, the report on the bioethics that you did on forensic science, a long time ago, you were at that.
[00:03:39] Dr. Carol McCartney: It makes me feel old now when people bring up
[00:03:42] Dr. Matthias Weinroth: But it's still relevant.
[00:03:43] Angela: It is still relevant and then that was back in 2007. The question is, have we progressed as a field since then? Have we gotten better or because of all of these new emerging technologies that we have, the forensic DNA phenotyping and biogeographical ancestry and forensic epigenetics, all of these new technologies. And then you have cases like the golden gate killer in San Francisco, that kind of puts a lot of these things more in the forefront of the public eye. Forensics wasn't really keeping up with ethical decision-making to begin with compared with the medical community and the speed at which technology is moving... Are we still behind the curve?
Well, it's interesting. Of course we say that was a long time ago. And like you say, if things have moved on, I think that the wonderful thing about that report was it was a point at which we took stock and said, hang on, let's have a look at what we're doing here, because we had the Marper case going through the European Court of Human Rights. So it was a juncture at which we said, we've got this new technology. We've raced ahead with it. We've built this massive database. We've taken DNA from people left right and center. Perhaps we should stop and have a think about this. So it was great at the time. And we had the Marper ruling. I think my concern then for like,that was wonderful, but two things have happened that I think there's a feeling that we dealt with the ethics then. That we had Marper, we've had that report. We've done that. So we thought about the ethics and that's all sorted. The other thing that I think has happened, which kind of led to the paper that you mentioned, is that people have delegated it. So they have appreciated the ethical questions that surround all of this, around forensic science, in the criminal justice process and in society. But they think that somebody else likethere's people out there that will deal with that. So the Nuffield Council on Bioethics, they wrote a report, so that's good, but there's other people whose job it is to deal with this. So it's not actually a concern for us because someone, somewhere else is dealing with that. So I think we haven't moved up. We haven't moved along or we haven't kept up with the times and we haven't kept under constant review the developments and the you know, new technologies and new techniques. But we've also thought that it's been dealt with and now it's somebody else's job, but it doesn't actually, I don't, I'm not convinced that the forensic community and obviously, I don't know them all personally, but the sense that you get is,ethics is somebody else's job. And so that I guess is one concern. Yeah.
[00:06:30] Dr. Matthias Weinroth: That is also one of the concerns that led us to writing this paper. The concern that, Carol just raised, this this kind of disconnect yes, we need to talk about ethics, and then the actual implementation of what it means to think and act ethically in forensic genetics. And that's, a starting point for the paper where we make the distinction within what you described as, more bureaucratic type of ethics, what we described as thin ethics. A really really thin understanding of ethics as something that can be a hurdle that, that is in the way. But essentially is a tick box exercise, uh, that, you know, people might find annoying, and, and just disruptive of the actual scientific and investigative process. And we we want to say, ethical decision-making is something that we do all the time and we just need to become more aware of it. And just how it impacts on our daily practices. As such, it's not a hurdle and it's not a tick box exercise. Ethics as lived practice is literally, reflecting on how we make decisions, what kind of principles, what kind of standards we apply,uh, what we consider to be what's done as we consider to be legitimate in the way that we make decisions. And even if, we find ourselves making decisions where we don't think it's a big issue, it's a big problem. It's not really a dilemma. We are still making decisions that can have far reaching effects. For example, when you make a decision about what reference data you use, it, it might seem to you at that point when you make that decision, this is really just a matter of, like a procedural matter, I'm using this particular type of reference data because this, database has the data available, easily accessible and I can work with it. But the way that the data have been compiled, and have been curated, might have a longer term effects on just how reliable and how useful, and essentially also how legitimate the technologies and the knowledge they produce can be in the end. A simple decision might have really long reaching effects and, we can see this happening at the time, but we can talk about, examples a little bit later on as well.
[00:08:39] Angela: Just in a general sense, I know that a lot of us that, you know, just day-to-day practicing forensic geneticists, you do just use the information that's available to perform that one thing that you're trying to do, whether you're just doing a basic calculation based on, population databases that were compiled from different groups, but you don't really... once you have it in place, you don't really think about where did it come from? Are there any issues with it? Is it skewed because we've, over-sampled a particular population over another, because we could exploit that population by, not really getting the informed consent that we needed to do it.
[00:09:25] Dr. Matthias Weinroth: Yes a related issue is of course that there are certain populations which are favored populations for data. An example is the, the Roma population in Europe, whose data have often been sampled and then used in the criminal justice system where prejudices have contributed. Not just social prejudices as an oh the Roma you know, they are, traveling people some might say, prone to criminal activity, that is a social prejudice, but also scientific prejudice, for example, saying, well, this seems to be a very, confined, genetically speaking confined population where we can get really good, pure data, whatever that might mean. And, and that of course is also not correct because when you are, uh, living across a whole range of European states, and of course you will have, intermixed with a whole range of European populations. So even the scientific prejudice there, can undermine the strength of the data that has been compiled and on which, certain analytical models are then being developed.
[00:10:24] Angela: And maybe that's why some of these analytical models are not holding up when you test them. So when we look at forensic DNA phenotyping, and it's off and it's not off all the time, but it's off enough for you to be very concerned about whether or not you're getting legitimate or valid information. And then of course, if you're already working on top of social prejudices that might exist and you're providing that information to law enforcement, are they then getting tunnel vision and focusing on searching for the wrong individual or, convincing themselves that they need to go to this particular group of people to find their perpetrator.
