In The Dark - S2 E3: The Gun

Episode Date: May 8, 2018

Investigators never found the gun used to kill four people at Tardy Furniture. Yet the gun, and the bullets matched to it, became a key piece of evidence against Curtis Flowers. In this episode, we examine the strange histories of the gun and the man who owned it.  Learn about your ad choices: dovetail.prx.org/ad-choices

Previously on In the Dark. had it down pat for me. So all I had to do was go there and they ain't mean a question to now. I was scared, but it was the police. So I'm finna go. I know I ain't did nothing wrong cause I don't never do nothing, get in no trouble. Everybody just disappeared. Rebacked off. Because it sounded like it's a threat, right? It was. It was. It was a threat. I just wish that I shouldn't have, I hate my name being, I don't like it. And I just want to live a normal life. I hate it happening. There's something I haven't told you yet.

Something else that happened in Winona, Mississippi, on July 16 16, 1996, the day of the murders at Tardy Furniture. That morning, around 10.30, maybe a half hour or so after the murders, a man named Doyle Simpson finished work. Doyle was 38. He was a janitor at a sewing factory. And that morning, after work, he went to pick up some lunches for some of his co-workers from a place called Fuzzy's Fried Chicken. When Doyle got to Fuzzy's, he told the people there

that when he'd gone out to his car at work just then, he'd realized that someone had gone into his unlocked car and stolen his gun from his glove compartment. It was a.380 semi-automatic pistol. On the morning of the murders, it didn't take long for Doyle's story about his gun being stolen to make its way to the investigators down at Tardy Furniture. By the time Doyle got back with the lunches for his co-workers, two investigators were there waiting for him. This story of Doyle's stolen gun would become one of the key pieces of evidence against Curtis Flowers.

It would be repeated over and over in six trials over 21 years. District Attorney Doug Evans, the man who prosecuted Curtis Flowers in all six trials, would tell jurors that on the morning of the murders, around 7 a.m., Curtis Flowers walked across town to the factory where Doyle Simpson was working, went into Doyle's car, and stole his gun, and that Curtis then used that gun to kill the four people at Tardy Furniture.

It was a simple story, and a clear one. And it helped lead the jurors to convict Curtis Flowers and sentence him to death, even though the investigators never found that gun. But the actual story of the gun, the full story, the one I pieced together over months of reporting, that story was anything but clear. This is season two of In the Dark, an investigative podcast from APM Reports. I'm Madeline Varon.

This season is about the case of Curtis Flowers, a black man from a small town in Mississippi who spent the past 21 years fighting for his life, and a white prosecutor who spent that same time trying just as hard to execute him. The case against Curtis Flowers came down to three main things. The route, the gun, the confessions. This is an episode about the gun. On the day of the murders at Tardy Furniture, July 16th, 1996, right away, law enforcement saw Doyle Simpson as a possible suspect and his gun as the possible murder weapon. Because when you go around a small town saying

your gun has been stolen on the same morning that four people were shot in the head in a mysterious killing, and it turns out it's the same kind of gun that was used in the murders, it would hardly be a surprise to learn that law enforcement is now considering you a suspect in that murder. That's what happened to Doyle Simpson. That day, investigators had Doyle come down to the police station for an interview. Doyle told the investigators that he was at work at the sewing factory at the time of the murders. We talked to a man who used to work with Doyle back then at the factory. His name is Kenny Johnson. He just would sweep up the floor or whatever

and take out the trash and things of that nature. And you'd see him in and out the building, you know, as he needs to. Kenny said he was at work with Doyle on the day of the murders, July 16, 1996. And that day, nothing was out of the norm. You know, he came and went as he normally does, you know, in and out the building to take the trash out or whatever, to the dumpster out back or whatever. So it would have been possible for Doyle to sneak out without being noticed.

And that meant that Doyle didn't have a strong alibi. And there were a few other things that got the attention of the investigators. First, a dusty beige car that looked like Doyle's had been seen downtown, parked about a half block from Tardy Furniture, around the time of the murders. And two black men had been seen outside that car. They looked like they might have been arguing. Second, Doyle's own half-sister, a woman named Essie Ruth Campbell,

told the cops that she'd seen Doyle drive past her work that morning around 9, at a time that Doyle claimed he was at work. Essie told us she was outside when she saw Doyle drive by. I went out there that long, you know. I just glimpsed the car, and I just noticed, I know his car because his car always stayed at my house all the time. And he always came visit me all the time. So I know this car.

