Our guest, Barbara O’Brien, explains why we don’t know much about conviction error outside of murder cases, making error rates for the vast majority of crimes: misdemeanors, sexual assaults, armed robbery, etc. a “dark ocean”.
What is Manifold?
Steve Hsu is Professor of Theoretical Physics and Computational Mathematics, Science, and Engineering at Michigan State University. Join him for wide-ranging conversations with leading writers, scientists, technologists, academics, entrepreneurs, investors, and more.
Corey: I’m Corey Washington, and this is Manifold. Steve Hsu is away this week. Our guest today is Barbara O’Brien. Barbara is a Professor at Michigan State University College of Law, where she teaches classes in criminal law and criminal procedure. She is also currently the editor of The National Registry of Exoneration, which collects, analyzes, and disseminates information about all known exonerations of innocent criminal defendants in the United States from 1989 to the present. The Registry provides a virtual home for exoneration stories and also an accessible, searchable statistical database about the cases. Today, we’ll be speaking with Barbara about her work with The National Exoneration Registry, new developments in the fight against wrongful convictions, and how race and class affect the likelihood that a person will be wrongfully convicted and how long that person will spend in prison. Welcome to Manifold, Barbara.
Barbara: Thank you for having me.
Corey: How do you define a wrongful conviction?
Barbara: Well, we document exonerations, and exoneration is a type of wrongful conviction, and sometimes we use the term wrongful conviction interchangeably with a false conviction. But you could think of a wrongful conviction as much broader than that to include any case where somebody was convicted of something that was more serious than what they actually did or convicted without proper process. But when we talk about wrongful convictions in this context, we really do mean somebody who was actually innocent of the crime. Either there was a crime and somebody else did it, or there was no crime at all.
Corey: So wrong guy, wrong gal?
Barbara: Yeah, in many of the cases it is…
Corey: Or no crime, perhaps?
Barbara: Sometimes it’s there’s no crime. In fact, we have quite a few cases in the Registry where there just wasn’t a crime.
Corey: This is an issue that actually goes back centuries, as I’m sure you’re aware. I think some of the first laws designed to prevent wrongful convictions involved cases where actually someone disappeared. And I guess they just sort of moved away. At one point in time, I think the original British case, three people were hanged for it and then the guy showed up.
Barbara: Yes. Actually, we do have, our database begins with exonerations that occurred from 1989 on, but we have another database that tracks historical exonerations that go back as far as we can find out about in the United States. It’s not as comprehensive. It’s a lot harder to find records, but in some of those cases the victim will show up at their funeral. And before the person could be executed they were exonerated, so we do see some of those cases in our pre-1989 database.
Corey: And this led to laws that require a body actually be found, I think.
Barbara: Yeah. Those rules, I don’t believe they’re universal, but that’s a pretty common rule, and at the very least as a practical matter it’s very hard to prove a murder without a body.
Corey: I don’t want to get sidetracked on this, but I do think it’s interesting that you brought up the fact that there’s a difference between wrongful conviction and exoneration, because I think there’s clearly a large group of people who are exonerated, but it sounds like there’s also even a much larger group of people who may be convicted of something, they may be a little bit guilty, maybe slightly involved, but convicted of something far more serious. And they’re never exonerated and they’re often never given clemency or not have their sentences reduced. Are you worried that the focus on exonerations may basically cast a shadow in a way, or at least lead to a lack of attention on this broader range of cases?
Barbara: Well, that’s correct. If you think of exonerations as a type of wrongful conviction, but there are … The kinds of wrongful convictions you’re referring to where someone was convicted of a murder when it really was a manslaughter or something along those lines, I think there’s two ways the focus on exonerations, it can go, and maybe both can be true. That the focus on actual innocence, pure innocence, might make people less sympathetic to the cases of over-conviction, which I think is a huge problem, particularly overcharging that can lead to very serious convictions for something that … Really should have been a lesser conviction. That’s possible. On the other hand, what I think you see happen is as people get interested in the phenomenon of false convictions, they’re exposed to all these different ways that the justice system is failing. Whether it’s by mandatory minimum sentences that lead to these Draconian outcomes, or cases where people are shuffled through the justice system, sort of assembly line justice.
Barbara: It really just highlights … It brings in stark relief these problems that are really salient in the cases of false convictions, but are actually undermining the fair administration of justice at all levels. And I think you can kind of see that. I do follow a lot of the popular press on false convictions, and what you see are the people getting interested in this idea that somebody gets convicted of something they didn’t do, but they’re also, then, really interested in the idea that somebody gets in the system and then they have a probation violation and then they can’t get out of the system. I think a good example of that is Serial season one, everybody knows this case that a lot of people think that is a wrongful conviction case. Serial season three is about the justice system, looking at it through the lens of one courthouse and all the different stories that happen in that courthouse in Cleveland, Ohio. And those are not necessarily cases of false convictions, but it highlights the uphill battle people face.
Barbara: Maybe they did something wrong, but then they get brought into the system and they miss a court date, and the next thing you know there’s a warrant out for them, and things just sort of snowball and get worse and worse. It’s going to be a lot of the same audience, who listens to Serial season one is going to listen to Serial season three.