[00:11:04] Dr. Matthias Weinroth: Yeah. I mean, you bring it together different steps, along the, the range of, of knowledge production, of how meaning is produced,using, for example, forensic DNA. phenotyping. I started out with what's the reference data that you build your models on, and you already jumped forward then to the point of, um, okay, and now you apply these models in your analysis, and now you tried to communicate,information to investigators. How can they deal with the information, um, what will they do with the information? Will they, you know, make it public, uh, and say, oh, look, there's this particular population group that we thinkwe can find a, that the suspect or the perpetrator in, and possibly not necessarily understanding or thinking about the impact this might have on that population itself,which might start from something that seems relatively innocuous, such as saying, well, we think the perpetrator comes from your communities. So now we expect you have to do, you have to go the extra mile to help us in our investigation.Uh, so that means you have to give us your DNA so we can compare it. You know, really there was no expectation that they would have to give the DNA as just an example, if you were not from that population necessarily. Uh, so this is, you know, some of the examples that can happen, that we have seen happen, in practice, um, in the UK and in the US and elsewhere. Genetic information points into certain direction. And then those kinds of populations are being particularly subjected to scrutiny,um, which might make sense perhaps from an investigative point of view, but it does bring along,things like, you know, guilty by association ideas, and, potentially also, as we've seen in 2016 and 2017,attitudes of the population towards minorities, where forensic, genetics and forensic geneticists,had a really precarious role to play in fueling prejudices, public prejudices, political prejudices, and not always lived up to the requirement to do so in order to protect people who are innocent but who by association fall under suspicion.
[00:13:10] Angela: So do you think that, and I think you touched on it a little bit in your paper. It's the pressure. There's also an enormous amount of pressure on the scientists to ensure that all these fancy new genetic methodologies that we have can actually produce results that then they can use in crime and policing issues. One of the things is, you know,do you propose all of the new possibilities around things like the forensic DNA phenotyping and the investigative genetic genealogy and all these types of things, especially in the biogeographical ancestry as well. Do you, you push it to your ultimate customer, which would be, the law enforcement agencies and the prosecution services or the governments in general, in order to get funding, to be able to function as well? Is funding an issue? I know in a lot of other areas in medicine as well, there's this push to... I need to get funding, if I'm going to be able to continue to do my research and develop these methods to produce the results.
[00:14:10] Dr. Carol McCartney: I think there's a very real issue about how ethical it can possibly be to be pushing technologies that aren't going to be that effective. That, that are perhaps riven with risks. So risks that Matthias has already highlighted, but just plain risks of being wrong or misleading or you're giving highly you know, data with an enormous amount of caveats that need attaching to it. So you mentioned phenotyping, biogeographic stuff. You can't give that information to an investigator who, with all the will in the world is not a geneticist, a forensic scientist, and so forth. If you're giving them information and then saying, oh, but there's all these caveats attached to it. What do you expect that investigator to be able to do with that? Go along with all the caveat? Like it would be, it's almost meaningless to give them information that's just not going to assist them. And they'll probably strip away all the caveats and say, oh, you told me it was somebody who's five foot six and of black origin. And so that's what I'm going to look for. And so you can't then wash your hands when they go and arrest a load of people that look like the person you said it was going to look. And then what happens when they don't catch them, they turn around and go, Hey, your technology was rubbish so it's not even, there's no point to you developing technologies that a) have to be blanketed in caveats or are so sophisticated that you can't understand it really without, some statistical background or, it's going to be beyond the wit of most,with all the will in the world and as generous as you might be to investigators that they're not going to be able to handle sophisticated data. And realistically think about there's a murder investigation. I want to catch a suspect. Giving me a list of, you know,perhaps he might have come from the Caribbean, realistically, how helpful is that? It, it strikes me as it's really not all that helpful. It's only going to be of minor assistance perhaps in that odd case here and there. So how much should we really be invested in, in pushing that technology.
