The best books on Forensic Science

Thank you for selecting five of the best books on forensic science. I’m sure many of our readers must feel they have a notion of what forensic science is from crime novels or crime drama on television—but as you noted in our earlier correspondence, you want to get away from what you called ‘fast-food forensics’. Could you tell me more about how the image of forensic science differs from the reality?

Yes. It’s hugely different. Over recent years, television has got a bit more accurate in the procedural sense, because there are so many forensic science courses—people believe they know about managing a crime scene, cordoning things off, all that kind of stuff. But the truth of the matter is that only a tiny number of people have ever been to a crime scene or been in a forensic science lab. Only a tiny number of people have ever actually been in a court, let alone given evidence. People fill these gaps with their imaginations, and this imagination has just gone wild in the last twenty years, and a lot of it is just untrue.

The uniqueness of fingerprints, for example—we’ll get onto this when we discuss my last book choice. It’s hugely convenient to believe fingerprints to be unique. What that means is, as a witness you go into court and you’re unchallengeable. As a lawyer or a cop, you’ve got a cast-iron case. Biometrics is another good example. People are sucked in by biometrics, they just believe in this technology. It’s so much easier to believe than it is to confront the reality, which is that it sometimes works and sometimes doesn’t.

You have worked in forensics for over forty years, including on some very high profile cases. You will be publishing some of these experiences in your upcoming book, Murder Under the Microscope. But what drew you to the field initially?

You know, it’s a very disappointing answer. It was basically the only job that I was offered. We’re talking about the late 1970s when employment opportunities were pretty poor. I just took it up.

I graduated in biochemistry and wanted to work in science. But one of the ironic things is that a great deal of what we call forensic science is only vaguely science. If you go in as a scientist, if you’re actually interested in science—things that happen in labs, and the kind of techniques and models that scientists use—then your first exposure to big chunks of forensic science is hugely disappointing. Because it’s highly subjective; often quite poor science in many instances, and in some instances no science at all.

There are, of course, exceptions. DNA is an exception. Chemical identity—identifying drugs, and things like that—is an exception. But a good deal of the rest of it, and this is a big theme of my writing, is craft. It’s a craft you can learn, it’s a craft you can be good at or bad at. It’s a craft you can make mistakes in. But, of course, the minute you have the title ‘forensic scientist’, people assume you have the authority of science behind you.

The courts treat science as a different species of knowledge. Because you have more authority, you’re harder to tackle as an expert witness. So, it’s very comfortable if you can get away with it. But actually, some quite simple questions can quickly undo the credentials of an expert or their evidence. Now, scientific evidence can be useful, but it has limits. And, of course, a lot of fictional representations don’t recognise those limits, and that’s where people get the bulk of their information on forensic science from.

The first forensic science book that you want to recommend is, I believe, a personal favourite. This is The Killer of Little Shepherds: A True Crime Story and The Birth of Forensic Science by—a past Five Books interviewee—Douglas Starr.

It tells two intertwined narratives: one is this really horrific serial killing in 19th-century France; it also tells you about the origins of forensic science and policing. If you go back to the early 1800s, that was the time when police forces around the world were being organised. Up until then it was private—just people who were hired as security, with no formal legal basis.

So The Killer of Little Shepherds is set around the time that the Sûreté was formed…

…the French criminal investigation bureau…

…and a man called Vidocq became a very famous detective. He’s kind of a French equivalent to Sherlock Holmes, although Sherlock Holmes is fictional while Vidocq was a real person.

It’s a horrific story, even by my standards: a man killed 11 people—mainly children or teenagers, boys and girls—and disembowelled them and raped some of them. He moved around southern France, and sometimes outside of the south of France. This is one of the big lessons which resonates with my career. Even in modern times police forces are not that good at cooperating. I’m not saying they don’t cooperate, but personalities often get in the way; everybody wants to solve the case. There are tensions and rivalries as you see in any other organisation. So one of the reasons this man could act under the radar was because he moved around. When it was happening in a different place or legal area, it became somebody else’s problem. Eventually he was caught more or less red-handed.