Corey: I think there’s been a lot of interest these days in basically looking back over our purely severe policies on crime and drugs of the past 30 years, and realizing exactly, people overcharge for crimes, and once you get into the system it’s very hard to get a job once you get out and then you’re essentially given a mark for the rest of your life that makes it really, really hard. And it seems like there’s at least an inclination now to look at a bunch of those convictions and maybe wipe them out, or in many cases reduce them.
Barbara: Yeah. I think of exonerations, and interest in wrongful convictions is sort of a gateway to these other issues. And one big development in the last decade has been prosecutors’ offices, in mostly major metropolitan areas, creating conviction integrity units where they’ll actually put prosecutorial resources and police resources into revisiting cases where there’s a plausible claim of innocence. And there’s talk in starting sentence review units, and some conviction integrity units actually don’t just stick to the wrong man or false conviction cases, but they are interested in reviewing all sorts of cases where there’s overcharging or over sentencing.
Corey: I recall I think the first prosecution integrity office was started in Texas, wasn’t it?
Barbara: I believe so.
Corey: The sort of ground zero for both death penalty and wrongful convictions. And perhaps wrongful executions too, as we may get into.
Barbara: It’s so interesting because on one hand, Texas is a pretty law and order place. They have a death penalty and they use it, but on the other hand they also have one of the best compensation systems in the country for wrongfully convicted people.
Corey: We have about the worst here don’t we, in Michigan?
Barbara: No, I think there are places that might be worse, actually. There are places like Pennsylvania that doesn’t have any system in place, but they’re working on that there. Texas has, and these were in reaction to cases of, just really stark in justice, has more liberal discovery rules than other places do. You do see this happen in cases where a really high profile terrible crime can lead to a new criminal law based on it. Megan’s law or something like that. The same thing can actually happen with criminal justice reform, so the Michael Morton act in Texas was passed in response to a case of wrongful conviction where Michael Morton spent more than two decades in prison for murdering his wife. And he didn’t do it, the state had hidden all sorts of evidence that implicated another person.
Corey: So straight forward Brady violation?
Barbara: Oh, yes, yeah. Really egregious brady violations.
Corey: Let’s just stop at this point and define brady violations.
Barbara: Yes, good idea. So brady versus Maryland is a supreme court case in which the court held that the government is required under the fifth amendment to disclose to the defense any evidence that is both exculpatory and material.
Corey: How can it be exculpatory without being material?
Barbara: Well, there’s a lot of problems with the materiality component because it’s always a judgment made in hindsight and so the judges are deciding would this have made a difference. But exculpatory doesn’t mean that it’s dispositive, right?
Corey: Sure, sure.
Barbara: Exculpatory could be, they talked to a witness, they have eye witnesses, and they talked to one witness who described the perpetrator very differently than the other witnesses. Or they get a lead on another suspect and they don’t follow it and they don’t turn that information over to the defense. It’s not necessarily a smoking gun that proves the defendant’s innocence. But it tends to favor the defense. Or it also includes impeachment evidence, so if the witness who is testifying against the defendant is getting a reduced sentence in consideration for their testimony, and the state doesn’t disclose that, that is seen as or considered as exculpatory evidence. It doesn’t prove the defendant’s innocent but it’s something that’s exculpatory in that it tends to favor the defense.
Barbara: And so in cases where the state doesn’t disclose favorable evidence to the defense, it doesn’t mean the defendant will necessarily get a new trial or a new sentencing hearing if the reviewing court does not believe that it was material. There’s a reasonable probability that the outcome would have been more favorable had the state disclosed this evidence to the defense.
Corey: That’s clearly a really, really sensitive judgment there on probabilities. On how accurate they are in making that judgment.
Barbara: Yes. They have to, by definition you’re engaging in hindsight and you’re looking back and saying, what would the jury have done with this? Or even more attenuated, you’re asking what would the defense attorney have done with this information? Because once it’s turned over the state doesn’t have to tell the jury, they have to turn over to the defense and the defense is supposed to make whatever use of it they can. It’s also a practical matter, brady material is by its nature hidden, if it wasn’t hidden it wouldn’t be brady material. So it’s just serendipity that you as the defendant would come across it and be able to make the brady claim.
Corey: The failure to turn over evidence in the brady violation led to this new law in Texas, the Morton law?
Barbara: It was a case of wrongful conviction in which the post conviction defense team uncovered some really egregious misconduct on the part of the prosecutor. That led to Texas passing the Michael Morton act, which puts a lot more onus on the prosecution to disclose evidence than the constitution requires.
Corey: This leads us to our general discussion about the causes of wrongful convictions. I think the first DNA exoneration was about 1987, I think in England they freed someone. That was the point where people really began focusing on wrongful convictions, we’ve had a lot of research since then about what leads to wrongful convictions and exonerations. One of them, I take it, is failure to disclose exculpatory evidence, what are the other leading factors that we know now cause wrongful convictions?