[00:16:26] Dr. Matthias Weinroth: Yeah, I um, I think the, the issue with those, highly specialized technologies that, and Carol, you just made that reference that they're, they're highly specialized, and you do need quite significant insight in order to communicate them well, to potential users such as investigators,and also where thestatistic or the probabilistic basis of the findings are so complex. Forensic scientists have been talking for some years now about how they could potentially,translate, the, the kind of likelihood ratiosand and other approaches that they prefer because they are more,uh, preciseand alternative forms of communicating probabilities, how they can be made understood. We know from research and experience that judiciary, and juries and, barristers solicitors,and police officers, investigators thought that, you know, if you throw at them a likelihood ratio, they, don't know what that means. It's just super complicated.And often there's not necessarily the time made to communicate what that means. So investigators and others might be looking for information that they understand. And if like certain companies in North America provide an image and a really easily accessible visual graphics, of information, you know,skin color, hair color, eye color, freckles, and so on so forth. Then of course, they'll go with that information thinking that this is reliable information, especially when there's also, information from scientists sometimes saying, well, you know, in the ideal case, this is reliable to 99.9%. But what that actually means is under ideal laboratory conditionsthe, the way that the technology worksis, is very high. But in the, if I may call it the real world of contamination and of decomposition of DNA samples and mixtures of DNA samples and so on so forth. Uh, those really high reliability rates, of course, don't apply anymore.And it can be quite difficult to then tell an investigator or a politicianwho's called for the use of these technologies well actually,you know, that's not the same thing. Using forensic DNA phenotyping in a criminal casewill only be useful in a very small number of cases. It's very rare if you don't have, if you don't have any eyewitnesses or you don't have any other circumstantial evidence to point into the direction of a potential perpetrator then you might use it. But then also to say, well, actually, the reliability of this technology isbased on various different factors, which we now need to talk through as well in order for you to get a good understanding of just how reliable this technology is. And this is a point where this notion of reliability meets the notion of utility, just how useful it is and Carol raised this as well, just how useful can this technology be? What needs to be the context within which it can be useful raises also another issue, that Carol has written about and that's integrity,of the, of the scientists of the forensic caseworker, and they need to be able to communicate when a particular technology is useful, or when it isn't, they need to be able to be transparent about the utility and the reliability of the particular technology at hand. I just want to briefly mention something and I think Carol, you might be able to say a little bit more about the Sydney Declaration. There was a paper published earlier this year and a number of forensic scientists have come together and written a list of seven principles, I believe it is, about what forensic genetics is and what, what needs to be taken into consideration here. And throughout those principles, one word comes up time and again, which I found so interesting and that's context, so, so very important, you know, determining whether something is useful, but also to determine in how far you can give information, what knowledge can you give to investigators? What is it worth for the investigation? The seventh principle talks a little bit about the ethics, of forensic scientists. Probably not as much as, as we would like them to, but it basically says,it's an ethical responsibility of scientists to understand context and distinguish it from analytical data. So yes, you need to know about context, but you mustn't let the context influence your analysis to such a degree that you're generating an analysis, which speaks tothe requirements very closely to the requirements of what the investigator wants to know. Rather, you should develop an analysis that speaks for itself. Aware of the context, but not mixing it up too much with the context. And then to be able to communicate the differences between the context and the analytical data that comes out of applying a particular methodology. And I thought this was a, this was a really interesting wayeven though they don't really mentioned terms such as utility and, uh, reliability and integrity, but it's, and it's an early starting point for thinking about ethics as, as lived practice, you know, cause we know that, it can be very difficult, to look at everyday activities that we all do and that forensic scientists, forensic geneticists do when they develop technologies and when they apply them,to reflect on, all the decisions that are actually going on that are being made. And it can be something as simple as saying, okay, I need to understand the context within which this trace was found,it will help me to do the analysis, but I also need to reflect on just how much this context might drive my analysisand and whether it might, color the information that might then come out in such a way that the information might not be, I don't know, as independent or whether it might be written for a particular viewpoint. If, you know what I mean, kind of try and carefully to, to phrase this here.
[00:22:24] Dr. Carol McCartney: I always think it just in terms of, in my head to put things simply I reflect on genetically modified foods in that if you talk to me about it in a laboratory setting, that we can tweak these genes to this plan and then increase the yield of this crop, whatever I'm like fabulous that's great. You know, great scientific development. Okay we're now going to change all the genetics of all these plants in this field here. Well, hang on a minute. What, you know, and this is where the debate starts because it's that difference between what can I do in a laboratory? What happens when it's out in the wild? Okay. And you can't divorce yourself, or I believe, and I'm sure Matthias and our other authors believe you can't divorce yourself from what's going to happen to that technology in the wild by saying, oh, but I'm in a laboratory and it's fine in a laboratory. Because what's the point of it in the laboratory? It has to go out into the wild for it, for there to be a point to it. So you, when you're developing, your phenotyping or your biogeographic ancestry testing and so forth, you have to take that extra step and think what's going to happen to this in the wild. And that's where the context is massively important. So you can't let it influence really the science that you're doing, but you can't divorce yourself from the ethical implications of what's going to happen to that technology out in the wild.
Yeah. So you mentioned GMOs and genetic modifications, I think you can look beyond food crops at this point with the CRISPR technology that's out. Right? So you have the ability to go in and, oh, well, let's fix these mutations that are specifically associated to cystic fibrosis, for example, which would be amazing for people that have it . But moving on from that, you could actually go and do designer babies. Right. Which isn't really something that you would, you know, like, oh, I can cure these diseases and these people can live normal, more healthy lives for longer periods of time, but then, there's going to be somebody out there that's going to go, well, I want my baby to look like this. So we were actually to the point where that's a possibility. But if you go back to what you were saying with, people that are actually doing the work in the laboratory, I think there is very much a focus on the technical aspect. We're focused on what we're doing. I've got to get, I have to produce a result. I have to produce a report. And even if they are looking at how is, how is my information going to be received by the investigator, even if you're concerned about that, a lot of times that communication doesn't come back towards the scientists. So the scientists provides information, they say, I'm available if you have any questions. I'm happy to further explain this and talk about the limitations related to my findings. But the investigators a lot of times don't come back. Like you said, they, they strip it away to the information that they think is the most pertinent without all of the caveats. And then of course, a lot of times your DNA reports are just read into the record in the courtroom. They don't even, they don't even ask you to go to court. It's the defense isn't,isn't taking any issues with the findings. They're happy to accept the information. And so the case moves forward without any scientist involved in it whatsoever. And of course, that's a big problem when we talk about the misuse of the genetic information or misstatement, it's not necessarily a misstatement by experts, but it's a misunderstanding by legal experts and it can lead to miscarriages of justice. Right?Yeah. Well, we even instigated a process in that, in this country called streamlined forensic reporting. The idea being well, we know DNA testing is reliable. We know it's effective. We can do this quite easily. It's all done on a computer, realistically. What's the problem with just get printing out a report that says match, okay. And that saves time saves money. And then we won't be bothered with any forensic scientists getting involved having to write reports and they cost money and it takes time. So we have a system whereby a laboratory will print out a result that says, this is a match. Okay. That gets put on one side of a form that gets given to the investigators . That then of course will be, given to any suspect, and their legal advisor saying, oh, we've got your DNA, okay at the scene. Or we've got your DNA on, somewhere pertinent. And then it you've already shifted the burden then on to you're now going to have to explain that. And of course, nine times out of 10 lawyers will go, oh well,they found your DNA. Okay. And so we've talked to some scientists who have, then, this has to be unpicked and it turns out it's not that good a match or it's actually a mixture. Or actually it was found on something movable that could have just been brought to the scene. It could have been transference, it could have been contamination. There's all these caveats, but they're not included in this print out. And people will plead guilty, cases will proceed on the basis of this kind of one line statement that there has been a DNA match and that's already going on because that's cheap and effective.