“A great deal of what we call ‘forensic science’ is only vaguely science”

In the French case, a man called Lacassagne, who was a professor of forensic medicine at Lyon University, played an important role. He was the tutor of an Edmond Locard, considered by many—not quite accurately—to be the founding father of modern forensic science. But he was one of Lacassagne’s pupils. Lacassagne was saying things like: ‘pay attention to details’, ‘observe carefully’, ‘try to be objective, theorise, then test the theory’, while the cops of the time, would just be running in and splashing about, knocking things over.

The things that really struck me about this book were that it’s really well written, incredibly well researched, and it’s a fantastic story.

What more could you ask of a book?

Douglas Starr is an American who started as a journalist. How he researched it in France… he must be a fluent French speaker, or reader, anyway. Many of the problems that arise in this case I highlight in cases in my book, which will be published in October, of recent cold cases. For example, Robert Black killed four children in the UK between 1981 and 1986 before he was arrested and convicted. One of the reasons he was able to do this was because he moved around the country operating in different police force areas.

People are still making mistakes or forgetting things, not cooperating, misunderstanding or not paying attention to the evidence, and so on and so forth.

Alexandre Lacassagne, wasn’t he the originator of the study of blood spatter?

He contributed to the development of blood pattern analysis. But these things weren’t happening around the time of the book. People were only just beginning to see that blood spatter might tell us how a person died, whether the body was moved, what weapon was used, stuff like that.

Lacassagne was a medic, a professor of what he called criminology. He was more focused on bodies, but he was also interested in things way beyond that.

You mentioned earlier the taping-off of crime scenes. I think this is the starting point for our next book on forensic science, Murder and the Making of English CSI. It’s by Ian Burney and Neil Pemberton.

They write the history of this notion of a ‘crime scene’—as a special place, a social creation, as an idea. They explained something to me that I’ve always struggled to explain to other people, which is: what’s a crime scene like? What can a crime scene be? My answer now would be, well, a crime scene is an idea, a concept. It can be anything you want. It can be a spaceship, it can be a car.

As a scientist, I work very closely with social scientists; much of my research nowadays is in the hinterland between natural sciences and social sciences. And this book explained things to me that I’ve misunderstood for a very long time.

They take only two crime scenes, but they look at them in a great deal of detail. They point out words and phrases that we now use routinely and assume that everyone understands them. So: the whole notion of a crime scene as a geographic location where something happens. They talk about how we demarcate it and start to look at it; how different roles start to develop for looking at different bits of it; how people start to take notes and record diagrams. How all that requires some kind of cooperation and discipline and structure—counter to most crime novels, or at least a certain genre of crime novels where you have the ‘great man’, the detective who comes in and somehow sees things that no one else can see.

“Crime scenes are not just procedural. They’re cognitive, too”

Burney and Pemberton point out that it is, in fact, a collective enterprise. Yes, people have insights, but the bottom line is: people can get it wrong. The way you keep on track is usually to expose your ideas to some kind of reflection or criticism, to talk to people.

And they say something very explicitly: that crime scenes are not just procedural. They’re cognitive, too. You look at them and think: ‘how would somebody get in here?’ or ‘how would you get out of here?’ You can come up with some really crazy theories, test them and then abandon them…. That’s quite a complicated business to do in a group. I’ve been in briefings where people have come up with absolutely wild ideas, and you want to say that to them—but you let it run. The evidence will eliminate it eventually, and they will quieten down.

Between the 1920s and 1950s, you get the development of the murder bag: the case, the kit with forceps, bags to put things in, and labelling, and the chain of custody. These are all ideas that were consolidated in this period. Continental Europe was way ahead at this time. A man called Hans Gross—whose name seems to have been forgotten in the UK—was really the first person who thought about the whole business of a ‘crime scene investigator’ who pays attention and thinks and hypothesises, and then tests those hypotheses. A very rigorous, rational kind of process.

I think they also look at what they call ‘body-centred’ forensics.

