Support Independent Journalism. Donate today!

Justice in America Episode 27: Junk Forensic Science

Josie Duffy Rice and guest co-host Zak Cheney Rice talk with Radley Balko, opinion journalist at the Washington Post and author of The Cadaver and the Country Dentist, about faulty forensic science.

Everyone who has ever watched a crime procedural believes that forensic science is the most reliable way to tell whether someone is guilty or not. But is that true? The reality is that a lot of forensic science is not exactly science at all. On this episode of Justice in America, Josie Duffy Rice and her guest co-host, Zak Cheney Rice, look at faulty forensic science. Radley Balko, opinion journalist at the Washington Post and author of The Cadaver and the Country Dentist, joins.

Radley Balko’s Book Recommendation, Ordinary Injustice by Amy Bach.

Additional Resources copy and links:

The Cadaver King and the Country Dentist, Radley Balko and Tucker Carrington

Faulty Forensics Explained, from The Appeal

We need to fix forensics. But how? by Radley Balko for the Washington Post

The Pseudoscience Behind Forensic Science, The Appeal Podcast

The 2016 PCAST report on Forensic Science in Criminal Courts

Pam Colloff’s Blood Will Tell, New York Times and Pro Publica.

How Dubious Science Helped Put a New Jersey Woman in Prison for Killing a Baby in Her Care, The Appeal

Justice in America is available on Apple Podcasts, Soundcloud, Sticher, GooglePlay Music, Spotify, and LibSyn RSS. You can also check us out on Facebook and Twitter.

Our email is



[Begin Clip]

Radley Balko: When a crime lab analyst’s year end review is done by a prosecutor, that’s a huge problem. There was a study that came out a few years ago that found I think, in fourteen, fifteen states and then dozens of cities, crime labs are only paid when there’s a conviction. If you’re acquitted, you don’t have to pay any court fines. If you’re convicted, you have to pay for the analysis the crime lab did. Well, think about what incentive that sends to people working in the crime lab. If I acquit this person, there’s no money that’s going to come into our office for this case. If I convict them, that’s $600, $700, that adds up after a while particularly if you feel like your lab is underfunded.

[End Clip]

Josie Duffy Rice: Hi, I’m Josie Duffy Rice.

Zak CheneyRice: And I’m Zak Cheney-Rice.

Josie: And this is Justice in America. Each show we discuss a topic in the American criminal justice system and we try to explain what it is and how it works. Thank you so much everyone for joining us today.

Zak: You could find us on Twitter @Justice_Podcast, we’re also on Facebook at Justice in America, and subscribe and rate us on Apple Podcasts. We’d love to hear from you.

Josie: So we opened the show with a clip from our guest Radley Balko. Radley is a journalist and an author, he writes about criminal justice for The Washington Post and his latest book, The Cadaver King and the Country Dentist is a really incredible story that is completely outrageous that deals with our topic for today, which is-

Zak: Junk science and the failure of forensics. And if you’re at all new to this topic, I promise you’re in for some surprises.

Josie: So today, we’re not going to have a word of the day because this whole episode is about an entire field of criminal justice that is almost entirely misunderstood and that field is forensic science. If you’re like many people in America, including us in our pre-kid days when we had free Saturdays and we watched SVU, CSI, NCIS or any other acronym show about the criminal justice system. And if you’ve watched those you’ve heard stuff like this before:

[Begin Clip]

Woman #1: Fingerprints, blood splatters and DNA analysis.

Man #1: The blood source being in this area when blows were struck.

Woman #2: The shape and distribution of these tiny drops reveal an astonishing amount of information.

Man #2: Violent death, a bloody murder, bloodstains on the wall are far from a random spattering.

Woman #3: I was focusing on the bloody footprints.

Man #3: You will dissect the evidence using the latest methods of forensic science. Every contact leaves a trace.

[End Clip]

Zak: Like a lot of people you might think that forensic science, by which I mean not just DNA but blood splatter, fingerprints, footprints, bite mark evidence, hair follicle analysis and other types of scientific evidence is exactly that: science. In pop culture and in the media these processes are described as very precise and law enforcement — no surprise — tends to describe it that way too.

Josie: But today we’re here to tell you just how faulty and fallible forensic science is, and how much of forensic science is at best unreliable, and at worst, just complete junk science. And we’re going to cover a lot of stuff — breathalyzers, Shaken Baby Syndrome, fingerprint analysis — but like always, we’re just quickly going to start with the basics.

Zak: According to the National Institute of Justice, there are twelve categories of forensic science, including toxicology, firearms, crime scene evidence, fire and arson evidence, and more. Some of these broad categories include practices that use quote-unquote “pattern matching” methods. That’s when the analysis tries to match evidence from a particular person or thing to another person or thing. This is the category that’s particularly problematic. The most common type of pattern matching is DNA analysis, of course, which usually attempts to match DNA from the crime scene to a suspect, but there are many others. 

Josie: For example, there’s fingerprint analysis, which matches fingerprints at the scene to the prints of a suspect. There is microscopic hair analysis, which attempts to match hairs found in one place with the hair of a particular person. There is firearm analysis, which looks at whether the bullets or shell casings at a crime scene come from an individual gun. And there is bite mark analysis, which tries to match teeth impressions from a suspect to bite marks found on a body. And there are many more types of pattern matching evidence, shoe prints, tire tracks, handwriting, so on and so forth. Then there’s a kind of quote-unquote “science” that analyzes a pattern to determine exactly what happened. So for example, blood stain analysis or arson evidence.

Zak: These types of forensic analyses have been used in countless cases, and resulted in the convictions of countless people. And it makes sense, right? Let’s say you’re on the jury in a case, and you hear a few different kinds of evidence. For example, you hear that the defendant told his friend he was going to kill the victim. But the defendant denies saying that, and the defense argues that that’s hearsay evidence and shouldn’t be trusted.  Then you hear that a random woman saw the defendant in the area on the day in question. Okay, you may think to yourself, that’s pretty damning. But the defense says the woman was too far away to see his face, or they claim she’s identified the wrong person and you think, it’s true, she was kind of far away, and yeah, maybe, she could have gotten the guy wrong. Then, the prosecution puts an expert witness on the stand and he says he’s an expert at microscopic hair analysis specifically and that he’s 100 percent certain that a piece of hair on the body of the victim matches the defendant’s. Well, that should probably do it. It’s a match. Of course this is the guy. Right?

Josie: Again, popular culture makes us really convinced that pattern evidence always demonstrates a really clear, a really simple and a really straightforward match. Here’s a clip from Last Week Tonight with John Oliver and it’s of examples from TV crime dramas talking about pattern matching evidence.

[Begin Clip]

John Oliver: Specifically, this story is about how we increasingly solve crimes using forensic evidence. It’s that thing that is just a staple of TV crime shows. 

Woman #1: Bullet from the torso on the left, bullet from the boat on the right.

Man #1: Two hearts beat as one. 

Man #2: Matches perfectly.

Woman #2: That’s a match. 

Man #3:  We got a match.

Man #4:  And they just found us a match.

Woman #3: Visible match.

Woman #4: We have a match 

Woman #5: Match.

Man #5: Were you able to determine which monkey bit him?

Woman #6:  The bite marks match those of the monkey found at the scene.

[End Clip]

Josie: But, not all of this evidence is actually ironclad. Take, for example, microscopic hair comparison — like the hypothetical that Zak just laid out — where examiners use microscopes to compare two different hair samples. The FBI’s microscopic hair comparison unit, that was a real thing, found, quote, “hair matches” in 2,500 cases over more than two decades. But a few years ago, federal authorities began to look into the practice, after The Washington Post reported that there was some real skepticism about the science. After the FBI began reviewing these 2,500 cases, they publicly acknowledged that that unit had given flawed testimony in 95 percent of cases. 95 percent. Of the first 268 cases that the FBI reviewed, 32 of the defendants had been sentenced to death and 14 of those 32 defendants had already been executed, in part because of the testimony that the FBI’s unit had provided.