Barbara: Perjury and false accusation is the most prevalent contributing factor in the exonerations that we see. Mistaken witness identification, forensic error, or even fraud, is a significant contributing factor. And just sort of official misconduct of all kinds. Whether it be witness tampering, coercive interrogations, it really depends. We often talk about the causes of wrongful convictions, we actually frame it at the registry as more of contributing factors, right? Because it’s really hard to say if this hadn’t happened we wouldn’t have had the wrongful conviction. And often any case of wrongful conviction has more than one factor, that’s very frequent.
Barbara: And really, we often think about what are the leading causes or factors associated with false convictions but it really depends a lot on the type of case. In cases, you don’t see too many false confession cases outside of murders. Because getting a confession from someone, getting a false confession from someone, if it’s the product of really extensive interrogation, I’m not talking about a case where somebody is mentally unstable and they voluntarily come forward and falsely confess. That does happen. But the cases where the interrogation produces a false confession, those are significant vestment of investigatory resources.
Barbara: you’re not going to get somebody in the room and interrogate them for eight straight hours because somebody broke into a car and stole something out of it. It’s going to be a case with very high stakes so you are more likely to see a false confession in a case that’s a murder case, like a really serious case. Whereas you’re more likely to see mistaken witness identification in a rape case, right? You’re less likely to see it in a murder case because you don’t have the surviving victim to make the identification, you may have other witnesses but you’re less likely to. So in cases of child sexual abuse that result in exoneration, those are frequently perjury or false accusation, and sometimes faulty forensics. It really does depend on the type of case, the type of contributing factor that you see.
Corey: And it seems like there’s actually an upstream cause in the case of these false confessions because the police essentially develop tunnel vision, become convinced it’s one guy who committed the crime then tried to extract a confession out of him. Is that right, they tend to zero in on somebody wrongly and basically, I’ll be blunt, beat it out of the person, or interrogate out of the person. Is that, that’s my image of how these wrongful confessions happen, it’s not that someone’s crazy, they’re sort of pushed to a confession by the investigators?
Barbara: Yes, I think that is true. We don’t have great data on the length of interrogation. We do keep track of it when we can tell from the materials that we’re working from how long the interrogation lasted. But you’re right they do tend to be pretty long. There are some cases where the defendant confesses falsely very quickly, those are the cases that involved violence. There are occasionally cases where somebody who has some sort of mental illness might confess but you’re right, that’s not the typical situation.
Barbara: It’s usually a situation where the detectives really think they got the right person and they’re trying to persuade the person to tell them what happened. But we see all kinds, we do see some cases where physical violence is used. In Chicago there’s a sorted history of that.
Corey: In LA also, with the ramparts team.
Barbara: Yeah, and unfortunately the ramparts, it’s hard to get data on and it’s hard to know even just how many defendants were implicated.
Corey: We should back up and give people the context on it.
Barbara: You’re right, we should.
Corey: The Chicago case was a series of police officers effectively taking black defendants down into the basement of these precincts and torturing them until they confessed. Now, were those mostly murder cases or were there other types of cases where the police were convinced they had the person and then effectively tried to beat a confession out of them?
Barbara: Yeah, they’re mostly murder cases, the exonerations that resulted from it. It doesn’t mean it wasn’t happening in other cases but as far as the exonerations and the registry, they are almost entirely murder cases. There was another scandal that is still unfolding in county, which is Chicago involving a detective Renaldo Guevera, he was accused and there’s pretty substantial evidence that he also was abusing suspects more recently. You would think after burgess’ case had unraveled that, you know like one bad apple. But there’s a bit of a culture problem in the county police department that this has continued to happen. There’s often a single bad actor who is the common thread in these cases but they’re working with other people too.
Corey: Yeah, I thought it was a team in some of these cases.
Barbara: Yeah. Burge was in charge of a team. There’s another scandal out of Cook County that came to light in the last few years. Sergeant Watts was going to one particular public housing complex and he was personally involved in drug dealing and shaking down drug dealers as well. And he would go to people he saw, he would just see someone, usually a guy, standing around or walking into his apartment. Stop him and ask him where the drugs were and if he had any drugs, and the guy would say I don’t know anything about that, he would say you’re under arrest, put drugs on him, and framed him.
Barbara: And defendants were complaining about this, saying the cop planted these drugs on me. But who is going to believe that, right? A few of them went to trial but most of them entered guilty pleas because that’s just not a defense that typically works. It wasn’t until the FBI got involved and investigated Sergeant Watts and some of his colleagues and they caught him on tape engaging in shaking down a drug dealer and everything started to unravel. I think we haven’t seen the end of those Watts exonerations yet.
Corey: So this leads us into a topic that I really wanted to get into with you. Which is the racially tinged character in many of these cases. It seems like the defendants, the target in these cases are black and I think that’s also true in the case of ramparts in LA.
Corey: It seems like there are many areas where there are allegations that there’s racial bias in the criminal justice system, and once of the chief empirical questions of whether these disparities remain after you control for confounding factors like rates of offending. But in these large scale cases of misconduct it seems that there’s very little doubt that the overwhelming majority of the targets are black. Would you agree with that?