Investigators also learned that Doyle had a connection to Tardy Furniture. He'd work there part-time, off and on over the years. It wasn't anything formal. Doyle would just fill in here and there when the store needed help with deliveries. Six days after the murders, law enforcement asked Doyle to meet with them again, this time so they could polygraph him. I just have the one-page summary of the results from this polygraph. The investigator asked Doyle whether he'd lied to the cops about the theft of his gun

and whether he had knowledge of the murders at Tardy Furniture. According to the investigator's report, Doyle showed deception on both questions. The investigator wrote, quote, It is my opinion that Simpson is not truthful about the theft of the gun and knows who committed the murders. Before we go any further, I want to tell you something about polygraphs, which is that they are not reliable, not at all.

They're so unreliable, in fact, that jurors aren't allowed to hear about them. Polygraph results are not admissible in court, and for good reason. Studies have found that polygraphs cannot tell you who's lying or who's telling the truth. All they can tell you is whether the person taking the test is anxious. All they can tell you is whether the person taking the test is anxious. And because a lot of people are anxious when they're questioned about a crime, there are a lot of innocent people who fail polygraphs. It happens all the time.

They thought that he did it. That's what they thought. That's what they didn't say he did it. They did? That's what they thought. That's what they was, well, they didn't say he did it, but I think on the first day, that's where they went with the investigation. I found a woman named Denise Kendall on her porch in Winona one afternoon last summer. Back at the time of the murders, Denise was dating Doyle. And Denise told me that after the murders, law enforcement had her come down to the station. They was questioning.

Did he ever get a lawyer at any point, do you know? I don't believe so. How did he seem with all this? Well, he didn't like the idea that they was trying to, I'm going to say, pin it on him, looking at him as a suspect, quite naturally. Investigators were particularly interested in Doyle's gun. They asked Doyle if he'd ever fired it. Doyle told them, yes, he had, he'd practiced firing it in his mother's backyard,

out on a country road called Poorhouse Road, on the outskirts of Winona. Investigators went to Doyle's mom's backyard with a metal detector to try to see if they could find any bullets that had been fired from Doyle's gun. They found a cedar post sticking out of the ground. It looked like it had been shot at. An investigator used a knife to dig a bullet out of the post. They went back again about two weeks later and found another bullet. The investigators sent those bullets to the state

crime lab. They wanted to find out if someone at the lab would be able to tell by looking at the bullets whether they came from the same gun as the bullets recovered at the crime scene. In other words, was Doyle's gun the murder weapon? Denise Kendall said all of this made Doyle even more nervous. He didn't like it. He didn't like it at all because they were out there picking out, picking the bullets, the shells or whatever goes in the tree when you shoot a gun. I don't know what's left in the tree, the bullet or something. So he didn't like that either.

Well, he knew they were trying to get him. I wouldn't have liked it either. If somebody come to my tree, a murder done happening, somebody come to my tree and start picking my bullets out of my tree that I'd been target practicing with a long time ago, and now you out here, ain't got no gun, picking bullets out of my tree. Doyle Simpson was clearly freaked out. But for Doyle, being freaked out wasn't anything new. Long before the Tardy Furniture murders,

Doyle was always looking over his shoulder, as if something terrible might happen to him at any moment. I talked to one of Doyle's half-brothers, a man named Johnny Earl Campbell. Doyle had a lot of shadows, like, scattered everywhere. I mean, you can't really tell about—I couldn't tell about Doyle. Doyle had something happen to him back in the past. I mean, Doyle got in trouble back in Louisiana years ago. But it was all down low, and nobody said nothing about it.

Johnny Earl Campbell told me that what had happened to Doyle years ago, down in Louisiana, had made him anxious, kind of jumpy, in a way that might have made him seem suspicious to the cops when they started questioning him in the Tardy Furniture case. And he told me that whatever had happened to Doyle down there, it had left Doyle with a deep, dark scar that stretched all the way across his neck. That's why he had a voice issue. Oh, he had a voice issue? Yes, he did. How did he talk?