Yes. As I’ve written myself, a body is a defining feature in a crime scene. If you go to a murder scene and there’s no body, then it’s a murder scene and you get on with it. But if there’s a body there it changes the whole thing.

Firstly, it brings in a certain practical urgency, because everyone’s waiting for the body to be taken away for a post-mortem examination. Also, you see the context and—as a blood pattern expert—I would always say I need to be there when the body’s there, to see exactly where it was. But quite often this isn’t possible.

The body—in a homicide—is the literal embodiment of the crime. So there are all these practical and cognitive, imaginative things associated with it.

Does a forensic scientist work closely with forensic pathologists? Is there much interplay between these two roles?

Yes. As a biologist, I would. Most of the work I’ve done would be in what are legally termed ‘offenses against the person’. If I were a chemist, I would do more road traffic accidents, blood alcohol analysis or burglaries (shoe marks, glass fragments), whereas most of my work was in sexual assault and homicide, all of which is dealt with in the higher courts.

So, yes, I worked with pathologists and I would very often want to know what injuries a person had sustained, to make sense of how blood was distributed on clothing or on a weapon. Sometimes I worked with them very closely, at the post mortem, and sometimes remotely via phone call, or reading a statement or report.

Let’s move onto book three, which is Maggie Nelson’s The Red Parts: Autobiography of a Trial. I love Maggie Nelson’s writing—her beautiful fragmentary memoirs Bluets and The Argonauts—but they are very concerned with literature and philosophy and art. I’m not sure I ever expected her to turn up on a list of forensic science books! Tell me about The Red Parts, and why you have chosen to include it among your reading recommendations.

I can’t possibly do justice to this book. I really can’t. It’s such an extraordinary book. She exposes herself as a writer and as a person. There’s no boundary for her whatsoever.

We are very much in the modern world, in cold-case review territory. I think this was published in 2007. It’s an extremely candid memoir, written by someone directly involved in a murder case—the victim was her aunt. But also, because of the timing (she comes into the case 35 years later), she can be a dispassionate observer as well.

So, Nelson’s aunt was a first-year law student who was brutally murdered in 1969. It was an unsolved crime for decades, but was reopened in 2004 when a DNA match identified a new suspect. Nelson had previously written about the shadow this horrible crime cast upon her family in Jane: A Murder.

Yes. In The Red Parts, she goes to court, she meets the cops, she sits through the trial. She talks about the decision whether to go to court or not, whether to look at the images of the post-mortem, which are being projected onto the court wall. She describes all the primary materials, sometimes in a very factual way—and then she goes on to engage with them in a very different and emotional way. It’s beautifully written.

One thing that really resonated with me was when she talks about how many people are just completely unwilling to face their revulsion about what’s happened in cases. We’re just completely unwilling to confront the fact that someone has done something to someone else which we cannot possibly conceive of. I don’t recall exactly what she says on this, but my take on it is that many people immediately resort to the notion of evil, which puts clear water between this and humanity. You know: ‘They’re evil, and therefore I don’t need to worry about it because I’m not evil.’ We can rationalise it away.

“Nelson expected to be confronted with a monster, and she just sees this innocuous person”

There’s an interesting link to Hannah Arendt, the philosopher. She used the phrase “the banality of evil” to describe Eichmann during his trial in Israel. Maggie Nelson writes: “Where I imagined I might find the ‘face of evil,’ I am finding the face of Elmer Fudd.” She’s saying the same thing: she expected to be confronted with a monster, and she just sees this innocuous person and doesn’t know what to make of it.

She comes up with this idea of bearing witness to the routine of the state: this process as a machine of prosecution and justice. All of this had enormous resonance for me. It’s a fantastic book—of the five, it’s probably my favourite.

It’s interesting that you selected it, given that you are an expert in the field. What did you take from reading a layperson’s account of a trial? Is it important to understand the emotional stakes, on the part of the victim and/or their family?

One of the things you have to learn quickly—or at least, I did—is that emotions should have little part to play in the work, although I can now engage with them in my writing. Not in the way that Maggie Nelson has, but at least to recognise them.