Barbara: Yes. Or Latinx people too. It’s a really interesting question in terms of what do you control for? If you imagine that you could say, the rates of most black defendants are overrepresented among murder defendants and so that explains the higher number of black defendants who have been exonerated of murder. But then you have to take a step back and think, well they’re victims of the disproportionate murder rate, in the same way that black victims are overrepresented among murder ratios.
Corey: Yes, that’s right.
Barbara: The people who didn’t commit the crime are also victims of that same disproportionate level of violence. But at the same time that is different than saying that they were targeted because they were black. It is a product of it and it’s not something that they’re doing that’s causing it.
Corey: Would you argue that there’s a uniform error rate across racial groups? You make an arrest it leads to a conviction, there’s some x percentage of people who are mistakenly arrested and convicted and that could be constant across racial groups. If you have say 50 percent more black defendants because black people are committing 50 percent more murders and if you find 50 percent more are being exonerated you would not be surprised, right? Because it’s going to be consistent. But it looks like what you’re finding in many of these cases is a far higher proportion than actually the offending rates, suggesting that there’s some other factor involved.
Barbara: Yeah, no, there are definitely other factors. Particularly in sexual assault cases if you look at the best crime estimates, black perpetrator white victim rape cases are unusual.
Corey: I think they’re about 11 percent of all cases.
Barbara: They’re a small percentage. Whereas a much higher percentage of sexual assault exonerations are interracial.
Corey: In half of sexual assault exonerations, eye witness misidentification let’s say, black men were convicted of raping white women. We actually need the data for how many are exonerated overall but it looks like in about half of cases a witness misidentification, it’d say black men accuses of raping white women.
Barbara: Right. People are terrible at eye witness identification, that’s how people are across the board. But we’re particularly bad at identifying a stranger of a different race, particularly white people identifying black people. That’s not the same thing as an overt racial animus but that does explain some of the disparity that we see. But we also have lots of cases with explicit racism where people became suspects because they were a black person in a white neighborhood, that they didn’t belong there, they stood out. Or it was a black man dating a white woman and that’s what brought the police’s attention to them. And some even more overt cases of explicit racism.
Corey: One of my few encounters with the police was when I was a graduate student in Berkeley and I was crossing the street and this police car comes up, stops in front of me and asks me to come over and put my hands on the back of the car. I comply and he pats me down and he says, yeah we have reports of a black guy who robbed, I think he said massage parlor but hard to say. Anyway, purely on that basis I was pulled over and who knows, if I looked more like the guy maybe I would have ended up handcuffed. It fits the pattern, I guess.
Barbara: Right. It shouldn’t be that surprising that we see these disproportionate, we see a higher rate of exonerations that black people are overrepresented in among exonerations than what you would expect. and if you start to think about it, there’s all sorts of things that could be driving it. Which is, if crime tends to be intra racial and there’s an eyewitness, and the perpetrator is described as a black man, a black man is more at risk of being falsely identified as the perpetrator.
Corey: Before we move on, I just want to drill into sexual assaults a little. Because I think it’s a pretty dramatic area, if that’s all right. I believe in the case of sexual assault exonerations that it’s not just eye witness misidentification that leads to a disproportionate number of black men being falsely convicted. There are other factors that go on, prosecutory misconduct seems like, and not just prosecutory but police misconduct seems like it plays a role in these cases.
Barbara: I think that’s true. Part of the problem with studying this is you often don’t have the smoking gun as to what the motivation was. We try to be as objective as possible in finding the presence or absence of certain factors and when there’s evidence that comes out that hadn’t been disclosed and there’s evidence the prosecution had, then we can call that prosecutorial misconduct. But in terms of their motivations unless they say something in any one case it’s really hard to say that we can prove this is racially motivated.
Barbara: But I think when you look at the patterns, even if you can’t identify a particular case and say this prosecution was racially motivated. Particularly if it’s not the overt explicit racism that you do see in some cases. If it’s a case where the defense attorney says you should really take the deal because they’re not going to believe you, who’s going to believe the black defendant accused of rape over this nice victim. Or a prosecutor or police detectives who are more inclined to accept a certain narrative because it’s consistent with their stereotypes. These are things that are not necessarily going to leave a paper trail. You’re not going to be able to follow the decision making and figure out what was driving it. I think we can be confident, maybe not in a particular case that this is what’s happening, but that this has to play a part on some level.
Corey: No doubt. We have a lot of anecdotes and it’s very hard to assume we have an unbiased sample of these cases, right? So you can pull out a case where it looks like we have evidence and we don’t know how many comparable cases there are where a police officer effectively set up a biased line up for a white defendant either. I agree, it’s difficult to get really solid evidence.
Barbara: Right, but If we think about it in terms of the context of history and there’s a big difference between saying, I am convinced that this person was convicted because of their race, versus saying based on our history in our not too recent past, it wasn’t that long ago where people didn’t even hide these motivations, I feel very comfortable saying that racial bias contributes to false convictions. The other problem with any study of wrongful convictions and by that I mean false convictions, is that we only know about false convictions that result in exonerations. We don’t know about the cases that the person doesn’t achieve exoneration and there’s so much luck involved in being exonerated if the evidence was destroyed, you wouldn’t believe how many evidence rooms in police departments have experienced floods over the years. It just seems like I always seem to be reading about that. The evidence was lost when the evidence room was flooded 15 years ago. There’s so much luck involved in getting exonerated and think about it, imagine your prototypical exoneration is a rape in which the only issue is who did it.