Zak: Or take fire pattern analysis, where fire investigators look at burn patterns to determine whether a fire was arson or not. Dozens of people have been exonerated in cases where analysts testified that they were certain that a fire was started due to arson. Eventually, burn pattern evidence was debunked, but not before people were literally executed because of it. 

Josie: Yeah. Or take, for example, bite mark evidence. Hundreds – literally hundreds of people — have been convicted on the basis of bite mark evidence. One guy, Michael West, was a bite mark expert in Mississippi who testified in dozens of cases all around the south helping convict people based on his determination that their teeth marks matched the marks on a body. Our guest, Radley Balko’s most recent book talks about Michael West at length and we’ll talk about him more with Radley later. But quickly, here’s a clip of Michael West from 60 Minutes talking about bite mark evidence. 

[Begin Clip]

Steve Kroft: How does it compare with other types of forensic evidence? Fingerprinting, DNA?

Michael West: If you look at bite marks, you don’t walk into a woman’s house and borrow a phone and accidentally bite her on the arm. You have to be in a violent contact with that individual to deposit a bite mark. In that light, my marks are better than fingerprints because they not only show that you were present, they show you were in a violent confrontation with that individual. The science of bite marks analysis is very accurate. The art and science of bite mark analysis depends on the clinician, the person who’s rendering the opinion. 

Steve Kroft: He’s found bite marks on a decomposed body submerged in a swamp, on a corpse that had been buried for more than a year, he’s even used a bite mark taken out of a bologna sandwich to get a conviction.

[End Clip]

Zak: Now, finding a bite mark on a corpse in a swamp? If you heard this from some random guy, you probably wouldn’t want to hang out with that guy anymore, but you might think he was lying. You’d definitely think he was lying but a dentist, in a courtroom, labeled an expert? You really might start to believe that there was something preserved on a corpse that had been in a swamp. We’re going to talk more about West with Radley, but here’s a quick story about how outlandish his claims could be — almost laughable honestly, if there weren’t lives at stake. In 1992, in Louisiana, Louise Keko was killed. The primary suspect was her estranged husband, Tony, but there was no physical evidence tying him to the scene.  So for a year, the case stayed unsolved, then, the sheriff called in Michael West to do a bite mark analysis, even though no bite marks were found on Louise during the autopsy. And Michael West said that, under a special UV light, using a technique he alone knew how to use, he might find something that hadn’t been seen previously. So they exhumed Louises’s body fourteen months after she was killed. Here’s a clip from 60 Minutes that begins with him talking about the supposed bite mark. 

[Begin Clip]

Michael West: In the visible light, it just appeared as a big black bruise. Under alternative light, the pattern of a bite mark appeared. We found that there was a curve pattern that matched four of Mr. Keko’s teeth to a perfect match on her shoulder. 

Steve Kroft: The only problem was Dr. West was unable to photograph or document what he’d seen, and the bite marks, he said, faded away when the tissue sample was stored in formaldehyde.

[End Clip]

Josie: If this sounds ridiculous to you, you’re not alone. Tony Keko’s lawyer also thought it sounded ridiculous and he argued as much to the jury. Here he is also on 60 Minutes.

[Begin Clip]

Eddie Castaing: This was the case of the disappearing bite mark. It wasn’t there the day after death in the autopsy photographs, and it wasn’t there on the day of exhumation until he put the light on then it appeared, then it disappeared ten days later. That wasn’t a bite mark. This was the case of the Emperor’s New Clothes. 

Steve Kroft: All the same arguments you’re making to me now, you made in the courtroom.

Eddie Castaing: Absolutely. 

Steve Kroft: And the jury still didn’t believe you. 

Eddie Castaing: That’s correct. 

Steve Kroft: Persuasive guy. 

Eddie Castaing: Very persuasive. A dangerous, dangerous witness.

[End Clip]

Josie: So Tony Keko was convicted of first degree murder, in large part because of Michael West’s testimony which remained the only physical evidence in the case. But then in 1998, Tony Keko’s conviction was overturned because the appeals court found that the bite mark evidence was questionable.

Zak: As it turns out, the evidence shows that bite mark evidence is simply not science. It’s just not. And West was one of the primary con artists out there pretending there was some sort of expert objective analysis happening. But there wasn’t.

Josie: The President’s Council of Advisors on Science and Technology, otherwise known as PCAST, is an advisory group of scientists and engineers across the nation who advise the president on science and tech issues. Back in 2016, when Obama was still president, they issued a report looking at some pattern matching evidence and how reliable it is. What they found about bite mark evidence was that it doesn’t even get close to meeting scientific standards. According to the report, a lot of times, analysts can’t even reasonably conclude that a mark on someone’s skin is a bite mark at all, much less match the bite mark to a person with reasonable certainty. Over the years, dozens of people, like Tony Keko, have been exonerated that were convicted based on this bite mark evidence.

Zak: And the report didn’t just implicate bite mark evidence, it found that, in general, pattern matching evidence is not nearly as reliable as people perceive it to be. Even stuff like DNA analysis or fingerprints is often much more complicated than one might assume. For example, when it comes to fingerprints, the report concludes that the science around matching is foundationally valid, but that it still has a “substantial” false positive rate, ranging from 1 in 306 to 1 in 18. 

Josie: Yeah and a 1 in 18 false positive rate is not insignificant. 

Zak: No. Not the best. 

Josie: Yeah. And remember the story of Brandon Mayfield?

Zak: Yes, Brandon Mayfield, guy who, Spanish authorities pulled a fingerprint I believe off of a bag, that was uncovered near the scene of the Madrid terrorist bombing in 2004 and they sent this fingerprint to law enforcement around the world, including the FBI. The FBI said we found a match, this guy Brandon Mayfield, he lives in Oregon.

Josie: Right. And so this guy had never even been to Spain and all of a sudden he finds himself accused of having something to do with a major terrorist attack that happened halfway across the world. And it was another example pattern matching evidence failing us, right? 

Zak: My understanding also is that after the FBI says we have a match, Spain writes in a letter that was dated in April that ‘we don’t actually think this is the guy.’

Josie: Right. 

Zak: Almost a month later, the FBI arrests Brandon Mayfield anyway.

Josie: Right. I mean, it was just an example of like, all the evidence is there that this is not a man responsible for any terrorist attack, right? But because these two prints match, there is basically a willingness to ignore all the other evidence that demonstrated that this was not the guy. 

Zak: Right. 

Josie: So, how does this happen? There’s actually a great explainer about this issue on — just shouting that out.

Zak: You may have heard of it.

Josie: You may have heard of it — That outlines this issue of faulty forensics really, really well. And it highlights a few things wrong with pattern matching evidence. 

Zak: First, the experts in these cases draw their conclusions from faulty assumptions. For example, this idea that handwriting, and bullet markings and dental impressions are unique. This baseline belief that every single person’s shoe prints or hair strands are visibly unlike anybody elses is a fundamental principle of pattern matching. It’s also based on nothing. There is no credible evidence to suggest that these categories are unique to each individual person or thing, especially unique enough to be discernible by human review. 

Josie: And even if there were a clear way to determine the uniqueness of the evidence in question, it’s not easy to know which quality of the thing you should be focused on. The analysts who look at this stuff don’t really know for certain whether the things they’re looking at to determine whether there’s a match are the relevant unique characteristics, for example. They’re only comparing two things, but there is no way of being able to tell whether they’re a million, trillion more matches of those exact two things out there. Right?