[00:27:44] Angela: That's frightening.
I find it frightening, Yeah. I, you know, I forensic phenotyping, I want to go back to something you were saying earlier, Matthias. It's really exciting. And I can see why investigators would think, oh, this is going to be really helpful in these no suspect cases where I haven't been able to generate an investigative lead by searching the national criminal DNA databases,right? I don't have anything. Do I have enough sample? Can I go back? Can I test and see what, who am I looking for? You don't have any CCTV. You don't have anybody saying, oh, the person that attacked me looked like this, not that eyewitness accounts are very reliable anyways. And when you look at phenotyping, it is okay on the extremes, right? If I'm talking about somebody who's has ginger hair and freckles and sunburns, right? Those are three of the markers you can look at versus somebody who's on the other end of the spectrum, where they're testing, and you can see that they only carry markers for a very dark complexion. And so the extremes, there's more reliability, but most populations of people are mixtures, right? And if you exclude the caveats from that information, you're doing more damage than good I would think .
[00:29:01] Dr. Matthias Weinroth: Well, the interesting thing about this type of knowledge is, when is it useful? So you, so there are different ways of talking about, what you do with, forensic DNA phenotyping, and, uh, similar technologies such as biogeographic ancestry testing. Some say, well, it's to reduce the suspect pool. But you're not really, you're not really reducing the suspect pool for what you're doing is directing the investigation in a different trajectory. And it's often not communicated in such a way. People talk about funneling, talk about reducing and so on, so forth, but you're not, you're prioritizing, you're setting a priority and just it's necessary just even there already to consider the communication, because if you're closing down some avenues on the basis of such information, which is, as you say, uh, relevant, uh, sorry, it's um, more articulate at the extremes, then at the mid points where you have a higher number of, people, let's say, in in Europe, you have a high number of people with a brownish hair, light brownish hair and middling skin color and,and, uh, blue and brown eyes and everything in between gray eyes and so on, so forth. You know, that kind of information is not really that useful. It gives you an idea of where to set your priorities next, but, it's not in the best case it gives you only that general priority. And the worst case,you take one of these analyses, and it points to a minority. And that's really where a lot of those extremely specialized technologies come in like phenotyping and biogeographic ancestry testing. What you're looking for to use them for is when you're looking for somewhat as a representative zone minority.So, you know, in Europe, you're looking for someone with really dark skin and really dark hair color, and really dark eye color, if you want to use this technology and if it's really useful.Only then is the priority setting, making, um, an investigative difference. Cause you don't have to ask 5,000 people. You might only need to ask 50 people in a particular geographical area. The problem here again is these technologies might have some value, but they also have the chance to create a lot more,problems in the way that they're used. So phenotyping should be about the appearance, but in the practical application, when you're looking for people in an investigation, you don't necessarily just talk about the appearance you make comparisons and conjectures about where that particular person of interest might be coming from. And in certain countries, certain nationalities will crop up. So in the Netherlands some years ago, there was this case where, two, young men were, um, caught on camera. So that's not DNA, but that's still visual uh, association of, externally visible characteristics. And they were identified in inverted commas [i.e. quotation marks] as people from North Africa. It turns out later on, they were actually Polish citizens. The, the thinking associated with crime and what the person might look like, who might be a person of interest in crime, it's often associated with certain nationalities, and that's not based on any crime statistics. That is, pure and simple bias. It's cultural bias. In Germany we had a case where people were saying, oh, there are lots of North Africans, in the marketplace. And, you know, there might be an issue that was a particular issue, around New Year's Eve in Cologne, I think. And it turned out, first of all, that these people weren't really an issue, but second of all, they weren't from North Africa, you know, a lot of the people that were, from, Afghanistan and from Syria and so on, so forth, who just happen to be there, but people were thinking in certain ways. And, and what's to say that investigators are not subject to those very same, cultural biases, and it might lead them to think in a particular way. I can give you the example of the Phantom of Heilbronn which is a case where a German investigation, which then actually expanded over parts of Europe, Forensic DNA phenotyping and biogeographic ancestry testing were used in order to, identify perpetrators that were unknown. A police officer was killed, and, another police officer was, injured. And you know, the, the potential perpetrator was then chased across Europe. And it turned out that a lot of the DNA that was taken at the crime scene, there were certain markers that came up time and again, at other crime scenes in France and Austria and Switzerland, and so on, so forth. So the police were thinking, okay, this is a highly mobile criminal, doing all kinds of different, criminal activities, ranging from murder down to, burglary and then theft on, so on, so forth. So they were asking, Austrian forensic scientists to please do an analysis of the DNA that was found in various different crime