Barbara: And you have a victim who identifies the defendant in a line up or a show up and that we learn years later when the DNA is tested that it couldn’t have been this person. We know we made a mistake. Here you have a situation where you have a eye witness who is the victim, who had a very clear opportunity to see the perpetrator often who has absolutely no incentive to not be accurate. They have every incentive to get it right and they still made a mistake. Now think about that in context of an armed robbery where you have the weapons effect, where the weapon that the perpetrator is holding draws the eyewitness’s attention away from their face, usually a shorter amount of time to observe the person.
Barbara: But you’re not going to have DNA in vast majority of cases to prove that the person didn’t do it. So you see far more rape exonerations than you do armed robbery exonerations but we have every reason to think that the same problems that would produce a false conviction for rape, that also applies in an armed robbery case. We’ll never know. We’ll just never learn about it. The only sample we have to look at is the university of exonerations, that’s our best proxy for innocence. But it’s skewed, we know it has to be skewed.
Corey: When it comes to race and sexual assault one of the statistics I used to quote when I taught the death penalty when I was a philosophy professor, is I point out that everyone sentenced to death for rape prior to 1950 in the United States was black. And that was just before they had outlawed the death penalty for rape. But it shows that there’s been a consistent component especially involving race of the victim in sexual assault cases. This probably brings us to an empirical question, how much do we know about the rates of wrongful convictions? I think we have good estimates in case of the death penalty based upon exoneration, using exoneration as a proxy for innocence. And what’s the number in that case and what are the estimates beyond that if any?
Barbara: That’s a million dollar question, right? Because of how little we know about false convictions because we have to use exonerations as our proxy, whenever I talk about this I would start with the caveat that there is so much we don’t know and it’s not because we haven’t tried hard enough. It’s just there’s so much we don’t know. But you’re right, death sentencing is a really interesting, people who have been sentenced to death are a really interesting population to look at. Because they get a tremendous amount of scrutiny, their cases are more likely to have counsel all the way through the process than someone who is sentenced for even a very serious crime with a very long sentence. So they’re not like other defendants in that way and we do see a much higher rate of exoneration among death sentenced defendants than we do among other populations.And there’s only two possible explanations for that, one is we make more mistakes in death penalty cases and that we are more likely to catch those errors. And they’re not mutually exclusive.
Corey: Why would you think you make more errors in death penalty cases?
Barbara: I don’t know if we do, but that would be the only explanation. That is one of two explanations of why we see such higher rates.
Corey: One of the reasons I’ve heard is that these cases are just so hot that the police really feel like they have to solve it. They’re going to get somebody regardless of whether they’ve gotten the right person.
Barbara: That is right. Both could be true to some degree. They’re not mutually exclusive explanations, but those are the only two possible explanations.
Corey: It seems like the second one is extremely plausible because so many resources are dedicated to this.
Barbara: That’s right.
Corey: Whatever you think about somebody, you don’t want them to be killed for a crime they didn’t commit.
Barbara: That’s right. But it is also possible that we do make more mistakes in capital cases because if you think about it from the perspective of just a prosecutor who is trying to make the best use of their resources, if you have a weak case of somebody breaking into a car, you’re not going to pursue this case and take it to trial. You might, there are always examples of it, but it’s less likely. But if you have a crime heinous enough to be capital, you are going to pursue it if you can, right? Even if there might be some flaws in your case, some holes in your case. So it is entire plausible that in capital cases an innocent person is more likely to be convicted just because the same level of evidence in a less serious case just wouldn’t be brought.But I would speculate that most of the difference we see is due to the greater scrutiny and that if we applied the same level of scrutiny to cases that did not result in a death sentence that we would probably have a lot more exonerations from that population as well.
Corey: I think there’s an actual estimate out there, was it you and Samuel together?
Barbara: And two bio statisticians as well.
Corey: Can you review those findings?
Barbara: Right. I went to university of Michigan law school along with two bio statisticians, Edward Kennedy and Chen Hugh. And we did a study using people who were sentenced to death and looking at exoneration rates among those prisoners sentenced to death, and used survival analysis. Which is something epidemiologists use.
Corey: It’s a really interesting application of that technique, kind of an inverse application.
Barbara: That’s right. Because we were looking at the likelihood of being exonerated, and that’s a good thing, but usually when you’re doing survival analysis you’re looking at mortality, which is a bad thing. So it took a lot of getting used to, think about it that way. We estimated looking at rates of exoneration among death sentenced prisoners, had they stayed on death row for a certain amount of time, I believe it was 20 years, you’d expect to see a rate of exoneration of 4.1 percent. Because people leave death row all the time for reasons not due to exoneration. People leave death row because they die of natural causes, because they get their sentence reversed, or they are executed.