It was like a wee sound when he pronounced words and everything, like a whisper. And that's what he would do. The story of how Doyle Simpson became a fearful man, how he got that scar that stretched from one side of his neck to the other, took place in a swamp on the outskirts of Edgard, Louisiana. That's after the break. Hi, this is David Remnick, and I'm pleased to share the news that I'm Not a Robot, That's after the break. it means to be human in an increasingly digital world. I encourage you to watch I'm Not a Robot, along with our full slate of documentary and narrative films, at newyorker.com slash video. Our reporter Parker Yesko went to the swamp outside Edgar, Louisiana, to see the place where Doyle became a fearful man.

A detective named Vernon Bailey brought her there. You don't want to walk in the grass in the air. Oh, no? No. What's there? All kinds of snakes in the grass. Snakes? In 1986, Doyle was 29.

He was living in Louisiana, spending a lot of time with a relative named Clyde. One afternoon, about two weeks before Christmas, in 1986, Doyle pulled up outside Clyde's house. That day of the incident, it was misty. It was like rain was supposed to be coming down. It was cloudy. Doyle was just sitting there in his car when all of a sudden, a man burst out of the house. He jumped into Doyle's car, put a gun to his head, and told him to drive.

Doyle didn't know it at the time, but this man had just slashed Clyde's throat and shot him in the head. Doyle and the man drove out, way out, into the swamps. And then the man told Doyle to pull over, and he marched him into the swamps. And then the man told Doyle to pull over, and he marched him into the swamp. The man took out a set of handcuffs and cuffed one of Doyle's wrists to a tree branch.

See that small tree? That skinny tree right there? Yes, straight ahead. Then he took out his gun and fired. He shot Doyle in the back twice. One of the bullets came out his neck. Then the man left, and Doyle was alone, bleeding to death in the swamp. If you was dull and you was handcuffed to a tree, what would you be trying to do?

Get free as quick as possible. You'd be for something. You'd smell your blood or whatever, you know what I mean? You'll smell your blood or whatever. You know what I mean? Alligator, snakes, possum, raccoon, wild hog. You name it. You don't want nothing to start jumping on you.

That's for sure. Doyle found a piece of broken glass, and he sawed through the tree branch to free himself. He crawled out to the highway and collapsed in front of a semi. The driver called an ambulance and got Doyle to the hospital. It didn't take long for the cops to find the person they thought did this to Doyle. It was a guy living down in Louisiana. And that guy took a plea deal, and he went to prison. The cops never figured out a motive for the crime. Their best guess is that it was a drug deal gone bad,

that maybe Doyle was involved in the deal, or maybe he was just in the wrong place at the wrong time. Doyle Simpson died six years ago. His family told me that after all this happened, Doyle was just uneasy all the time. He kept looking over his shoulder. He was scared that the man who had kidnapped him would come back for him and finish the job.

And they told me that's why Doyle felt like he needed a gun. But after the murders at Hardy Furniture in 1996, Doyle Simpson was not honest with the police about that gun. Doyle told the cops he didn't have the serial number for the gun. He didn't even know what brand it was. All he knew was that it was a.380 semi-automatic pistol, and he got it from an uncle in New Orleans. One problem with Doyle's story was that, as best I can tell, Doyle didn't even have an uncle in New Orleans. But Doyle did have a brother who lived there. His name is Robert

Campbell. He still lives in Louisiana, in the town of Hammond. Our reporter Parker went to talk to him. Robert said that in August of 1996, about a month after the murders at Tardy Furniture, Doyle called him and asked him for a favor. Doyle had called me, man, tell him you sold me the gun. I said, man, I ain't lying for you, bro. Too many people got killed. I ain't lying. I said, I ain't sold you no gun. Oh, man, go on, tell him you sold me the gun. No, I ain't telling them people that. I mean, that's my brother, but I ain't gonna lie for him. So, hey. I mean, that's my brother, but I ain't going to lie for him.

Robert told Parker that sometime that month, the cops showed up at his job asking about Doyle. They asked me, did I know Doyle? I said, yeah, I know Doyle. Doyle's my brother. And they asked me, did Doyle buy a pistol from me? That's what he asked me. He said, Doyle, you sold him a.380. I said, well, Doyle lied to you then.