While writing, I came across some really interesting cases that show that the police are far more emotionally driven than they would ever admit. There’s one case in particular—the murder of Rachel Nickell—where it’s very plain to me, having researched and written about it, that the a key driver of the investigation was affect. It wasn’t rationality—it was just the horror of the case. So emotion plays an enormous part here.

I imagine it is no different to being a nurse or a doctor: if you get hugely emotionally engaged with your patient, I can’t imagine you would do your job very well. So you do need to manage that, I guess, otherwise you end up like some people I know who, thirty years into the job, end up with PTSD. They get flashbacks of crime scenes they were at.

Maggie Nelson is a layperson, but is also a very good observer. She used lots of factual stuff, but at its foundation it’s sort of ethnographic. She saw and she watched and she reflected. She doesn’t give us any flights of fancy in here—that’s not to say she didn’t use her imagination. What’s in this are some very pointed reflections on what people did then, what they are doing now, whether this process was the right thing to do, whether you should confront it or run away from it.

It’s also riven with family tensions. The big issue is whether they’re going to be in court for the verdict: guilty or not guilty. She describes the tension in that process. She’s just a great writer.

Fantastic. Well let’s look at book four on your forensic science reading list. This is Killer in the Shadows: The Monstrous Crimes of Robert Napper by Laurence Alison and Marie Eyre. This book deals with a case that you just mentioned, the brutal murder of Rachel Nickell and a related miscarriage of justice.

Yes, it’s a big chunk of my next book. But yes, Rachel Nickell was killed on Wimbledon Common in 1992. I got involved in 2006, I think.

Napper killed Rachel Nickell, he stabbed her 47 times. He was still stabbing her after she was dead. In broad daylight, in the middle of London on a summer’s day. She had her two-year-old son with her at the time.

And the police absolutely fucked up the investigation, pretty much from top to bottom: fitted someone up. They used this man called Paul Britton who was a forensic… let’s call him an offender profiler, which is a term they used then, but nobody would use now. Britton was naïve, but the cops were the opposite of naïve. This profile led them to Colin Stagg, and they just put on their blinkers after that and didn’t listen to anything else.

“The police absolutely fucked up the investigation, pretty much from top to bottom”

Stagg was tried at the Old Bailey in 1994, but the judge kicked out the case. He was just appalled by it. He really leaned on the police and the prosecution. He obviously thought the case was nuts and should never have brought near a court. Then the case remained unsolved.

Apart from the Stephen Lawrence case, it was the single biggest investigative failing in the many failings of the Metropolitan Police. I used to work in the Metropolitan Police forensic lab.

Now, in 2006 I was asked to review this case. Napper had actually killed another woman the year after Nickell: Samantha Bisset and her four-year-old daughter, in absolutely horrific circumstances. So horrific that at least one of the persons at the crime scene never worked again. Utterly horrific.

Just awful.

The upshot of the story is that Napper came to court in 2008. Napper is now in Broadmoor.

This is the high security psychiatric hospital in England, where a number of notorious murderers have been detained.

There was no trial. He pleaded guilty to the murder with diminished responsibility. By that time he had already been convicted of the previous murders. They reckon he could have raped as many as 100 women. Given what we were saying earlier, it’s interesting that they used, as a subtitle, ‘The Monstrous Crimes of Robert Napper.’ I think most people would have no trouble in describing Napper as a monster.

Laurence Alison and Marie Eyre, I think Alison’s student at the time, are much more interested in Napper than the crime; I was reviewing the crime—although I always want to try to figure out what he was doing. During my review I had a really smart detective from the Met who was my interlocutor; they wanted an independent review because they had had so many problems with the case. The main problem was that evidence could potentially have been contaminated, because they didn’t know cases from Napper were actually in the lab at the same time as the original Nickell case—because they hadn’t linked the cases.

The situation in the lab was so bad, that when they told me what had happened I basically thought to myself: ‘You’re fucked. There’s no way on Earth this can’t be contamination.’ Then they said to me: ‘If you say this is potentially contamination, there will be no trial, no prosecution.’ So no pressure then!