scenes. And the the Austrian lab did an analysis of forensic DNA phenotyping, biogeographic ancestry and they were saying, okay, what we're seeing here is a female person, from Eastern Europe and we think that's the person that you're looking for. The police interpreted this as, okay. So we have a highly mobile criminal, Eastern Europe. Okay, must be from this Sintian Roma communities. So now we're looking for people from those communities. Now in the end, it turned out that the DNA that they were chasing and the investigation and that the, analyzed belonged to a Polish woman. So the analysis of a Eastern European woman, based on phenotyping information biogeographic ancestry information was fairly correct. However, this woman wasn't a criminal, at least not involved in any of the crimes that, the DNA was found at, that she worked in the factory that was packaging, the cotton swabs that are being used, in in swabbing. Her swabs were contaminated with her DNA. So the forensic scientist might say, hey, our analysis was successful. You know, we did say it's a person from Eastern Europe. It looks like this, and this and this, for the police, of course it was unsuccessful this analysis because it led them in the completely different way. They suspected a Roman Senti, and we're looking there and it turns out, you know, this was not at all relevant, not at all useful for their investigation. So this is a really good example where we can see, uh, where technologies might be useful, but they don't have to be. In fact, they can be really quite damaging if they're not used to in the correct way..
[00:35:26] Angela: I remember hearing all about this, I was in the US at the time. So do you think that is it, is it a matter of, we shouldn't be using these technologies or we shouldn't be solely relying on them? So for example, if you're going to go down the track of forensic DNA phenotyping biogeographical ancestry to help you identify a suspect or to narrow your pool. Should you have two different groups working on the investigation? One that's following the genetic investigative lead and another that is remaining with standard investigative techniques that are unaware of the genetic side of things. Because it can be a useful tool in certain circumstances, if the recipients of the information are properly informed and understanding all of the caveats, but then you do have that bias effect. So how do you use the technology and prevent the bias?
[00:36:24] Dr. Matthias Weinroth: Yeah. it's a question of training to a degree, but also, uh, training of investigators, as well as of, forensic scientists. The, because there are some positive examples of how these kinds of technologies were applied with very careful community relationships applied as well. In the way that these, the, the findings were communicated, the findings were followed up on with the Turkish community, where there was more careful, communication by the investigators. So it didn't lead to something that was informed by what you could call moral panic. Basically saying, oh my God, you know, here we have this terrible crime and now we have to use everything in our power, regardless of the damage we might cause, to use it. You know, this is the usual argument of, well in the balancing of, catching a killer protecting innocence, of course we have to always, on the side of catching the killer. But that's that's a really unhelpful balance. It's a really unhelpful, unhelpful notion of we have to balance particular rights against each other. So the it's a really it's it's a, I don't know. I think these technologies can be useful, but they need to be used very, very carefully. And we're trying, in our paper is not to prescribe when and how to use it, these kinds of technologies, even though we do talk about three different types of technologies, we raise the point of what might need to be considered in the development and values. Carol... Well, one of the things that, and you'll be where to where this Matthias, is that I'm always asking about effectiveness and almost bringing to the table a sense of realism, which is what we're kind of asking for. Well, there's one important point to make is that I don't think, and this is very often set up like this and, and for my whole career, I've been seen as being pesky or a nuisance because the notion is that I somehow don't want to catch criminals or, somebody once said at a meeting, or you're obviously a fan of rapists, which is just it's nonsense. The idea that if you want to, if you're on the crime control end of the spectrum, you want to catch criminals and, you know, and murderers and so forth that somehow you're on a different end of the spectrum to me, where I actually want things to be done ethically. The two things aren't in conflict, they shouldn't be in conflict. We're not saying, don't do these things, please don't catch these criminals. We're saying, can you do that but could you do it ethically please? And also of course the important thing is if you do things ethically from the start, you will probably more likely to then end up with an effective technology that's something we can use. You know, if you do something, completely unethically, then somebody somewhere is going to turn, okay, now you can't do that. look what happened with Marper. We started creating this massive DNA database, putting everybody on it, and then of course we had to go, uh hang on a minute that's not actually lawful. We breaching a whole bunch of human rights there and we had to roll back. So, one of the things I would really like is if we let's learn that lesson and say, let's do this, increment it and do it ethically from the start. Let's not get carried away with ourselves. Let's not think, wow, are we going to catch all these criminals because now we can tell which ones have got ginger hair, because that's just nonsense. You know, the idea we're going to catch a load of criminals cause we can tell which ones have ginger hair and blue eyes.
[00:39:59] Angela: The gingers already have enough problems as it is.