I told him just like that. Doyle lied to you. I said, I didn't sell sold him a.380. I said, well, Doyle lied to you then. I told him just like that, Doyle lied to you. I said, I didn't sell Doyle no.380. I have a copy of the typed report from law enforcement's interview of Robert Campbell. And it says exactly what Robert just said. That he told them that he didn't sell Doyle the gun and that Doyle was lying. Robert said he knew where Doyle actually got the gun. Because after Doyle had bought it, he told Robert the name of the man who sold it to him. It was a man known as Three Finger Ike.

We just called him Three Finger because he had three fingers. You know, he was born with three fingers. So we called him Three Finger Ike. So that's who he got the gun from. That's who he told me he got the gun from. When did he tell you that? He told me that when I first seen the gun. I said, man, where'd you get that gun from? Man, I bought it from Ike. Law enforcement talked to a friend of Doyle's who went with Doyle to Ike's place

when Doyle bought the gun. The man told the investigators that Doyle also bought some crack at the same time. Law enforcement also interviewed Ike, and Ike confirmed that he'd sold Doyle the gun. On August 14, 1996, about a month after the murders, investigator John Johnson had Doyle brought in again for questioning. Doyle was Mirandized. He didn't have a lawyer with him, but he agreed to talk. The interview was recorded. I don't have a lawyer with him, but he agreed to talk. The interview was recorded.

I don't have the audio, but I do have the transcript. On this interview, it is not friendly. John Johnson tells Doyle, you know that we are investigating a murder. Four murders down here at Tardy Furniture Store. A place where you worked and knew the victims. And knowing this and asking for your cooperation, you were lying to us about where you got the gun. Would you please tell

me why? Doyle replies, because I was scared. Johnson says, why are you scared? Because I told you that I didn't have anything to do with it and y'all kept pressuring me, man. Well, we are only asking for the truth, because you keep lying to us, and we knew that. Doyle says, I told y'all the truth. Johnson doesn't appear to buy it. He tells Doyle, why are you going to all this trouble to lie to us, after we have explained to you that this gun has been matched and made positive identification. And here Doyle interrupts, because this business about law enforcement having matched his gun to the murder weapon? He said that was news to him.

Doyle says, you did not tell me that. Johnson says, yes, I am, and I am explaining it to you now. It's not clear whether Johnson actually knew this. The crime lab report that would later identify Doyle's gun as the murder weapon hadn't even been written yet. But law enforcement aren't required to tell the truth when they interrogate someone. Johnson keeps going. Involving four murders, your gun has been made positive identification as being the gun that killed these people.

People that you supposedly have worked with and worked for, that have done favors for you. I don't know, do y'all know that it is the same gun? Yes, we do. The crime lab has made positive identification on it. We've already explained that to you several times. Doyle says, I don't have nothing to say.

You don't have nothing else to say? No. Okay, that will conclude this statement. John Johnson doesn't arrest Doyle Simpson. He allows him to leave, and the investigation continues. And this whole time, in July and early August of 1996,

while law enforcement is investigating Doyle, they're also investigating Curtis Flowers. And like with Doyle, they didn't have anything solid on Curtis either, nothing that would absolutely prove that Curtis committed the murders. None of the route witnesses had come forward yet. That wouldn't happen until a little while later. But law enforcement was already looking into Curtis. They'd asked Curtis if they could fingerprint him, and he agreed. They wanted to find out if Curtis's prints matched the ones they'd found on the counter at Tardy Furniture and in Doyle Simpson's car. But none

of the prints matched Curtis's. They took Curtis's clothes, not just what he was wearing on the day of the murders, but other clothes too. Several pairs of shorts, a t-shirt, two pairs of shoes. They sent the clothes and shoes to a lab and asked them whether any of it had the DNA of the victims on it. The lab did not find any. I couldn't find any record of law enforcement collecting Doyle Simpson's clothes or shoes to perform similar testing. There's also no record of law enforcement testing Doyle's hands for gunshot residue. But investigators did get Doyle's fingerprints to see if they matched the ones found at Tardy Furniture. They didn't match.