Zak: Right. For example, if I look at microscopic hair samples, and I’m determining a match depending on the color of the hair sample — I don’t know how common it is for two people to have the exact same color hair. What’s more, the two people could actually have very similar but slightly different hair colors that look the same to the naked eye. I wouldn’t know. There are very few guidelines in these fields regarding the scientific methods that should — in a perfect world, or even a slightly better world — guide rigorous analysis. There are also no objective standards. It’s all subjective and up to the interpreter. This even applies in some areas that we think of is concretely, super reliably, scientific like DNA cases. There are essentially three broad categories of DNA analysis. There is single source, which involves a sample from a single individual. There is simple-mixture, which involves DNA from two individuals. And there is complex-mixture, which includes samples from multiple people, often in unknown proportions. 

Josie: Most DNA analysis is either single source or simple-mixture, which generally means the analysis is pretty objective and the interpretation, as the PCAST report stated, involves, quote, “little or no human judgment.” But the same can’t be said for complex-mixture samples. These samples take significant interpretation and the PCAST report talks about this a little bit, about how complicated complex samples are. Basically, if a DNA analyst is looking at a complex-mixture sample, they basically have to determine whether the suspect’s DNA profile could be present within the mixture, what the probability of that is, there could be many different DNA profiles that could fit within that same mixture. In other words, the chance for error when you’re analyzing a complex-mixture profile is significantly higher, maybe millions of times higher than it is when you’re analyzing a single source profile or even a simple-mixture of profile. 

Zak: So. Here’s a story that highlights the terrible impact this evidence can have. It’s a story about a guy named Joe Bryan. There’s a chance that you’ve heard of Joe Bryan. There was a big story on him in The New York Times in 2018, written by Pam Colloff, who we featured on Justice in America last season. She’s one of the best reporters in the nation on the issue of faulty forensics.

Josie: So back in the mid-80s, Joe Bryan lived in a small town in Texas called Clifton with his wife, Mickey. We’re talking about a very small town, by the way, one traffic light literally. That kind of small. Anyway, Mickey was a teacher at the local elementary school, and Joe was the high school principal and was just the two of them, no kids. Then, one day in October of 1985, Mickey didn’t show up to work. She was later found dead in her bedroom. 

Zak: Joe was more than 100 miles away at the time, at a conference in Austin. Everyone had seen him the evening before his wife was murdered, and he was there, still at the conference, first thing the next day. But eventually—Pam’s article does a great job of laying out exactly why—Joe became the primary suspect. He was arrested and put on trial. 

Josie: Now, when you think about it, Joe being the killer was just sort of hard to believe, given the fact that he had a pretty airtight alibi that put him a few hours away from his house during the time of the murder. To quote Pam, “Prosecutors asked the jury to believe that between 9:15 p.m. and the following morning, Joe slipped out of his hotel in Austin; drove 120 miles to Clifton, at night, through heavy rain, even though he had an eye condition that made night driving difficult; shot his wife, with whom he had no history of conflict; drove 120 miles back to Austin; re-entered the hotel; and stole upstairs to his room — all in time to clean up and attend the conference’s morning session, and all without leaving behind a single eyewitness.” And it’s just not really an easy story to believe, right? I mean, what are the chances that this guy was able to kind of drive home and murder his wife and go back to the conference without anybody noticing?

Zak: Plus, there was plenty of evidence that didn’t implicate him – a cigarette butt that was found in the house, hairs that matched neither spouse, and more. But the prosecution presented an expert witness that seemed, to the jury, incontrovertible. He was a detective named Robert Thorman, and he went on stand and said he was an expert in blood stain analysis. 

Josie: You should really read the story to find out exactly how Thorman pulled it off, but he basically supports the prosecution’s theory that Joe killed his wife through this blood spatter analysis that he tells the jury. There was this flashlight with these teeny tiny dots of blood on it that was found in Joe’s car, and Thorman tells the jury that the little dots of blood were back spatter, a pattern that indicated a close range shooting. And to the jury Thurman’s an expert and this flashlight found in Joe’s car is pretty much just the smoking gun. And it’s really what destroys any hope for Joe Bryan. That’s how he ends up in prison for his wife’s murder, though he is almost certainly innocent. That was 35 years ago, and he’s still in prison today. 

Zak: This story really highlights so many of the issues with forensic science. First, the legitimate concern regarding what it means to be an expert in blood stain analysis. It used to be, once upon a time, way back in the early 1970s, experts in blood stain analysis were actual forensic scientists. They were researchers with backgrounds in science and scientific methods. Understanding blood stains well enough to analyze them required more than a simple class, it required high level math, and an understanding of fluid dynamics.

Josie: Whatever that is. 

Zak: The well known field of fluid dynamics. 

Josie: Yeah. To be clear, fluid dynamics is surely a real practice. In fact, it’s so real and so scientific that we have no idea what it is.

Zak: But then in 1973, one criminologist decided to start teaching these concepts to just regular old law enforcement officers where a science background of any sort isn’t required. So now to be a blood stain expert, you need no background in science, no background in forensics, nothing. Usually all you have to take is one week long course offered to anyone. In fact, Pam Colloff, the reporter who wrote the Joe Bryan story for The Times, took one of the classes in Oklahoma, she writes about in the article. The class cost $700 and most of the people in the class were law enforcement. And get this, the teacher told the students that they’d get their certificate of completion, even if they failed the final, which would have helped me a lot in college. It’s no surprise then that blood spatter analysis is also highly unreliable.

Josie: This is a major takeaway from Joe Bryan’s story. And it really underscores the need to rethink what it means to be an expert. There are very few things I could do for a single week and say with confidence that I am now an expert in this thing. I literally was trying to think of some earlier and I can’t think of anything that if I did it for a week, I would be confident going in front of a jury and saying, you know what, I’m 100 percent certain of this thing. So it’s kind of unbelievable what happens in court all the time. It’s unbelievable that it’s even legitimate. These expert witnesses often control the fate of the defendants’ entire life, because they’re seen as really scientific, and their evidence is seen as really impenetrable. But to present someone as a scientific expert, because they took a one week course, strikes me as one of the many major failures of the justice system. The other thing is that these experts also often use their experience as proof of their expertise. So in other words, they’ll say ‘I’m an expert. I’ve been testifying in these cases for 20 years. I’ve testified in hundreds of cases. I’ve analyzed hundreds of footprints or bullet markings’ or what have you, but if that analysis has been flawed for 20 years, then are you really a reliable expert, just because you’ve wrongly analyzed something 100 times? 

Zak: I don’t think so.

Josie: I don’t think so either.

Zak: The other thing at play here is confirmation bias. If the prosecution, including the expert, thinks they have their guy, they will interpret information in favor of that theory. They’ll see clear matches where they don’t exist. It’s not unusual to have two quote-unquote “experts” testify in one case, one for the prosecution and the other for the defense and draw completely different conclusions about what the blood stain analysis means. So did Thurman really believe so strongly that the minute splashes of blood were undoubtedly backscatter? Who knows? It’s possible he did. But believing something doesn’t make it true.

Josie: And there are no real standards to evaluate how exact a result is. So in other words, Michael West can’t say ‘I’m 92.4 percent certain that this is the guy,’ right? It’s all subjective. There’s no real measurement here. It’s like you trying to tell if I’ve been in the freezer eating ice cream for breakfast again, there’s not really a way for certain you can know that. 

Zak: Well, I would say with most people it wouldn’t be certain, but the ice cream lids you leave on the countertop-

Josie: Okay, okay. See this is not forensic science, you should not be able to convict me of murder over your analysis.