Corey: This was one of the disturbing implications of this paper which is that if you leave death row and get sentenced to life in prison, that actually might be worse for you because people stop paying attention to your case. There’s a good chance a whole boatload of people who are actually innocent and just serving life sentences now.
Barbara: I obviously have no idea how many but I think that there are definitely people out there who have been re sentenced and removed from death row, who had they stayed on death row would’ve been exonerated. In part I think that because if you’re an attorney arguing for someone who’s been sentenced to death that they should get relief, often the legal arguments you make require some sort of re examination of the evidence. In that was the error harmless? So the way to the evidence matters in whether you can get relief on a purely legal claim. What that means is somebody who has not a great case against them, it was a close case, they’re more likely to get relief on a legal error that doesn’t result in exoneration.The people who leave death row and get re sentenced I think probably on average those are the people with on average weaker evidence against them.
Corey: Wow. And thus more likely to be innocent.
Barbara: That seems very plausible to me.
Corey: Do we have any numbers beyond the 4.1 percent for the likelihood of a wrongful conviction?
Barbara: There are studies that try to do similar types of exercises looking at cases in which, examining DNA exonerations and trying to draw estimates from that. It’s just incredibly difficult to make estimates. The only reason I think that we felt comfortable doing so in the context of the death penalty, and even then with a lot of caveats that it’s just an estimate, is that those are the cases that a, the records are the best. You’re most likely to get the underlying records, the opinions tend to be published opinions, there’s more written petitions that you can get access to, and also that they get the kind of scrutiny that they do.
Corey: It’s not just in terms of justice, you have outside groups, innocent projects that take an interest in cases and do an incredible amount of external due diligence. So it’s sort of an external actor in some sense driving these changes.
Barbara: That’s right. We call these actors professional exoneraters. They are innocence organizations and also conviction integrity units in prosecutor’s offices that are huge drivers of exonerations. There’s been a proliferation of them, I believe when I last looked there were 56 conviction integrity units in the United States, still a very small percentage but a lot more than there used to be. There’s also roughly the same number of innocence organizations, there may be some more. Some are better resourced than others but all of them get more requests for assistance than they can help with, so they have to triage. And part of their triage is just figuring out which are the most likely to actually lead to exoneration.But actually a lot of them have a requirement that you have to have a certain number of years left to serve on your sentence.
Barbara: They have only so many attorneys to go around.
Corey: So if you’re close to the end your sentence they’re not going to give you attention and thus those people are likely to be lost.
Barbara: There are people who have been exonerated even after they’ve been released but that’s much less common. And if you talk to any of these attorneys it’s not that they think that an innocent person who only has a year or two left to serve, we say only a year like that’s nothing, that’s actually a long time. Or who has actually been released on parole, it’s not that they don’t have a lot at stake, if they’re innocent they deserve exoneration, it’s just that they have to pick which cases they’re going to put their resources behind.
Barbara: And depending on where you are, if you’re an innocent defendant who has been convicted, whether you’re ultimately exonerated may depend on just is there a well resourced innocence organization in your jurisdiction that’s willing to help you. These are factors that have nothing to do with the underlying strength of your claim, it’s just serendipity. Just like if there’s biological evidence that existed that could exonerate you but it’s been destroyed in a flood or it’s just a department that had lousy evidence retention policies. That’s totally out of your hands but that could make all the difference in whether you ever get exonerated.
Corey: Just not wanting to waste resources on cases that they perceive as comparatively low value, this is leads to an interesting issue of wrongful execution. Because of course innocence projects don’t dedicate a lot of effort to figure out whether someone who’s executed is innocent because they’ve already been killed. But I believe there are laws passed in some jurisdictions prevented the post death testing of DNA. Is that right?
Barbara: I don’t know the specifics but part of it is just trying to find the best place to use your resources where you can do the most good. But also just legally how do you get into court and say, hey judge, compel the state to turn over this evidence to get tested, when the case is closed? There’s no active case anymore, there’s often no legal avenue in which to do it, the state isn’t going to agree to it and the court doesn’t have a case before it. So there really isn’t a mechanism.
Corey: That’s bizarre. It seems like one would want to know.
Barbara: Yes. Or do you?`
Corey: Yeah, I guess if some people have different interests they probably don’t want to know.
Barbara: Yeah. We do see posthumous exonerations. We do have some posthumous exonerations in the registry.
Corey: People who have been executed or people who have died?
Barbara: People who have died. And it’s often a case where there’s a family member who has fought very, very hard and often the process has gotten started already, the legal mechanism for doing so, the petition for post conviction petition has already gotten underway. They’re a lot less common. But when somebody has been executed that’s loaded.
Corey: I was a bit obsessed with this a couple years ago so I dug into the literature on wrongful executions. There are a couple cases that really stand out, Carlos Deluna really stands out, Todd Willingham really stands out. I think there’s been a lot of literature written about both cases. There’s a real consciousness now and as far as I know, part of the reason the American population has begun to turn against the death penalty, at least from surveys, people worry about wrongful executions. I don’t know if these cases have driven it but it seems like the number of exonerations has slowly steeped into public consciousness as potentially leading to the possibility of someone will be wrongfully executed. It seems to be effecting public opinion, judging on what people tell surveys.