I've read all the documents I could find from the investigative file back then. And in these documents, you can see law enforcement going back and forth between these two suspects. Curtis Doyle, Doyle Curtis. I've talked to people who were interviewed during this time in the investigation, like Kittery Jones, Curtis' cousin. All they wanted to hear was about Curtis and Doyle. They would start asking, you know, little questions like, you know, did Doyle and Curtis hang together, stuff like that. Law enforcement also considered the possibility that Curtis and Doyle could have committed the murders together. Curtis and Doyle did know each other. Doyle was a distant relative of Curtis Flowers.

He was Curtis's mother's stepbrother, and they would see each other at big family gatherings. They weren't close friends, though. Doyle was 12 years older than Curtis. According to Curtis, law enforcement had even tried to use him to get evidence against Doyle. Curtis testified about this in the first trial. And in that testimony, Curtis said that the sheriff wanted to wire him up to see if Doyle would say something to him about the murders. But Curtis had refused to do it. The only constant in this part of the investigation was the gun, Doyle's gun. The only question was, who pulled the trigger?

Law enforcement kept investigating Doyle Simpson for a little while longer, but by September of 1996, about two months after the murders, there aren't many investigative notes that mention Doyle. Instead, law enforcement now appeared to be focused entirely on building a case against Curtis Flowers. focused entirely on building a case against Curtis Flowers. I don't know what caused this shift. What made law enforcement turn away from Doyle Simpson?

There's no report in the file saying that law enforcement determined that he was not the murderer. There's no new piece of information in the file that would prove that Doyle couldn't have done it. I did find something curious, though, in the file. A handwritten note from an investigator. It's not signed, so I don't know who wrote it. It's dated August 20th, 1996. August 20th was six days after Doyle Simpson was interrogated by John Johnson. It's the interrogation where Johnson told Doyle that his gun was the murder weapon. This note from August 20th is brief.

It says that Doyle Simpson had called. The note lists a bunch of names and addresses of other people in Winona. And then it says, quote, Curtis, acting funny. By the time Doug Evans brought Curtis Flowers to trial, Doyle Simpson, the suspect, had disappeared. He was replaced by Doyle Simpson, the state's witness, the helper, who, though it was difficult to turn against his own family, was helping law enforcement discover the truth,

which was that Curtis Flowers had stolen Doyle's gun from the glove compartment of his car and used it to murder the people at Tardy Furniture. To prove that, it would help for Doug Evans to be able to show that Curtis Flowers knew that the gun was in Doyle's car. That Curtis wasn't just wandering around town, going to a parking lot of a sewing factory in the hopes of finding a gun there. Evans called Doyle Simpson to the stand, and Evans asked Doyle a question. Evans asked, did Curtis know that the gun was in your car? And Doyle told him, yes. Under cross-examination, Doyle's story was more shaky. He said the only reason the gun was in his car on the morning of the murders was because he'd put it there the night

before, because he was going to take it to have it cleaned. Defense attorney Billy Gilmore said to Doyle, so there is no way that Curtis Flowers would have known that gun was in that car that particular morning, was it? Doyle says, no, sir, not as I know. And then, just as quickly, District Attorney Doug Evans got back up on redirect, and he asked Doyle, Curtis knew you normally kept the gun in that car, didn't he? And Doyle said, yes. By the time of the first trial, the crime lab results were in, too, and Doug Evans told the jurors that a firearms expert had determined that Doyle's gun was the murder weapon. The defense tried to tell the jurors that Doyle had been a suspect, but when they questioned law enforcement on the stand about it,

law enforcement downplayed it. They said they ruled Doyle out as a suspect almost right away. State investigator Jack Matthews told jurors that even when the investigators found out that Doyle had lied about where he got his gun, quote, he wasn't a suspect at the time. And that contradicts law enforcement's own paperwork in the investigative file. They were listing Doyle as a suspect using that word on reports at the crime lab as late as September. In this closing argument in that first trial in 1997, Doug Evans told the jurors not to get all tangled up

in what the defense was trying to say about how Doyle wasn't trustworthy. Evans told the jurors, quote, they want to try to confuse the issue by pointing the finger at Doyle Simpson. Evans said they can't put the blame on the poor fellow that owned the gun because he didn't do it. Doyle never did, you know, speak about it. He really didn't. And, you know, after it happened, Doyle kind of shied away a little bit.