Zak: The point here is not just about subjectivity and a lack of standards, both of which are relevant here, it’s also about human error. When the DNA is complex or the fingerprint is slightly smudged or the sample is tiny or the evidence is accidentally contaminated or the testing equipment is wrongly calibrated, that all has a major impact on the reliability of the tests and the reliability of the conclusions.

Josie: So you may be wondering why the courts let in this evidence that has been pretty much clearly disproven. And the thing is that courts aren’t scientists. Courts rely on precedent. So they want to know if another court in the past has said that this evidence is okay. And as a result, they basically never stopped using bad evidence, because it’s always been let in before so it will be led in for the near future. A judge isn’t a scientist, they don’t know if the process is sufficiently rigorous or exactly how reliable it is. They just rely on precedent.

Zak: So we can’t cover everything today and there’s a lot more worth talking about but a few more things to note about forensic science. For one, do you know what is probably the most well known type of forensic evidence? Breathalyzers. The things that determine whether someone is too drunk to drive in most states if the breathalyzer reads over .08, it means you can be arrested and charged with a DUI.

Josie: Actually in our state of Georgia, you can be arrested even if you blow under a .08, so any alcohol reading can get you arrested. Here they call it a DUI Less Safe. But anyway, I digress. The point is that breathalyzers are used as reliable evidence of whether someone has committed a crime. I mean, drinking and driving is committing a crime, right? But just in November, a New York Times investigation showed that breathalyzers are and I quote, “often unreliable.” The Times said that they, quote, “generate skewed results with alarming frequency, even though they are marketed as precise to the third decimal place.” In Massachusetts and New Jersey, judges have thrown out more than 30,000 breathalyzer tests just in the past year. 

Zak: The problem here isn’t the machines, exactly. It’s, again, human error. It’s the failure to properly oversee the use and care of the machines. This is what happens when sensitive tools and practices are put in the hands of law enforcement, who, not only don’t have the knowledge of forensic scientists but also often don’t have the resources. The machines often haven’t been calibrated properly, which means they can show results that are often up to 40 times too high. And, according to The New York Times article, sometimes lab officials use stale chemicals to measure how much alcohol a person has consumed, which messes up results. In Massachusetts, apparently, one machine that police regularly used had rats nesting inside.

Josie: It’s like, how is that even possible? So as a result, these breathalyzer machines that we rely on constantly are regularly again, quote, “with alarming frequency,” just wrong, completely wrong. And to put this in context, about 1.5 million people a year are arrested for a DUI in America alone.

Zak: I just want to point out that many people in Massachusetts have apparently been blowing into machines that rats were nesting in so- 

Josie: Yeah, that’s also, right. What kind of diseases do they have now?

Zak: That’s another legal question.

Josie: Right.

Zak: Another area of forensic science that is much less reliable than, maybe, most people tend to realize is the science behind Shaken Baby Syndrome. You’ve probably heard those horror stories of babies who were shaken to death by a parent or a caregiver. For years, it was accepted science that if a baby had brain swelling, bleeding on the surface of the brain and bleeding behind the retinas, it could only be caused by shaking. But it turns out that a number of conditions can cause those three injuries to happen all at once. In 2015, The Washington Post found that at least 16 shaken baby murder convictions had been overturned.

Josie: One other thing to note about forensic science before we talk to Radley is about crime labs. Crime labs that test urine or drug samples or blood. These crime labs exist in counties all across the nation and very often they’re seen as places where evidence is processed objectively, and the same as everything else, there really is no room for error. A few years ago, there was a story in Massachusetts — also in Massachusetts, a lot of Massachusetts today — about a woman named Annie Dookhan and who worked in a crime lab in Massachusetts and Annie Dookhan and was probably the most effective crime lab technician in her office, right? She processed a ton of samples. She was really efficient. And it took years for them to really realize that she was actually just making up the results of many different test samples that she said she had done. It turned out that Annie Doohkan hadn’t processed a ton of the things that she said she had processed. A lot of people were sitting in prison or jail convicted of the crime of possessing drugs because Annie Dookhan said she had tested the sample and it had come out positive. But it turned out she hadn’t tested many of those samples, thousands of those samples. It was another example of forensic science gone wrong, where people think that what happens in a crime lab must be objective and scientific and reasonable but sometimes that’s not the case at all.

Zak: Here to talk more about this with us as Radley Balko, a columnist with The Washington Post who has been looking at issues around forensic science for years. His book, The Cadaver King and the Country Dentist, looks at the junk science career of Michael West, who we talked about earlier in the episode, and a medical examiner named Steven Hayne, who performed the vast majority of Mississippi’s autopsies. 


Josie: Thank you so much for joining us Radley, we’re here to talk about forensic science. And you in particular have done so much reporting and research on this issue and you also co wrote this great book that I love with Tucker Carrington called The Cadaver King and the Country Dentist. So I thought we could start by you just telling us a little bit about that book. 

Radley Balko: Sure. Man, where to begin? So, back in 2005 ish I’d written about a police raid in Mississippi, where the police raided both sides of the duplex but they only intended to raid one side. There was a guy on the other side named Cory Maye who woke up to police breaking into his home and thought he was being robbed or attacked and he shot and killed one of the police officers and then immediately surrendered with bullets still left in the gun. I think they found a roach in the apartment. You know, clearly a mistake on the part of the police. And Cory was convicted of capital murder, which is the intentional killing of a police officer and sentenced to death. And that’s kind of when I found his case. And there were a lot of things wrong with the case but one of the things that I found that was wrong with the case was this medical examiner who had testified that the trajectory that the bullet took to the officer was going down, which contradicted Cory’s story that he was lying on the floor, afraid and fired upward. And since, you know, his testimony was, his case kind of boiled down to whether the jury found him credible or not when he said he didn’t know they were police, it was pretty damning testimony from this doctor, Steven Hayne. And I talked to other medical examiners and figured out that it also was not scientific. There are a lot of variables that can affect a bullet’s trajectory, including if someone is sort of squatting, when they’re shot, which this police officer probably was because he was running up steps. And also it’s the position of the shooter and the angle of the gun and all these other variables. And so what he said was, it wasn’t technically wrong, but it was misleading and it led to Cory’s conviction. And even then I had been writing about this stuff enough to know that if you found an example like that you probably didn’t find the only one. So I started calling other defense attorneys in Mississippi and sure enough, they all had, you know, their own Hayne stories and from there, I started calling other medical examiners around the south and I wouldn’t even have to finish sort of my introduction, I would say, ‘Hi, I’m a journalist looking into a questionable medical examiner,’ and they would say, ’Oh, you’re talking about Dr. Hayne in Mississippi, right?’

Josie: Wow.

Radley Balko: And incredibly, nobody had really written about it and this is very early in my career and, you know, I sensed that I had kind of a huge story here. And so I wrote a big expose for Reason magazine, where I worked at the time and from then on Tucker, actually, when that came out, Tucker had just started as director of the Mississippi Innocence Project, it had just kind of gotten off the ground. I remember calling him and he tells the story about how he basically had a phone and a tablet and a pen at the time, didn’t have a computer yet in the office, and he had just been reading a couple Supreme Court cases involving Hayne and from then on, I’ve written thousands and thousands of thousands of words on Dr. Hayne and his sidekick, Michael West and their sort of forensics quackery, for lack of a better term. And a few years ago, it just occurred to us that Tucker would have a case or I would have a story and there’d be this frustration that this wasn’t limited to one case, there just wasn’t a way to kind of tell the whole story to give people the proper context and perspective in a single court filing on his part or a single article on my part, and and we needed a book, like a book had to tell the comprehensive history behind this, including going back decades before Hayne was even in Mississippi. And you know, we kind of decided we were the people that needed to tell it. Thousands of cases over about a 20 year period that these two were involved in and corrupted.