[00:40:02] Dr. Carol McCartney: And let's be realistic about effectiveness as well, because let's think broadly about, we have to make priority decisions about where we invest our resources. So is this a sensible place to invest our time and resources? Or is there something that could be more effective? That would actually solve more crimes and so forth. So yes, in, and I think of course these technologies are probably, you mentioned no suspect cases or what we call cold cases and yes, it's vital that, we, we have technologies that we can bring to bear on these cold cases because they're obviously, it's awful that there's murders that go unsolved. But do we then roll out or say, oh, we can do photo fits of criminals now and so forth and spin this kind of idea that we can do things that are going to be of general use, that they are very limited. They are very expensive and time consuming and are really only have limited use. That I think becomes almost a societal question then. Do I want my resources effort, the policing and the criminal justice system put in these technologies? Or actually are there other priorities because we have, if money was limitless, if resources were unlimited, yes, go for it. You know, as long as you're doing it ethically and so forth, but that's not realistic. So all these questions, they're not in conflict to being ethical and catching criminals, you can catch criminals, but you have to do it ethically. And there are societal questions that I think you can't over oversell this technology as being a silver bullet, that is. You know, I've done a lot of work with other people as well, Aaron Amankwaa, about the effectiveness of the DNA database. And when you actually have a conversation where people about how many people they think we are catching using just conventional DNA testing, they wildly overestimate it because we've been sold this idea that DNA testing for a start is objective, reliable, infallible, and so forth, which is not true. And that somehow we're catching all these criminals and that's not true either.
I mean, we are catching some and in some cases, the, you know, a lot of, post-conviction testing where people are exonerated. And it's you find the true perpetrator was actually in the database for some small petty crime and you just didn't identify them.
[00:42:40] Dr. Carol McCartney: Yeah. One of the most frustrating things of course, is that we do have this technology and sometimes frustratingly it isn't used to good effect. I mean, we had a massive case here years ago, a young mother was murdered horrifically in a park in front of her toddler, Rachel Nikell on Wimbledon Common. And we got criminal profilers involved and it was this whole big investigation, which basically we ended up putting on trial, the wrong guy, meanwhile, all the time there was a suspect who'd been named by his own grandmother. He was, known to have committed other crimes and so forth, and he should have been a suspect and the police just kept somehow never quite getting round to taking his DNA. So they went to his house even, and he wasn't in, so they just left again. And then, so there was all this went on. Meanwhile, we're busy prosecuting an innocent guy. Who fortunately was not convicted. But yeah, it took ages for us to actually get around, to take in the DNA and testing it of this guy that who seemed an obvious suspect. And then of course it matched and in the interim he had killed another mother and her daughter. So because of the tardiness or whatever it was about the police didn't get around to taking the DNA of this suspect, another couple of people died, so it is frustrating. So it, isn't just the technology. You have to make sure it's being used properly. So, you know, the systems are integrated. So if you have a suspect that you take his DNA and then it's tested because otherwise, all the technology and laboratories in the world are wonderful, but if they're not being used properly. I always go back to the wonderful, metaphor that was used by, Ken Pease and Tim Leary said about the sandwich. If you've got a wonderful filling, DNA testing, but two slices of stale bread, which is policing and the trial system, you've still got a rubbish sandwich so got to get rid of the stale bread.
[00:44:46] Angela: I think I'm going to have to title this episode, a stale sandwich. Um, you touched on a lot of things in there. It doesn't do you any good to have this amazing technology if you're misusing it and prosecuting the wrong person. I mean, like I said, we miss things and more people end up dying because your focused in the wrong direction, because you're not understanding or you're misapplying or you're overstating what this technology means.
[00:45:17] Dr. Carol McCartney: Yeah. I mean, we have had the odd case in the UK at more so in the US. There was a wonderful case here where a guy got convicted of a robbery, quite a serious robbery because he was... there was basically a chase. He was caught wearing these trainers and that, these trainers then matched at the crime scene, the treds, the footwear impressions and so forth. And he swore blind that someone had been running away and stopped him and made him put these trainers on. That does sound like a crazy story. I admit you go yah right that happens all the time. People come up to me in the street and tell me to swap shoes with them. But it turned out the guy got convicted. Do you know what it turned out to be true. That they did DNA testing on the inside of these trainers. And they matched it to another guy who, you know, had a string of robberies and was matched the, because they had a description as being like a six foot tall black man. And this was like a five foot tall white guy. That was convicted. And so we, we overlooked that bit cause the, you know, the trainers, the trainers sort of thing, and who would have said, I'll let us DNA test the trainers. It was only when the guy was in prison and he'd lost all his appeals was that our lawyer went okay, let's, do this testing and yah he was subsequently exonerated. Obviously for exoneration cases and so forth, we've had cases, many, cases where, DNA can exonerate people. Obviously it would be preferable if we could exonerate, just not convict them in the first place...
[00:46:55] Angela: yes.
[00:46:56] Dr. Carol McCartney: ...would be far preferable.
Well, even all of these technologies that we're talking about, whether we're talking about traditional DNA profiling with STRs or some of these newer technologies and newer methodologies, like the forensic genetic genealogy and all that sort of things, the, you kind of forgetting the fundamental fact that just because somebody's DNA is on something is not equivalent to them having actually committed a crime. That, the source of the DNA is completely separate from the activity. And it has to be considered as a separate mutually exclusive proposition. I mean, we do have not rules as such, but a convention that say that you cannot convict somebody with DNA evidence alone. Okay, but that has been stretched in some instances too well, we've got your DNA on something suspicious. And actually you can't explain it and you've got previous convictions you're a bit of a dodgy character, we don't have any other evidence, but hey, let's ask a jury. And if you can't convince the jury, there's a, a viable, an explanation for why your DNA is on this, then you know, well the jury convicted you and there must've been something else. So it can be stretched. Yeah, but, convicting people on DNA alone, I think should always be a no-no. If somebody is guilty of a crime and all we can find is a DNA match, that's got to raise a question.