This is one of Doyle's relatives, a man named Antonio Campbell. I talked to him out on his porch one night. He never did hang around much at all no more than. Why do you think that is? I think I can't really pinpoint it, but I think that

he know he just told a story on Curtis. And I think that if Dora should have came up and just told the truth about the whole thing before he deceased. I'm just going to be honest with you. I think that Curtis should have been out a long time ago. I thought a lot about Doyle Simpson over the past year, and I still don't know what to make of him. Almost nothing in the investigation that involved Doyle was recorded on tape.

The notes are brief, and for some really key moments, there aren't any notes at all. The way I see it, there are a few options for what's going on with Doyle Simpson. Maybe someone really did steal Doyle's gun. Maybe the gun was never stolen, and Doyle gave it to someone else. Maybe Doyle did commit the murders himself and went around town saying that his gun had been stolen because, for some reason, he thought that would make him seem less suspicious. Maybe Doyle had turned against Curtis to save himself from being charged. But when you have all of these options,

what that really means is that there is no clear evidence that any one of them is true. But there is one other option that I haven't mentioned yet, and that is that maybe Doyle's gun wasn't the murder weapon after all. And to understand why, you have to know a bit about how law enforcement goes about matching bullets to guns in the first place. When you fire a gun, the bullet travels down the barrel of the gun.

And as the bullet travels down the barrel, it's picking up all kinds of scratches. Because the inside of the barrel of a gun is not smooth. It has grooves cut into it in a kind of spiral pattern. is not smooth. It has grooves cut into it in a kind of spiral pattern. The reason for this is so that when a bullet passes down the barrel, it picks up some spin so it can be more accurate, like a well-thrown football. So if you look really closely at a bullet that's been fired, you'll see some lines on it, some scratch marks that were made when the bullet passed through the barrel. Those scratch marks are what examiners in the crime lab are looking for when they're trying to match a bullet

to a gun. And people who do this kind of analysis believe that you can tell whether a bullet came from a certain gun because the grooves inside the barrel of a gun are different from one gun to another. One way to think of it is that it's like a gun has a fingerprint, and every time a bullet passes through it, the gun leaves that fingerprint on the bullet. And so, if investigators are trying to figure out whether two bullets came from the same gun, they'll take those two bullets, they'll put them next to each other under a microscope, and they'll decide whether the lines on the two bullets, they'll put them next to each other under a microscope, and they'll decide whether the lines on the two bullets look similar.

And if they look similar enough, then the examiner will declare a match. These two bullets came from the same gun. I talked to a man who does this for a living. His name is Andy Smith. He's the vice president of the Association for Firearm and Toolmark Examiners and the supervisor of the firearms unit at the San Francisco Police Department Crime Lab. I wanted to know exactly what an examiner is looking for on the bullet. Of course, it's easier to show you a picture than it is, you know, to describe it.

Of course, it's easier to show you a picture than it is to describe it, but there's actual widths to the lines that occur. There's spatial relationship between each of the lines in relation to each other. Yeah. So do you actually measure the width of the lines? We do not. Using a comparison microscope, we're not physically measuring the width of those lines. So it's an optical comparison. I mean, we're just doing it visually. Like it comes down to looking at it, basically. Yes. Like you're looking at two things under a microscope and you're making a decision about whether these appear to be basically the same,

like similar enough to say that they came from the same gun. Yes. What Andy Smith told me is that there's actually no set of criteria for what constitutes a match. It's not like you're required to have the same number of lines or the same distance between them. You just have to find what firearms examiners call sufficient agreement

between two bullets. And what sufficient agreement means is for the individual examiner to decide. In 2009, there was a big report that came out from the National Academy of Sciences. It changed the way a lot of people look at forensic science. The report showed that a lot of what passes for science in the courtroom is full of errors and overstatements and might even rely on practices that aren't scientific at all. This National Academy of Sciences report led to the creation of all kinds of groups and commissions and studies to find out whether certain branches of forensic science,

including ballistics, are valid. One of those groups is called the Center for Statistics and Applications in Forensic Evidence. The director of the center is a woman named Alicia Karakuri. She's also a professor of statistics at Iowa State University, and she spent a lot of time looking into ballistics in particular. So I gave her a call. She spent a lot of time looking into ballistics in particular.