Zak: Yeah. So the book focuses a lot on bite mark evidence specifically, among other examples of kind of junk science. Can you talk a little bit more about bite mark evidence specifically, and how it’s been thought of science when it isn’t?

Radley Balko: So bite mark evidence, it’s a part of a larger field they call pattern matching fields forensics. And it really just boils down to an analyst gets a sample from the crime scene and a sample from the suspect and they just kind of eyeball it. And they say, ‘Yeah, that’s a match’ or ‘No, it isn’t a match.’ And the real tell when you know you’re in sort of your possible junk science territory when it comes to forensics is they can’t generate a margin for error. They can’t tell you how many times they get it wrong. That’s a problem. And then also any field where you regularly have two experts giving testimony at the same trial when they’re diametrically opposed to one another. Because, you know, you very rarely get two DNA experts that will vehemently disagree about how many markers are a match or nobody’s ever going to disagree about what a person’s blood type is, right? But it’s when you’re looking at something subjective. And the problem really with bite mark, even within the field of pattern matching forensics bite mark analysis is considered out there because the whole discipline rests on two fundamental theories. One is that we all have unique dentition, like we all leave unique bites. And two that human skin is capable of preserving that uniqueness in a way that it can be measured and compared and used to identify someone. And every test done by actual scientists on those two premises has shown zero evidence for them. There’s no evidence that we all leave unique bites and there’s really no evidence that human skin which is spongy and fungible and moves in different ways and starts healing almost immediately after you inflict a wound, but everybody heals differently, there’s just no way to say that a bite mark in human skin implicates one person to the exclusion of other people.

Josie: So one of the things that we were talking about earlier was about TV and movies and all of these sort of like NCIS and CSI and all these things that sort of portray these junk science methods as completely reliable. How do you think that this culture of — it’s not just law enforcement, right? — it’s just the general culture of forensic science and the cultural touchstones of forensic science, influences how believed it is in a courtroom?

Radley Balko: I think it has a huge impact. There’s a phenomenon that lawyers call the CSI Effect and that’s the impact these shows have had on what people expect of forensic testimony in the courtroom. And interestingly, the most common group that complains about it are prosecutors, because they argue that these shows have conditioned people to think that forensics is kind of foolproof, and it’s like you get a shoe from the crime scene and you put it in your forensics machine and it finds, you know, clay that was only native to this particular county, and therefore, it could only be this person and juries want that kind of certainty. Whereas scientists, legitimate conscientious scientists tend not to speak in certainties. Where you run into problems on the other side is when you get forensic charlatans like Hayne and West, who are willing to speak in those kinds of certainties, are willing to kind of eschew the science and give juries that certainty that they crave and they become very convincing. One of Michael West’s greatest hits was, there was a sting operation that a criminal defense attorney pulled on him, the attorney for a guy named Ray Krone in Arizona and in that case, he had been convicted based on bite marks and he was later exonerated by DNA and his attorney was so pissed off about this that he decided he was going to conduct a sting operation on a famous bite mark analyst. And so he went after West, and he sent West photos from the crime scene of the victim of a clear bite on the victim’s breast and then he sent him a dental mold from his own investigator that obviously had nothing to do with the crime. 

Josie: Right. 

Radley Balko: And he sent West the letter with both of these pieces of evidence and saying that basically he was the attorney for the family of the victim, ‘Here’s the crime scene photo, here’s the dental mold of who we think is a suspect, can you match these up?’ And West sent back a 40 minute video, I mean, I still have the video, where he not only says that these are a match, he says the odds of anyone other than the owner of this dental mold leaving this mark are astronomical. If you don’t know the premise and you watch the video, he sounds like he knows what he’s talking about. You know, he uses very technical terms, he’s got a big thick southern drawl and knowing him as I do from researching the book, he’s kind of an idiot but he sounds smart. He’s very good at sounding smart and sounding like he knows what he’s talking about to juries. And he had a huge impact. He put a lot of people in prison.

Josie: Right. It’s interesting, because even doing criminal law work at all, it was a long time before I realized you couldn’t trust this. I mean, it never occurred to me that you couldn’t trust bite mark testimony or something, you know, I’m not a scientist, maybe that’s real. When you actually explain it, it obviously sounds kind of ridiculous, but you can understand how in a jury room if a scientist says it, you assume they know what they’re talking about.

Radley Balko: Yeah. And that’s really the problem with this and a lot of other fields. It’s, you know, you don’t blame juries for finding a charismatic, persuasive person persuasive. The problem is that we’ve entrusted judges to be the gatekeepers of what is good and bad science of what juries actually hear and judges aren’t any good at that either. Judges are trained to be lawyers, they’re not trained to be scientists. And by making them gatekeepers, we’ve basically put them in this position of having to do scientific analysis. And, most of the time, they just look to what other courts have already done. And that’s how our legal system works. We want consistency, we want precedent to guide us going forward. That isn’t how science works. Right? Science is constantly kind of revising and altering the theory as we get new evidence. So the two systems really aren’t compatible. And the problem is the easy thing for judges to do is just to let it all in and think that our adversarial system is going to work it out and that is not what has happened. And every single scientific body that’s tried to analyze whether or not bite mark evidence has any sort of scientific foundation has found that it doesn’t and yet to this day, every single time it’s been challenged in court as unscientific the defendant has lost and the bite mark evidence was admitted.

Josie: Gosh.

Zak: Can you give us a sense of how often this kind of junk science is used in cases?

Radley Balko: So, you know, bite mark evidence, I would say we’re probably somewhere in the hundreds of cases-

Zak: Nationwide?

Radley Balko: Right, nationwide, you know, it’s been, I think the first case bite mark cases were in the mid seventies and it really kind of ramped up in the nineties, particularly because West was such a big evangelist for it. But, you know, beyond bite mark evidence for example, if you’ve looked at hair fiber analysis, which again, is highly subjective, it’s about sort of eyeballing two pieces of evidence and deciding whether or not they match. I mean, we’re talking tens of thousands of cases, there are thousands of cases at the federal level alone, but then all those federal analysts trained analysts state labs across the country. I mean, the really kind of terrifying thing you find as you start looking into these cases, is that you can almost pick any case at random, any murder case where some sort of forensic evidence was used, and you could look at what kind of evidence was used, you could look at who testified and you can start sort of peeling back the onion and you can just fall down these wormholes. I can tell you my wife wrote about a case — my wife’s also a journalist who covers a lot of similar issues — and she found a case where there was some very questionable blood spatter testimony. And this was a case I believe, like in the nineties maybe, and she looked at the police officer testified and his qualifications, and he had taken this, I think it was like a two day course in blood spatter analysis, similar to what Pam Colloff has exposed when it comes to this particular discipline, but then she looked up the woman who taught that class and this was, I’m going to get some of the details wrong I’m sure, but the gist of it was she was sort of like a paralegal who just decided she had this interest in blood spatter and so she started, you know, teaching herself and then just started teaching classes and certifying people. And she was wrong about a lot of it, and she taught thousands of people and this has happened to me too, but the remarkable thing is, this is, you know, one case that one journalist started looking at and you start going back and working your way backwards, and you find this sort of inverted triangle of other cases that were probably tainted by what you found in this particular case. And you start to wonder if anyone’s ever gotten a fair trial in this country at some point you know?

Josie: Right.

Zak: You mentioned that West was sort of a major evangelist for use of this kind of science. Have you found in the course of your work that a lot of these specific areas kind of generate from like one particular person who kind of becomes like the go to expert who then has just kind of massive influence and causes the stuff to be used in different courtrooms, in different cases across the country? 