[00:48:29] Angela: That's that goes back to what you were saying earlier. Matthias about the context, stressing, the context being important. I've always told the prosecutors, when I get the pretrial meeting, you cannot consider your DNA evidence in a vacuum. And if that's all you have, you don't have enough. So it, whether or not that gets across to people, I don't know, because a lot of times they have a pretrial meeting everybody accepts the DNA and you don't even know how the case turns out. So sometimes it's not exactly fair to the person that's doing the work in the lab. They're excluded from either side of that sandwich and maybe they should be more included in those parts of the process to help guide the investigators and the prosecutors to make sure that they really are understanding the information they're trying to give them.
[00:49:17] Dr. Matthias Weinroth: I mean, that, that's a good point, about this misconception of substantive equivalent. So, you know, so yes, uh, profiling and databasing, and simple matching is, is quite well developed highly reliable, but that doesn't necessarily mean that other DNA based technologies are as equally well-developed and as equally, highly reliable. Um, you know, so just to say, we've got the DNA and that's where we can say this is, you know, that doesn't hold if you've just got trace, that might give you information on what the person might look like, or that is highly reliant on cultural denominators, such as, geographical ancestry ideas. Um, that's to say that someone might come from, East, East Asia say, is in itself. Not that, not that really useful, unless you're looking for people whom you imagine to look like what we think are East Asians. And that's a cultural assumption of what an East Asian person might look like. So we need to be careful in thinking about, you know, DNA technology isn't always, isn't the same as any other DNA technology that the substantive equivalent isn't given, isn't something we can assume just because DNA matching works, uh, DNA phenotyping or forensic genetic genealogy, or, forensic epigenetics is something that we can use, quite as, confidently, as DNA profiling generally. I think it's it's quite important to keep that in mind. And that might... what's quite important is, in the public debate might get lost to this idea that these are new technologies, which are not the same as existing technologies. These are emerging technologies, you know. Genetic genealogy. Yes. We have a few cases where they have worked. But how many cases, exists that they haven't worked in? We don't really know because, failures are not communicated. We tend to hear about individual successes, and that can be really really good and helpful, but it would be good to know about the failures as well. You know, the Phantoms of Heilbronn where different technologies, have not worked. I'm not saying that these are not being discussed in the scientific community. I'm pretty sure they are, but they're not communicated to those who might be under the impression that these new and emerging technologies are just as reliable as existing, DNA technologies. And I think that's, that's that's an issue. That's a problem that we've seen time and again, in public debates where expectations have been raised quite significantly, without consideration for the detrimental effects, on policing, on the perception of science in policing, on the perception of the legitimacy of policing using certain technologies, even on the on the perception of, those people who might be subject to the use of these technologies. So that's not as Carol was saying earlier, it's you know, we need to protect, we want to protect the rapist and the killers, this is completely wrong. These technologies can be useful, but if you don't use them carefully, they can cause a lot more harm where innocent people get hurt or even killed. And then you have another case of a murder, which, is might've been caused by someone taking the information in the wrong way. For example, police officers in the US shooting at someone because they are black and they're holding a mobile phone in their hand.
[00:52:42] Angela: What you're saying there actually, not the part about the person with the mobile phone, but the information about the using this new technology, it does harken back to the beginning of DNA databases with just the DNA profiles in them to generate investigative leads and the argument that exists and continues to exist in the US there've been several pieces of legislation around it, where if we'd had this technology available to us, we would have prevented eight rapes from happening. And now it's with rapid DNA, you can have people's DNA samples collected at intake at the custody suites, and you can immediately search that person, generate a DNA profile and search it against the database and identify somebody as soon as they're arrested as being linked to these crimes, rather than waiting through laboratory backlogs to get this information to happen. And only you can use it for crime prevention, as well as making sure you've got one aspect of catching the correct person, rather than going 200 miles down the wrong road and looking for the wrong person, and wasting a lot of resources and stuff. I think some of the things that you're saying with... you're talking about the ethics and following the right process to ensure the integrity of the investigations. And by highlighting these issues, you're not saying that you want this rapist to go free. You're saying you want to catch the person who actually is the rapist and not this other guy who happened to be running down the street and somebody made him put them on a pair of trainers.
[00:54:21] Dr. Matthias Weinroth: Absolutely.
[00:54:22] Angela: I'm mixing up our stories there. Right. Um, but in, in your paper, I'm conscious of the time now. So I wanted to say in, in your paper, on the ethics as a lived practice, you highlight, three key lessons to help forensic geneticists navigate the future with all of these technologies as they continue to develop. Would you like to share and touch on those.