So I gave her a call. So today, this is still a largely subjective science, in quotes. What did you mean, hypothesis, you experiment, you confirm, or put into question that hypothesis, improve your models, repeat experimentation, etc. Most of the forensic sciences really do not follow that process. And so talk about science is a little bit of a misnomer. Alicia Karakiri told me that the biggest problem is that there's no proof that each gun actually does leave its own unique marks on a bullet. It's never been tested in the kind of massive peer-reviewed study that would be required to try to prove something like that. I asked her about the way that examiners decide whether two bullets match,

about this term that they use to describe what they're looking for, sufficient agreement. Yes. Yes. Why are you laughing? Because that is such a fuzzy-wuzzy, non-scientific concept. It really does blow your mind. And so this is one of the problems, that there is no good definition of what it means to find sufficient agreement.

So what's sufficient agreement for you may not be sufficient agreement for me. you may not be sufficient agreement for me. And so you have this very undesirable situation where two examiners looking at the exact same samples might reach different conclusions. I wanted to know what Alicia Karakiri would think of how ballistics evidence was used in the Curtis Flowers case. So I started to take her through what the investigators had done. Like they went and got a bullet out of a piece of wood in someone's backyard. So they believe that that bullet that they got came from this stolen gun. And how in the world, how did they know to go look for a bullet in somebody's backyard on a tree? So they asked the guy who was a suspect.

So they say, well, have you ever fired this gun? And he's like, of course. And so he tells them he's fired it in his mom's backyard, like on this country road. And so like a bunch of people in this family would go there with like different guns. Like, although allegedly this was the only 380 semi-automatic pistol that was involved in this target shooting. And so when the investigators show up, they use this metal detector. They end up in this post area, and then they take up knife, and they pry out a bullet from this post and send it off to the crime lab. Okay, which of course means that, first of all, the bullet must have been somewhat damaged

from being shot into a post. Right. And second of all, who knows how many more marks they introduced in the bullet by prying it out with a knife. I wanted to know what she'd make of what happened next. Investigators tried to match these bullets from the post to the bullets they had collected from the crime scene. But unfortunately, the examiner who looked at them couldn't say whether or not they matched. But then, about a month after the murders, the DA's investigator, John Johnson, went back to the crime scene with three other investigators. By then, the store had been cleaned up. John Johnson testified in court about this. He said he knew they hadn't gotten all the bullets. He went to the back of the store to where Bertha

Tardy's body had been found, because he'd remembered seeing chipped paint on a brick column back there. And within five minutes of walking in the store, Johnson said, He'd found a bullet. It was inside a mattress. And Johnson took out a knife and got it out. He told the jurors the bullet was pristine. I told Alicia Karakiri about this. And so they go back into the crime scene. It's a store. It's a furniture store. And it's people been in and out since then. It not secure and they find in a mattress a bullet and so that's the bullet that they use to make this id are you serious yeah oh this is uh i'm speechless let me put it this way i mean imagine right so you found all these other bullets nothing matches to anything and then then weeks later, you go to his mattress

and find another bullet, and lo and behold, this is the one that matches. The crime lab didn't find any blood on that bullet from the mattress. I wanted to talk to the investigator, John Johnson, about all this, but Johnson didn't respond to my requests for an interview. Finally, I wanted to find out what Alicia Karakiri made of how this evidence was presented at trial, like how a firearms examiner named David Ballish had described his findings to the jurors. He says, this bullet from the crime scene came out of the same gun as this bullet from the post across town in this person's backyard. And so he says, this is what he tells the jury, he says,

when I identify it, that means I am 100% certain that these were fired in one gun and no other gun on the face of the earth. That's what he says. Is this a case that's going on now? Yes. Well, you cannot say I can tell these two things are the same or were fired by the same gun with 100% certainty. First of all, there is no such thing as 100% certainty anywhere. And second of all, even most firearms examiners today will agree that saying to the exclusion of every other gun in the universe is insanity.