But at the end of the case, when I’d reviewed it—and it was a horrifically detailed and complex thing to do—I decided that it probably wasn’t contamination. I couldn’t possibly rule it out. You can almost never rule out contamination, there’s always the possibility, but it was unlikely.

“This guy targeted women with children. That’s really, incredibly unusual”

It takes a long time to get your head around these cases. There’s a huge difference between getting directly involved in a case and going to the scene and seeing it first hand, and just having a lot of documents and photographs. So I was constantly meeting up with the Met detective to ask, ‘what about this?’ and ‘what about that?’

There was an interesting conversation we had while I was building up for my review, when I said to him one day: ‘this guy targets women with children. That’s really, incredibly unusual. I’ve never come across this before.’ And he told me, smiling, ‘I’ve been told not to discuss that part of the case with you, because we want a completely independent review.’ I said, well, I didn’t need to know, I really didn’t. But it seems to be rather obvious.

All the documents in the case were disclosed to me. To my surprise this included correspondence between the Home Office, the police, the Crown Prosecution Service and the Forensic Science Service (FSS), who had carried out the early forensic work. Later the police transferred the case to another forensic lab, which found DNA evidence that the FSS had missed. The case papers showed how the FSS tried to cover up their failings. I also got the distinct impression that the Home Office were trying to distance themselves from the row because they were concerned that it might impact on the planned privatization of the FSS.

Alison really opens up what might be in the mind of Napper. And we can only speculate, but it’s a very dark place, a very strange world, where there’s this connection between sex and horrific violence and humiliation and torture. It’s pretty horrible.

I’ve used a quote from Alison as an epigraph. Because in the outside world these people are described as monsters who are hiding in the dark shadows. But Alison says, no. They’re right in front of you. There are serial killers there in plain sight. I thought, oh, that’s cute. That’s much more chilling that the man hiding in the shadows. Napper worked as a cleaner in a school. They all thought he was a bit odd, sure. But there he was in plain sight.

What a horrifying case. You’ve touched on this a little already, but could you talk about the risk of post-traumatic stress in this job, and what you have to deal with on a day-to-day basis? I mean, is this something that a person considering going into the field should worry about?

I wouldn’t over-worry about it, because you are a bit remote from the really serious action. You do have to go to crime scenes, and there are bodies—and some of those bodies can be in pretty horrible circumstances. But that’s only a certain type of forensic scientist.

Nobody would ever make you go to a crime scene. You could choose not to. It would curtail your work a bit, but the people who really tend to suffer from PTSD are cops and crime scene investigators who see this day in, day out.

How many bodies have I seen throughout my career? Not that many, to be honest. I don’t know how many, but perhaps only 30. I’ve seen some pretty horrible photographs—and even photographs could set you off. But even then… it’s a subset of a subset of forensic scientists who get exposed to this kind of stuff, whereas cops and crime scene investigators and pathologists are exposed all the time. So I don’t think it’s a huge problem for forensic scientists.

“If you engage in a highly emotional way, you are probably in the wrong business”

But if you have that personality type, where you engage in a highly emotional way, you are probably in the wrong business. You do need to be able to distance yourself from this. You need to be able to cope with the details of how somebody might have been raped or tortured or murdered, and you need to be able to listen to what has happened and think: ‘what does that mean for my investigation?’ Only on a tiny number of occasions have I been involved in cases where I’ve heard the details about a case and felt that I didn’t want to hear any more.

In the Robert Black case—a serial child killer—I had to think: ‘what has he done to these children?’ And I can remember myself just thinking: no. A bit of me was saying: ‘You really don’t want to think about this.’ But it would tend to be cops who get the real brunt of this because they deal with the victims—and the offenders, you know, people who will lie to you, and boast or brag, when you believe they’re guilty. So that’s a much trickier psychological environment than just reading the story in a case file.

And, I suppose, these extreme cases like Robert Napper or Robert Black must come up relatively rarely.

Oh, yes. I mean, I’m not sure how many serial killers I’ve dealt with, but you could count them on one hand, probably. They’re pretty rare. Whereas if you pick up a crime fiction book, every second book has a serial killer. There soon won’t be any people left!