Radley Balko: I don’t think there’s one person that it all stems from, I do think there tends to be, so for instance, bite mark evidence really kind of exploded after the Ted Bundy trial, because it was a key piece of evidence that was used to convict him and so now all of a sudden, everybody wants to be a bite mark expert. There being similar cases with like tool mark evidence. So this idea that like, you can look at the pry marks on a door handle or a door jamb and you can trace them to one screwdriver, you know, to the exclusion of all other screwdrivers that becomes popular after a couple of serial killer cases are allegedly solved from that. So I think it’s usually one kind of big high profile case that launches kind of a field of forensics into the kind of the mainstream and then everybody wants to do it. West’s specialty, he attended a seminar, we traced it back, we found he attended a seminar where another bite mark analyst was talking about ultraviolet light and how it might be possible to use ultraviolet lights or find wounds or marks on skin that can’t be seen with the naked eye. And West takes us to this radical extreme where he claims that he can, well, let me give you an example of just how radical this is. So there’s a case, the Larry Maxwell case that we write about in the book, and there was a triple murder in Louisiana and three elderly people were killed and police focused on this guy, Larry Maxwell. He actually had a pretty solid alibi. He was a chef on a naval base and he was on security cameras going in and then leaving at the time the murders allegedly occurred. Whatever reason the police wanted him for these murders, and so they bring in West —  all three people were stabbed — West finds a knife in the apartment of the victims that he claims just by eyeballing the stab wounds and looking at the knife, he claims that this knife and only this knife could have created these stab wounds, complete bullshit from the start, right? But all that does is establish the murder weapon. Now he’s got to tie it to the suspect. So he goes to the jail and he gets out his ultraviolet light and his yellow goggles and this is a procedure that he calls the West phenomenon and he claims that he can find marks on Larry Maxwell’s hand on his palm that could only have come from gripping this particular knife and stabbing someone with it. 

Josie: What?

Radley Balko: Now this was about a week after the crime had been committed. So you know, Larry Maxwell has gripped a lot of things in that week including sell bars right? And, you know, West, in other cases, claimed that he could match the marks on somebody’s hand to a particular purse strap, right? To the exclusion of all the purse straps. This is kind of the method that he perfected. And of course, West was never good at generating new suspects that the police didn’t already sort of suspect, right? He was very good at confirming the person they already suspected. That was what he did. But just the conclusion to the Larry Maxwell case, it gets even crazier when it got closer to trial, West claimed that the photos he took of Maxwell’s hands while he was doing his procedure that he had accidentally overexposed them. And so, not to worry, West took Maxwell to a copy machine and Xeroxed his hands and then wrote in the marks with a magic marker from his memory. And he said you’re just gonna have to trust him that he knew what he was doing here. Now that case was one of the few where there was too much even for the judge and he threw out the testimony, and Maxwell was later released, but he was in prison for two years, he was in jail for two years awaiting trial. And he later sued for, you know, violations of his civil rights and it was thrown out in federal court because somebody had written a letter on West’s behalf saying that what West had claimed in that case, was clearly wrong, however, it wasn’t so wrong that it was sort of outside the field of science. It was sort of a disagreement among experts sort of thing. Now the person who wrote that letter was Steven Hayne, who would become West’s sort of business partner and partner in crime through all of this.

Josie: Why? Why lie like this? What is it about going as far to say that you can tell what he gripped? Did he believe his own story? Do these people believe their own or what is it about this that makes him want to do that?

Radley Balko: That’s a great question. I mean, I’ve struggled with that for a long time, we struggled with it when writing the book that these people actually believe their own bullshit.

Josie: Right.

Radley Balko:  And, you know, I think a lot of it was probably misplaced trust that police hunches are usually right. And so I think there was a sense that they were helping put the bad guys away. You know, Hayne and West — Hayne more so than West— but Hayne made a lot of money. We tried to crunch the numbers and he was making at least $1.5 to $2 million a year at sort of the height of his career. But I think a lot of these, particularly a lot of these bite mark analysts, there’s kind of a guardian angel sort of complex that they take on. When you criticize what they do, they always bring up sort of the child killers that they helped put away. And so I think there is a sense of like, they’re on the side of righteousness and goodness, I can point to a lot of cases where they got the wrong person, you know, nobody’s sort of pro-child killing or child sexual assaulting. We want to make sure that we get the right person. And you know, that’s one of the central themes of our book is because they focused on the wrong guy in one of these two cases that animate the book, because they focus on the wrong person and because Hayne and West came in and confirmed their incorrect suspicions, the real killer remained free, and he had assaulted and murdered another little girl about a year and a half later in the same neighborhood. So yeah, I think they think that they’re doing the right thing. And I think there’s a lot of misplaced trust in kind of the hunches of law enforcement. 

Josie: Yeah.

Zak: So we’ve talked a lot about bite mark specifically, but there’s all kinds of evidence that’s error prone in criminal cases. Can you talk a little bit about eyewitness testimony? And kind of how easily it’s misunderstood by jurors and the average person?

Radley Balko: Right, yeah. So I mean, we’ve known for decades that eyewitness testimony is incredibly flawed, and the courts have been really slow to accept that. There are some famous and really compelling videos you can find online, there’s one pretty famous one where you’re asked to sort of watch this scene and your attention is focused on one thing, I think it’s like a sporting event or something, but halfway through the video, sort of a man in a bear costume sort of runs across the screen and most people don’t even notice it because your attention is all kind of directed in this other area. And it illustrates kind of just, you know, how bad we are at picking up details in the background while we’re looking at something else. Of course, nobody knows where a crime is going to be committed or when it’s going to be committed. We’re usually looking at something else when it happens. But it seems like really compelling evidence because somebody saw it, right? But you know, yeah, I mean, we’ve found the power of suggestion is incredibly powerful when it comes to altering eyewitness testimony. Lots of studies have shown one thing that is particularly compelling is when an eyewitness suggests someone to get a lineup and they suggest maybe it was that person, the police officer then tells them, they got the right person, their certainty goes through the roof, and they become adamantly certain it was that person, you know, going forward and there are studies showing, even when they get the wrong person, if you confirm that they got the right person, they become more certain in their opinion. So yeah, I mean, it’s incredibly, you know, the crazy thing, there are things that we can do to improve it dramatically. Things like the person administering the lineup should not know who the main suspect is. You want lineups that are sort of randomized. You want as much blindness and ignorance going in as possible. And there are lots of police departments that have adopted those procedures. But there are a hell of a lot that haven’t and most courts have not compelled them to, including the US Supreme Court, they had a chance to a few years ago, and they declined. Prosecutors, sometimes they’ll say these procedures would make it more difficult for them to arrest people. But if it makes us more certain that we got the right person, but that also is going to result in fewer arrests, it seems like that’s the bargain that we should probably take. So far, the courts have been reluctant to compel it.

Josie: One topic I have been kind of obsessed with recently as the elected coroner and you wrote a piece about this a few years ago called “abolish the coroner” in The Washington Post. So can you just talk briefly about why elected coroners are such a problem? 