Yeah, and he repeats this. So this is a case that has had six trials. So he repeats it as recently, yeah, which is all other issue. So that's 2003. And then in 2010, testifies this as the last trial, the latest one. And he's in the same, you know, same exact thing. So he says, I'm 100% certain there is no margin. If I identify them as coming from the gun, that's an absolute identification, 100%. Well, I mean, in this firearm examiner's defense, the last was in 2010, you say? Yes. Okay. so in 2010, they were still saying that nonsense, but that is absolute nonsense. And today, I sure hope that the same examiner

would be less categorical and say, I cannot exclude the possibility that these two bullets were fired from the same gun. That's about as much as he can say comfortably. I mean, that's about as much as the science today would allow. I am 100% certain. I am 100% certain that these bullets were fired from one gun.

I called David Balish, the firearms examiner who testified that the bullets were fired from one gun. I called David Balish, the firearms examiner, who testified that the bullets were 100% match. Balish is an expert witness from Michigan who says he's testified in at least 400 trials all across the country. I'm 100% certain that is my opinion. What made you 100%? The fact that I've been doing this for an awful long time and I know what an identification looks like. And what you have to do is when you're looking at these,

you have to get to the point as a firearms examiner, convince yourself that there's no other gun on the face of the earth could have left these marks. And that's what I do. How do you do that? Well, you have to do mental gymnastics, I suspect. And when you, you know, when you look at it long enough and you see how the markings are, they have to be in the same place at the same spot. But you understand why that takes place. But that's how you come to the opinion. I talked to Balish for a long time, almost two hours, and basically what he was saying was that this whole process of how he matched the bullets in the Curtis Flowers case, it came down to mental gymnastics. It's the sort of thing where you have to know it when you see it.

I wanted to know if David Balish was aware of all the criticism that this type of forensic science has received in the past decade or so. Yeah, have you read that National Academy of Sciences report that came out in 2009? I think I perused it one time. They sort of want to be ambiguous. You know, they're trying to be politically correct, I suspect. What do you mean? to be politically correct, I suspect. What do you mean? You know, people now want to be able to say that everything can be assigned, you know, a percentage or have to be absolutely sure that

all of this is one way or the other. You know, you want to make it a science, and it's never been called a science. It's been always called an art form using scientific materials and equipment, and it's always been an opinion. But it should be based on facts, right? I'm sorry? But it's an opinion based, it should be, it's an opinion based on facts, right? Or based on science? Well, I don't know. Sometimes, you know, the fact in somebody's mind may or may not be a fact in somebody else's. The reason that I wanted to talk to you is because it does seem like you were so certain at trial, and yet the state of the science says you can't be certain. And so I just was curious if you had changed your view of it.

If you would say, well, actually— Not at all. Not at all. Okay. David Ballish says that if Curtis Flowers is tried a seventh time, he expects he'll be called to testify. Meanwhile, the gun that was used to kill the people at Tardy Furniture has never been found. Still out there somewhere.

The case against Curtis Flowers comes down to three main things. The route, the gun, the confessions. Next time, the confessions. Next time, the confessions. In the Dark is recorded and produced by me, Madeline Barron. Senior producer, Samara Fremark, producer Natalie Jablonski, associate producer Raymond Tungakar,

and reporters Parker Yesko and Will Kraft. In the Dark is edited by Catherine Winter. Web editors are Dave Mann and Andy Cruz. The editor-in-chief of APM Reports is Chris Worthington. Original music by Gary Meister and Johnny Vince Evans. This episode was mixed by Veronica Rodriguez and Corey Schreppel. To see photos and videos, and to find out more about what happened to Doyle Simpson in the swamp,

check out our website, inthedarkpodcast.org. Hi, this is David Remnick. I'm proud to share the news that three films from the New Yorker documentary series have been shortlisted for the Academy Awards. And they are Incident, Seat 31, Zoe Zephyr, and Eternal Father. And they all immerse you in the finest cinematic journalism, exploring themes of justice, identity, and the bonds that shape us. These extraordinary films, which were created by established filmmakers as well as emerging artists, will inform, challenge, and move you. I encourage you to watch them along with our full slate of documentary and narrative films at newyorker.com slash video.