Great, so we’ve made it to book number five, and we’re retaking our thread of the history of forensic science. This is Suspect Identities: A History of Fingerprinting and Criminal Identification, by Simon Cole.

Yes. This, for me, completes the circle—although things have moved on a bit since 2001, when this was published. It’s a more academic book—I mean, it’s very readable, it’s not an academic treatise, but I guess unless you’re really interested then you might think it a bit dry. But he’s a very good writer and he has good sources. What he asks is: how did we get to this idea that we can identify people? Where did it come from, how do we do it, and does it work? That’s the short of it.

We need to identify people because if you’re going to put somebody in prison, or even accuse them of a crime, we need to be confident it is the right person. So it’s a really important aspect of criminal justice. Most people don’t realise that, actually, most fingerprint identification is simply used to confirm the identify an offender when they are arrested.

“The idea caught everyone’s imagination: that you can measure the body, and that can be used to identify a person”

What Cole does, is he goes right back to Bertillon. Bertillon was the French anthropometrist who believed you could identify a person by measuring the body; he took photographs and he would measure noses, eye distances, all that—early ‘facial recognition’ stuff. But this was completely impractical. You couldn’t store this stuff and identify people quickly enough. So it was quite quickly supplanted by fingerprints, because they were left at crime scenes. Not many criminals leave photographs at crime scenes, or at least not before CCTV!

But the idea caught everyone’s imagination: that you can measure the body, and that can be used to identify a person. So Cole goes through Bertillon, and some other quite scary people like Lombardo, who would say that ‘the shape of the skull tells you that this person is a criminal’—all this eugenics stuff.

Along come fingerprints, and there are eugenicists involved in that as well. Francis Galton, Darwin’s cousin, was a huge eugenicist, he believed in a criminal ‘type’.

Right. Galton made a composite images of ‘the face of crime’ by combining the faces of lots of different criminals. He thought that if you looked like the composite face, you were more likely to become a criminal.

If you go back to the time of The Killer of Little Shepherds, or a little earlier, this was a time when you got these huge movements of population, immigrants moving from the countryside to the cities. There was more crime, and they wanted to identify criminals. Everything was getting organised, systematised—all these policing organisations were being developed.

So Cole goes over the history of all this, and dissects the connections between this and the emerging belief in uniqueness—that your face is unique, for example. But even if your face is unique, in modern facial recognition the measurement of your face requires a machine, and a machine is created by a human, and inside the machine is an algorithm written by a human, and its output is interpreted by a human. None of these things are perfect.

Yes, we’re beginning to hear a lot of these stories of bias in algorithms used in artificial intelligence and machine learning systems too.

They’re all subject to error. But uniqueness was just such an enormously convenient idea. So what happened—in the 1930s, in Scotland, and there was a kind of parallel evolution elsewhere—it was asserted that fingerprints were unique, and are therefore infallible.

Actually, that doesn’t logically follow. There was a legal judgement in the thirties in Scotland where the judge said, ‘well, you know, I don’t like you saying these things are infallible, because nothing is infallible.’ He should just have stopped there, but he went on to say, ‘but I guess they’re practically infallible…’ No! Science doesn’t deal with uniqueness. That’s not a scientific question. To establish uniqueness in any real way, you would have to measure everything.

The second thing is, you don’t need it. I mean, I’ve gathered evidence, I’ve given really important evidence in many criminal trials which was nothing like unique. It was highly arguable. But it was on point, it went to the heart of the case: did this happen, or did that happen? Was this person there? Or was someone else there? The courts don’t need uniqueness, they never have. But it’s enormously convenient.

“The Shirley McKie case opened up this whole argument over the uniqueness of fingerprints”

Cole exposes all this really quite well. There’s a little vignette in my book, where I describe meeting a fingerprint expert at the Old Bailey, and he spends all his time telling me how unimportant my evidence is, and how important his evidence is—because it’s conclusive. Which is just another word for ‘unique’.