Barbara: I agree, I think this gets back to your question at the beginning when you asked about is the focus on actual innocence detracting from some other justice issues. But if you think about it in terms of the death penalty, imagine if 4 percent, so just go with the 4.1 percent. That’s an unacceptably high number, let’s even say we doubled it, right? And we say it was 8 percent are actually innocent, it’s still most people who are convicted and sent to death row are actually guilty. The danger that comes from executing an innocent person often makes people say it’s not worth it even if 9 out of 10 are guilty, it’s just not worth it. But we can’t do it right, it’s not worth it to have this penalty. We can do without it, there’s lots of states that don’t have it, they’re not falling apart.
Corey: Yeah the numbers have been going down over the past few decades.
Barbara: Right. A number of different things I think are driving that. One is I think public opinion is shifting on the death penalty, it’s not as strong as it used to be and part of that is driven by the innocence movement, the fact that people are recognizing that the errors are more common than what we would consider acceptable. And also crime has gone down and that’s obviously a big part of it. The quality of representation in a lot of places has improved dramatically.
Corey: Yeah, I think this actually is an undiscussed component of exonerations, I think many people have said, the people that are falsely convicted are almost always poor and almost always have a public defender. So it looks like if the equality is growing up, you would see fewer wrongful convictions. Is that because of these external groups coming in?
Barbara: Well, yeah. I would say we don’t have financial information on the exonerations in the database but suffice it to say that we have our share white collar criminals who are exonerated. But for the most part it does seem to be people of lower socioeconomic groups. It’s also hard to assess the quality of their representation based on the information that we have. In death cases, I’m thinking about North Carolina which is a jurisdiction I know best in terms of the death penalty. There were cases in the nineties and the eighties where somebody would be appointed, there wouldn’t be a public defender, a public defender at least that’s what they do all day. And if they are well resourced enough you can often do no better than a public defender because that is all they think about all day long and they’re really well trained.
Corey: But the key caveat there is well resourced.
Corey: Often these public defenders have 40, 50 cases on their docket.
Barbara: Oh, more than that. Way more than that. There will be cases where somebody who his only experience is-
Corey: A tax lawyer or something like that.
Barbara: Or he’s done real estate deals and they don’t appoint more than one lawyer for a death case.
Corey: And death cases are very expensive to actually investigate.
Barbara: Yeah but they wouldn’t actually put the resources, the court wouldn’t approve funds for an expert or proper investigation and so North Carolina has a defense commission where they mandate that there have to be two attorneys to represent somebody charged with a death case, they have to have a certain level of experience, there’s money for experts. You see the quality of representation go up.
Corey: So that’s changed you think by statute in general or just by public pressure?
Barbara: Well, the public pressure is what leads to the statute. I think it’s not universal but it’s much more common now to see more rigorous defense standards for capital cases. When i think about the limits of exonerations as a proxy for innocence, it’s the best proxy we have. But where we’re really missing is in the low level misdemeanors. We have absolutely no idea how many of those cases are false convictions, we have some in the database, but when you think about the under resourced public defender’s office, or even worse, the court appointed attorneys who are taken on a contract basis who have very low caps on how much they’re allowed to bill for a particular case. There’s no resources for investigation, there’s very little time to do any consultation, and the stakes are seen as very low so people are encouraged to plead and they say then you’ll be done, you don’t have to serve any time. But there’s all these collateral consequences that they suffer once they’ve been convicted even of a misdemeanor.
Barbara: I think that’s the real great unknown. I think of it like the ocean, how much do we actually know about life in the ocean? That’s the deep depth that we don’t know, I know it’s bad. I believe it’s really bad the rate of false conviction among misdemeanor defendants is high but we’re going to miss most of it.
Corey: Your paper actually connected with the recent discussion of sexual assault because you see this number quoted consistently which is about 8 percent of sexual assault accusations are false. Having read your paper it’s quite clear to me that people have no idea.
Barbara: We don’t have any idea.
Corey: I think some have actually looked to that figure and they could not trace it back to a source. Actually there was a source but the source had basically no evidence. The only way to figure out whether someone is telling the truth or not is to follow the cases to the end, have a large sample and see how many are exonerated. In that case you would get a lower bound on it. But that’s never been done on sexual assault.
Barbara: We get asked this all the time. They’ll look at how many cases involve perjury or false accusation and they’ll look at sexual assaults or rape cases and say look at all these cases that involve perjury. But it’s often a police officer lied about some aspect of the case and we would count that as perjury or false accusation but in terms of victims coming forward and lying, saying this happened when it didn’t, we have cases like that, it does happen. But anybody who claims to know how often or what percentage of them are, basically what I can say confidently is it has happened. It is some number of cases but I would never venture a guess as to how often it happens. Based on exonerations, anyway.