Sure. Yeah, let me just, preface this with, um, the rationale for the paper really was, we can see, which is really good, a lot of technical harmonization, with these new, with, existing forensic, DNA technologies, as well as emerging ones. There's a harmonization of methodology and, um, an analysis approaches across forensic laboratories and, uh, service providers, uh, not all, but the responsible service providers, they share the same kind of information. They share the, the kinds of methodologies that they're using and they're disclosing them. And what we need really is also a shared strong, ethical code, across forensic laboratories across, uh, forensic service providers, just as though, as a shared kind of approach to methodology, there should be a shared, like a strong, universal ethical code, and the, uh, the way that we approach ethics as a consideration of, standards of rights and obligations, and that applies to, scientists and, and their rights and their obligations as scientists. So the three lessons that we, boil the conversation and the paper down to where, the first one is really to think ahead. To and we call this anticipatory capacity, to, to be able to, to anticipate, not to expect, but to anticipate based on experiences, based on awareness and based on communication with others, the kind of, impacts of, research and a forensic work. So scientists need to be able to anticipate how technologies will impact on individuals and on, on groups of people who might be drawn into investigations. And they also need to consider the wider community relationships, and criminal justice more more generally. So it's really, really important to embed forensic science and forensic genetics within the criminal justice system and within society. It's not something that exists outside of these two. And often we, we kind of see you there saying, oh no, this is science, this is objective, this is this is, facts, but this is, scientific information generated by human beings within society for society within the criminal justice system. So it's really really important here, that forensic geneticists, and that's the first point, again, need to engage with ethics at all stages of, the data life cycle. And they need to be able to think ahead. The second point is that it's really important to think about the power of these technologies to cast suspicion. And I think that's, it's always seen as something really positive, which it is, but it can also have unintended consequences. So it's really really important to think about the power of forensic genetics, forensic epigenetics, to cast suspicion over a wide range of individuals and communities. For example, those who may share visual, characteristics or, family ancestry or lifestyle habits, and, forensic geneticists need to be aware of the, the active role that, uh, their research and technologies can play in either strengthening or threatening, fundamental legal principles, such as the presumption of innocence, the equality of arms and the legal burden of proof. The third point is about clarity and open communication about the real world, capacities of technologies, such as the ones we've discussed today, which is really, really, important, and which we often don't see that, clarity and open communication about how these technologies, what they can deliver outside of an ideal case laboratory, proof of principle situation, and, just a testing where you already know the answers. And you're just trying to see what you come to. These, these are important, these kinds of tests, but we also need to see, examples from work with, samples from outside and those are happening, but we need the clear communication here about the limitations. Yeah, so, so the point really is a clear articulation of technologies. And that helps to build legitimacy, and public trust, around these technologies and around the work of forensic science, forensic genetics, that it can contribute, and deliver in the criminal justice context.
[00:58:48] Angela: Yeah. There's we didn't touch that much on the epigenetic side of it, in the lifestyle choices. but I'm going to put a link to this paper in the show notes. So people will be able to read, the full information that you guys have worked so hard to put together. And I know that you've also been building on a lot of the principles in here and continue to do work around these areas. I guess I'm already talking to you, do you have any closing remarks and then I'll pass it over to Carol and let her have opportunities.
well, I, I brought a lot of the important points together just in the summary. But what I also want to say is, ethics, and I think that's a really important point, ethics is not the law. Ethics is not religion. Ethics is not what society accepts. Ethics is not, I feel therefore that's right. Ethics is about principle at reasoning about actions. It's about standards of rights and uh, how one should behave. And that's really really important. Just because something is legal doesn't necessarily mean it is ethically sound to do, doesn't necessarily mean it's legitimate. So a really really important point that I want to kind of add on to the summary of the paper that you kindly gave me the time to do.
[01:00:00] Angela: Absolutely. My pleasure. did you have anything?
[01:00:04] Dr. Carol McCartney: Yeah, I think they're really important points to end on. and just really a plea that this is an ongoing conversation, that the point about communication and that we're not trying to delegate that all forensic geneticists have to be ethical and it's all up to them. we want to be involved, all the slices of bread in the sandwich and the feelings should all be working together. So this is a conversation that should be broad ranging, involve lots of people like, you know, I'm a lawyer Matthias, I'm a social scientists and so forth. But forensic geneticists are a key part of the conversation and I would urge them to, to join other conversations, but also invite us into your conversations. And likewise, so that, it's really exciting and yes, we all want to catch criminals, so let's just see how best can we do that? And this is taking that conversation forward. It's not trying to hamper it or slow anybody down. so yeah, I hope that, yeah, it's not seen as challenging, but actually exciting.
[01:01:14] Angela: yeah, I'm hoping that with this episode, we'll be helping start that conversation and make sure that people in the field know that experts like yourselves are available and happy to help when we come across these difficult decisions. So we've solidified the science. We know what the limitations are for pristine samples. We don't know what it's going to happen to it when it goes out into the wild. And maybe we can't conceptualize those because we're not as, we don't have the same foundation in the social aspects of the implications of what we're doing and we need help. We need external help. And a lot of those, a lot of those areas, and well hopefully you guys will actually start having more people in the genetics field reaching out to you and talking to you. I know you've done some collaborative work and inter interdisciplinary work, before, and hopefully this will continue along that line of conversation. I will be sharing your biographies and links to, a lot of your publications. And especially with some of the, you know, we talked about the new Nuffield Report and also I want to link to some of the other papers that you guys have done, and the interdisciplinary work with EURFORGEN and things like that too. So with that, thank you guys very much for taking the time out of your schedule to be on the show today. I really do appreciate it.
And if in the future you guys have some more work you'd like to share, I would be more than happy to have you back again. I really enjoyed the conversation and I think we probably could talk for a couple more, at least another hour on this topic. But thank you again
[01:02:59] Dr. Matthias Weinroth: Thank you.
[01:03:00] Dr. Carol McCartney: Thank you.
[01:03:02] Dr. Matthias Weinroth: Absolutely.