Radley Balko: Right. So we get extensively into this history in the book and the coroner is such a fascinating position. It’s a government job that we’ve sort of inherited from England, and even in England, and we get into a lot in the book, it was the coroner, there’s no association between coroner and death, like the root of the coroner has nothing to do with death. Originally, the coroner was somebody who sort of collected taxes on behalf of the King and over the course of centuries of English common law it kind of adopted or morphed into this position where it kind of just took on odd jobs that nobody else in the King’s government was doing. So coroners had weird responsibilities like they were to, any sturgeon, or some species of valuable fish that washed ashore, or were caught by fishermen, were considered so valuable that they’re automatically the property of the King. It was the coroner’s job to make sure that nobody kept that and the fish if they caught one or found one. All these very strange kinds of positions. Anyway, it ends up being associated with death because one of the English Kings and post a death tax and part of the death tax involved if someone was murdered then the murderer and the village of the murderer had to pay extra tax and so it became the coroner’s responsibility to investigate who committed the murder so they would know which village to tax. And so we inherit this position that sort of over the years inEngland then becomes its primary responsibility becomes investigating causes of death but its origins are in this weird tax thing, but it gets inherited in the US and it’s one of the few government positions that a lot of state constitutions are specifically prescribed. While investigating deaths as a part of it, it has all these weird other things. So in Mississippi, for example, which is where the book kind of takes place, for a long time, the coroner’s job was to investigate any suspicious deaths, and also to round up any stray livestock that had gotten out and that someone had complained about. And so there are interviews of coroners in like the eighties and nineties they say you get a call in the middle of night and you don’t know if it’s because somebody has been murdered or it’s because you have to go round up some pigs and take them back to the rightful owner. But it’s an elected position and it requires, there’s no prerequisite that you know anything about investigating deaths. In fact, the people that usually run for the job tend to be the owner of the local funeral home, which is not about death investigation, it’s about knowing how to embalm the body after death, but they tend to take the job because you’re the first person on the scene after suspicious death, you get the first crack at the family for funeral and embalming services, right?

Josie: Oh my gosh.

Radley Balko: And the job becomes extremely corrupt, it becomes sort of an entry level into politics for people, it becomes a patronage position. And, you know, you can imagine like during the Industrial Age, as the Industrial Age comes on, you have people dying in factories, and now determining whether a death was a suicide or an accident or negligence, a lot of money is riding on what diagnosis you come up with. And so coroner’s getting a lot of kickbacks, they start taking bribes and then we enter kind of the civil rights era, or actually, the Jim Crow era, and coroners are kind of developed, particularly in southern states, but actually all over the country, as a way to kind of sweep lynchings under the table. The coroner becomes kind of a part of propping up white supremacy and propagating segregation by refusing to investigate lynchings. I mean, we found cases in the book where there’s this magical phrase that coroner’s would use every time there was a lynching, they would say “murdered at the hands of persons unknown,” right? Well, in some cases, the people who did the lynching were photographed posing with the dead body in the local newspaper, right? I mean, it wouldn’t have taken much investigation to actually figure out who did it but because the coroner had control of the crime scene and was the person who did the investigation and had ultimate say over how the person died, their conclusion ended the matter if they decided that we would never be able to ascertain who committed this crime then that closes the case. More often what you’d also see is somebody found hanging from a tree and the coroner would say they died by suicide or they died by natural causes. There’s a quote we use at the beginning of one of the chapters in the book where this woman, I can’t remember the exact quote, she said something to the effect like ‘My brother was shot 10 times and hanging from a tree and they said he died of a heart attack.’ And so that’s how you see the position used. And with states that still have that system, the coroner is still in charge of the crime scene. They’re the first person on the scene, it’s their job to secure the scene. And while we don’t, at least in most cases, sweep lynchings under the table anymore, they can have an enormous influence over say, what we found in Mississippi was a perfectly healthy 19 year old black kid dies in the back of a police car and the coroner would say, ‘Oh, it was a stroke or heart attack.’ And then there are also, obviously, all the other cases where the coroner has a strong incentive to tell the prosecutor what they want to hear in order to get convictions. And so you see the corruption kind of on both sides there. But at the core, it’s a position that evolved to serve people in power. One of the cases we find in the book is when US Attorney, in the federal government, wanted to come in and prosecute a case from the sixties, which was a civil rights murder. And you know, the guy who they were trying to convict was unquestionably a member of the Klan, he probably committed or almost certainly committed the two murders which they charged him for, and he probably committed a lot of other ones. They brought in Hayne there was basically an order for the feds to get jurisdiction, they would have had to show that when the bodies were thrown in the river, they were in another state. So you get the crossing the state border issue and then the feds have jurisdiction then, and Hayne came in and said that he could tell after they dug these bodies up after decades that the two people were alive when they were dumped in the river. And there’s just no way, right? Now in that case, there was some karmic justice, right? The guy deserved to be convicted of something, he had not been convicted earlier, but there’s still something sort of just uncomfortable about finding somebody willing to say something that he had no business saying in order to kind of get to that end. And I think it was a good illustration of how this position and the coroner was the one who sort of contacted Hayne and got him to say that, I think it’s a good illustration of how this position had evolved to kind of suit people in power, even though kind of their objectives may, what was the politically prudent thing to do has switched, at least in that particular case. There’s obviously still a lot of racism in Mississippi and this system still disproportionately obviously affects marginalized groups. 

Josie: Right. 

Radley Balko: But you know, yeah, it’s a system that was easily abused and easily manipulated to service people in power. 

Zak: I don’t think I’ve ever seen a coroner’s campaign video before. What does that look like? What kind of campaign materials does one present to the public as a coroner candidate?

Radley Balko: I mean they’re very, very localized election so you tend to see ads in local newspapers for coroners, but we found some and Hayne and West were very, you know, the coroner officially, but usually the prosecutor, the way it works in Mississippi is it was all contractual basis, the autopsies were done on a contractual basis. So there’s a suspicious death, the coroner gets the crime scene, decides it’s suspicious. Now they need an autopsy. And so they would contract the autopsy out to a doctor in private practice. Hayne did 1,500, 1,800 autopsies a year, somewhere between 80 to 90 percent of the autopsies in Mississippi over a 20 year period. But he only briefly held an actual public position. The rest of time he was a doctor in private practice. And so what would happen is they would contract the autopsy out to a doctor in private practice, the prosecutor would usually typically say, ‘Here’s what I think happened.’ And if that doctor comes back and says, that isn’t what happened, or more likely, there’s just not enough evidence for me to say that’s what happened, that makes the prosecutor’s job more difficult, and that doctor probably isn’t going to get the next referral and so the system selected for somebody like Hayne, if it wasn’t him, it would have been somebody else. They needed somebody who was going to tell the prosecutors what they wanted to hear. And that’s how he was sort of able to dominate that system. And Hayne became very powerful, became rich and so there became this kind of mutual back scratching going on where when we get referrals from coroners, in exchange for that he would endorse them when they ran for reelection. And he and West would regularly take out newspaper ads endorsing somebody for coroner over somebody else. Actually, West himself was a coroner for a little while, too.

Josie: So it seemed for a brief moment, it seemed like there might be some addressing this junk science thing when this PCAST report came out a couple years ago, and there was some energy, I mean, I guess it’s not the number one issue for most people, but there was some sort of energy towards maybe highlighting how unreliable a lot of this evidence was. And then both under Obama’s Justice Department and Trump’s there’s been some reticence to address this. That’s another thing you’ve written about. Why do you think that is and why do you think that people are so uninterested in kind of highlighting just how bad this evidence is?