I was quite young and thought, well, surely they wouldn’t call me to court if my evidence was of no relevance. The courts are quite busy; they only call witnesses that have something to say. When I went back to the lab, I told my colleagues and they all just burst out laughing. They said, ‘He’s one of the mad fingerprint people who think they’ve got the answer to everything.’ It’s just dogma, it’s just faith. No science in it at all.

Why is this significant? Well, have you heard of Shirley McKie? Or Brandon Mayfield. These are the two big cases. Shirley McKie was a Scottish police officer who was accused of leaving a fingerprint at a scene. She didn’t, she was never at the scene. It was a mistake by the fingerprint experts. The debate was a huge problem, went on for more than a decade, and eventually the Scottish Government paid out more than £750,000 to Shirley McKie. That opened up this whole argument over the uniqueness of fingerprints, and error, accuracy, faith and dogma and exposed the whole thing.

The Brandon Mayfield case had some very interesting background factors—he was a US citizen and a Muslim, which I’m convinced had a bearing. The Madrid Bombing of 2006 was a horrific bombing that killed well over 100 people. The Spanish government’s criminal justice authority sent out all these fingerprints that they found associated with the crime. One went to the FBI, and they identified this American Muslim lawyer called Brandon Mayfield.

“Nine times out of ten, eight times out of ten, the forensic evidence is useful but not critical”

The Spanish experts eventually say, ‘we think the fingerprint is from another guy, who happens to be an Islamic terrorist.’ But the FBI says, ‘no, no, no. You’ve got it all wrong. It’s this guy—Brandon Mayfield.’ Eventually they have to admit they’ve made a complete error, they’re absolutely wrong. There goes the infallibility of fingerprints. They have to pay out $2 million for making an unfounded accusation. There’s a huge fuss about it, all around the world—as Paul Simon had called it, many years ago—the myth of fingerprints. The bubble had burst.

Since then, the whole business of fingerprints has been cooling. There’s been an enormous transformation from this dogmatic belief, to what it should be: fingerprints are a perfectly good way of identifying people, but you need to be honest. Sometimes errors are made, and you need to pay attention to certain kinds of cases which are problematic.

Almost all the misidentified fingerprint cases involve difficult fingerprints to deal with. They’re slightly distorted, things like that. They require a very careful procedure to make sure you’re not biased. What happened in the Mayfield and McKie cases was that they made mistakes and then they dug in. So that’s the link back.

Yes, this does seem to be an overarching theme. So, as we proceed with DNA evidence and ever more cutting edge technology, is this history being presented as a way of telling us to proceed with caution?

There’s an element of that. The problem is that forensic science is seen as this monolithic body of knowledge, and it isn’t. What is?

So the first question you should ask is, what kind of forensic science is involved? If you say DNA, most DNA profiling is pretty safe stuff. Not perfect, but pretty safe. This is amongst the best evidence that you will get in a court case. But that’s only a certain type of DNA. All the new science that’s coming out, you have to treat it with caution, as you say. You have to be careful.

DNA mixtures, which are very common in sexual offences, are much more complicated and require much more caution, like the distorted fingerprints. So you should never just believe this stuff, nor should you automatically disbelieve it. You should ask fairly seriously questions about the science—and here’s the thing—and the person (the expert witness) who’s telling you the story. Because some people have credible credentials, and other people haven’t. Even the people with credible credentials sometimes make mistakes.

So you have to unpack it: Why do you believe this? Why do you believe that? That’s what should happen in a good court case. Although things only seem to go wrong in forensic science when the evidence is really important. Then, because then everybody digs in, there is a really deep argument. But nine times out of ten, eight times out of ten, the forensic evidence is useful but not critical. You know: you get a DNA match from a crime scene, it leads you to an offender, then you establish that there are a couple of witnesses that link the offender to the crime scene and he drives the same kind of car. You’ve got a partial numberplate. Then, the DNA evidence is what I would call ‘investigative evidence’: it takes you to someone. It’s not the case against them, it’s only part of the story.

So: caution, yes. Honest scepticism is what you need in all these instances, and that’s something that can be missing quite a lot of the time.