Radley Balko: Yeah. So Obama, you know, he assembled the Forensic Science Commission. He also, like you said, the President’s Council of Advisors on Science, PCAST, yeah, they put up this report that was incredibly damning, and really specific about the problems in forensics. And you had this Forensic Science Commission that was made up of actual scientists, along with some lawyers, but the idea was that we were going to, you know, figure out where the problems are and work to correct them or fix them. But when the PCAST report came out, as you said, Obama’s own Attorney General Loretta Lynch, basically sort of dismissed it and said, ‘We’re not going to follow these recommendations.’ And you know, I think it’s about control. I think the Justice Department, the FBI does not want outside people telling it what forensics it can and can’t use. They want to be able to use whatever they need to solve crimes, to get convictions. So then the Trump administration takes over and Sessions immediately disbands the Forensic Science Commission, actually declined to renew their charter and instead they say, ‘Well, we’re going to handle forensics internally’ and they appoint this career prosecutor to oversee sort of the use of forensics in the Justice Department and a guy who, who actually was on the Forensic Science Commission but if you talk to other people who were on that commission, they say basically, his role was to object to everything they wanted to do. So this is the guy that Trump then puts in charge of forensics in the Justice Department. And the FBI is considered probably the kind of gold standard of crime labs, but it has had a lot of problems. There were the anthrax cases that got two wrong suspects, and ruined two lives over their investigation there. And then there’s the hair fiber analysis that I mentioned earlier, where you’ve got tens of thousands of cases that may have been corrupted. For a long time, the FBI was claiming that every box of bullets of ammunition had a unique chemical signature, and so if you found a bullet at the crime scene, and you took this signature and it matched a box of bullets at a suspect’s house that that bullet could have only come from that box. Turns out, it was complete crap. There was never any science behind that. So when Rod Rosenstein gets out in front of the prosecutors, as he’s given various versions of this talk, but he’s basically saying, ‘Just trust us, we’re handling this internally, you know, we’ve got our best people on this. We take science very seriously.’ We shouldn’t trust them. I mean, that’s an enormous amount of power, the ability to kind of use these different methods of analysis to put people in prison or even sentence people to death, that needs outside supervision, it needs to be treated with skepticism, and the FBI’s own record, I think, really hammers that point home.

Zak: So what is the answer here? What should we be doing differently to address these issues?

Radley Balko: Oh, man, you know, I don’t think there’s a single answer to that question. I did a whole online symposium at The Washington Post with fifteen or so people who’ve been advocating for forensics reform and there’s no, you know, they all think the system is broken, and there wasn’t agreement among them on how to fix it. So one of the main problems I mentioned before is that judges are the gatekeepers of good and bad science. And this is a huge problem because they’re trained to think like lawyers, not scientists, and they do a really bad job of that. But I don’t know what the solution is to that. And some countries have what they call a hot tubbing, where the state has an expert, the court has its own expert, the defense has an expert and the three experts get together and decide what they all agree on and that’s the only evidence that’s admissible. That’s interesting. There’s also the idea of just the court having its own experts that only report to the court so they don’t report to defense or prosecutors. If we’re going to keep the adversarial system, there should be equal funding on both sides. So defenders should have the same amount of money to hire an expert, whatever the prosecution spends, but even then, I don’t think the idea that letting bad science get in front of a jury, and that letting the defense put on somebody who says that this is bad science, that’s going to solve the crime. Because I think we don’t want juries hearing bad science. And a lot of times defense attorneys are in the awkward position of, you know, they can put on an expert who’s going to say, ‘This whole field is corrupt and shouldn’t even be presented to you’ or they can get their own expert in that corrupt field to say, ‘This person is wrong about their conclusions’ and usually they can’t do both. And that’s a problem. So, you know, if we’re going to use the adversarial system, I think both sides need to be equally funded. I would hope that we could try to move away from it. They’re also sort of assuming things aren’t even going to change that much. Crime labs should be independent. The idea that crime labs report to prosecutors, are sort of within the law enforcement sphere, it’s a huge problem. I always tell the story, there’s a great report by the Raleigh News & Observer a while back about corruption in the North Carolina state crime lab, and they found this security video, these two blood spatter experts who did this experiment with blood spatter with respect to a certain case and they did it seven, eight, nine, ten times until they finally got the result that the prosecutors wanted at which point they high fived one another. But the defense was never told that it took them eight or nine times to get this result. 

Zak: Or that they high fived each other.

Radley Balko: Or that they high fived each other. (Laughs.) But this is a problem though when crime lab analysts’ year end review was done by a prosecutor, that’s a huge problem. There was a study that came out a few years ago that found I think, in fourteen, fifteen states and then dozens of cities, crime labs are only paid when there’s a conviction. If you’re acquitted, you don’t have to pay any court fines. If you’re convicted, you have to pay for the analysis the crime lab did. Well, think about what incentive that sends to people working in the crime lab.

Josie: Right.

Radley Balko: If I acquit this person, there’s no money that’s gonna come into our office for this case, if I convict them, that’s $600, $700. That adds up, you know, after a while, particularly if you feel like your lab’s underfunded. One last recommendation is a guy named Roger Koppl, who has studied a lot of this and I think this is kind of a brilliant solution. It’s this idea of what he calls rivalrous redundancy. And so right now, the incentive I think for crime lab analysts is to give the state the results they want, right? They’re considered the part of the state’s team, you know, they’re considered on the prosecution side. And Koppl’s idea is that you hire what you call an evidence handler. And this is a neutral person. And what the person does is every case, you know, goes to the crime lab. But in every third or fourth case, he’ll take a piece of the evidence and send it to either a private crime lab or maybe a crime lab in another county or city and they’ll do their own analysis. And the official crime lab never knows when they’re being double checked, right? So they don’t know which case it is. So now the incentive is not so much to please the state or please prosecutors or please police, now the incentive is you don’t wanna be embarrassed by getting it wrong, and you never know when you’re going to be double checked. And the incentive for the rivaling lab is to find mistakes, right? You want to point out when they got something wrong. And, you know, he ran the numbers and the cost of what states pay for, I think it was like two wrongful convictions over 20 years, you could find a system like this. So you know, that would be a very easy way to kind of flip the incentives upside down and remove cognitive bias I think in a way that would really keep these analysts honest. 

Josie: Right. So thank you so much Radley, this was like really illuminating, and super helpful, and I’m really glad you came on. 

Zak: It was great thank you.

Radley Balko: My pleasure. Thanks for having me.


Josie: Thank you so much to our guest Radley Balko for joining us today. For show notes and for more resources about the issues we talked about on the show, please visit And don’t forget to check out Radley’s book bonus as well. You can find it wherever you listen to podcasts or on

Zak: And thanks for listening to Justice in America. I’m Zak Cheney-Rice.

Josie: I’m Josie Duffy Rice.

Zak: You can find us on Twitter @Justice_Podcast, like our Facebook page at Justice in America and subscribe and rate us on Apple Podcasts.

Josie: Justice in America is produced by Florence Barrau-Adams. The production assistant is Trendel Lightburn. Research assistance by Nawal Arjini. And Radley’s interview was recorded at Argot Studios, the engineer was Paul Ruest. Thank you so much everybody for listening and tune in next week.


Radley Balko’s Book Recommendation

Josie: I’m Josie Duffy Rice. And this is the Justice in America book bonus. I’m here with Radley Balko. He’s an opinion journalist at The Washington Post. Bradley, tell us what you’re reading right now.

Radley Balko: So actually, I’m working on a new book about criminal defense. So I’ve been reading a lot of court watching books. So the one I’m making my way through right now is written over 10 years ago I think, but it’s called Ordinary Injustice by Amy Bach. The really kind of incredible thing about it is she did some court watching. She looked at the state and the premise of the book, as you can probably glean from the title is that we get all these stories about big stories about wrongful convictions and police brutality. But it’s, it’s really the kind of day to day injustice, the minor cuts and scrapes that really wear people down and become sort of really burdensome on their ability to live their lives. You know, she wrote this book and it made a big splash, but nothing has changed. And it’s kind of depressing to read that and think that like we’re still here. Yeah, exposure didn’t do much. Well, maybe in a couple of years. counties that she went to I think things got a little bit marginally better, but it didn’t. You know, you hope kind of like all the podcasts going on right now like this one and the serial podcasts and that educating people about the outrages and the injustice that go on will foment change. And you know, we’ve seen a lot of that with progressive DBAs and so forth. But I think the kind of day to day machinery of the system, I feel like we still have a really, really long way to go.

Josie: Yeah. That’s a great answer.