Corey: I want to turn now to some of the possible solutions to the problem of wrongful convictions. Technology has changed a lot since we realized this is a larger problem, we now have DNA testing, there’s more cameras everywhere. So it seems like right now there’s something of a conflict between people who are working against wrongful convictions and civil libertarians, right. It seems like one of the best ways to reduce the number of wrongful convictions would be have every male on the planet contribute DNA to a database. Therefore we’d be able to tell whether someone was wrongfully convicted of rape.
Barbara: Not necessarily. I hear what you’re saying and I don’t disagree with some of it. If I’m accused of a murder and there’s blood left at the scene you don’t necessarily have to know whose blood it is to know it’s not mine if I’m the suspect. I just need to be excluded. Yes, to then catch the person who left the blood at the scene you might want that.
Corey: That’s right. Often having a clear perpetrator can take the heat off someone who might be a suspect, right? If there is say, blood at the scene, and it’s not mine you don’t know whether I’ve got a collaborator or something like that. But if you can trace it back to me and you find out I’ve actually been accused of these other things also, that definitely takes the focus off the person you may be thinking about. It just gives you more data, right?
Barbara: Right, it does. I’m just saying I don’t think it’s quite at stark. I think your camera issues, there was an officer in Baltimore who turned off his body camera to go put drugs somewhere and then turned it on and said, oh look I found these drugs right where the defendant was. But he didn’t realize that it kept recording.
Corey: Oh, really?
Barbara: Yeah, he wasn’t actually hiding what he thought he was hiding. So it was exactly the thing where defendants say hey, he planted the drugs, and you say come on, he didn’t plant the drugs. Well, actually we have it on camera, he did. Cellphone data is another one I think you had mentioned before that I know anywhere I go someone could figure it out if they had access to my phone because I have my phone with me all the time. But these things are happening anyway, right? The surveillance, in some parts of the world anywhere you go you’re going to be on CCTV.
Corey: A friend of mine in Beijing claims you can’t commit a crime in Beijing and get away with it for more than 24 hours.
Barbara: But a lot of this is due to private actors, right? This is a bigger question than any sort of criminal justice reform can answer. Or just a question of do we avail ourselves to this?
Corey: Have you seen any benefit from this increased amount of data in exoneration people? I’d say an increased number of people exonerated through cellphone data or even cellphone footage or CCTV. Or even increased amount of data available in DNA databases. We’re now seeing people convicted through DNA databases but I’m going to assume you’re going to start seeing a few cases of people being exonerated once you actually track the person down.
Barbara: That’s right. I think it’s going to take a few years for us to start to see that because exonerations take a long time. Also, as the technology is available then it’s available at the trial level and central to our definition of exoneration, for most of our exonerations unless it’s a pardon based on innocence most of our exonerations are some new evidence that wasn’t presented at the original time of trial. I think part of the problem with some of these technological advances is they’re maybe not as solid as we think they are, so the cellphone tower ping, your cellphone pings a certain tower and people think that means you had to have been there. But actually it’s much more complicated than that, it can ping multiple towers so it’s not as probative as we thought it was.
Corey: All this seems to be changing as bandwidth increases it seems like you’ll probably get more granular data on people’s locations.
Corey: My personal theory is that over time not only you have cellphone data you’re going to actually have wearable clothing to be able to tell not only where someone is but what they’re doing at that point in time. So they’ll know from my synthetic wifi connected shirt whether I’m running very fast or I’m moving. Once you get data at that granular level that may be a boon. It’s essentially accelerometers on your iPhone.
Barbara: It’s incredible how much data I have allowed about myself. It’s like, well I don’t care if somebody knows how many steps I took today.
Corey: My guess is for defendants, right, if there’s a class of people who don’t have access to this technology now they may have access to it in the future even at their lower socioeconomic level. And thus we’ll have data of people who are more likely to be accused than you or I of these crimes. And that may be potentially exonerating.
Barbara: Yes. On the other hand, I was just reading a case today that we’re about to post where a person was convicted of a murder from the mid 1980s, I think it was 1986. And he just got exonerated this month and he had four or five alibi witnesses that he was in a different city at the time. But the jury didn’t believe it, they believed the person who said, he was my cellmate and he was in jail on another charge and he confessed to me.
Corey: But presumably technology would override, one would hope.
Barbara: I don’t know, I’ve seen enough cases, and it’s not always the case, every case is a little different and every jury is a little different. But I’ve seen enough cases where people had multiple fairly disinterested alibi witnesses who could corroborate the time they were with the defendant because they went to a store together and they have the receipt from the time that they were there and what they bought. Yet somehow they get convicted, so it does happen. I think that in the case of the central park five which is an interesting case for many, many reasons.
Corey: I think most people are generally aware of it but can you give us the background?
Barbara: Over time my students have been less so. But now with the movie they tend to know it. There was a jogger in New York in the late 1980s, I think it was 1989, who was viciously attacked and raped and beaten and there were five young boys. I think they ranged in age from maybe 15 to 17, but pretty young, who had been in the park that night and were taken in for questioning and most of them gave statements that they had participated in the rape and very vicious beating of this woman. Before trial, the DNA was tested and it excluded them. It was one person but they still got convicted. They still got convicted on the basis of their confessions and it wasn’t until, I guess it goes back to the point you were making before, actually somebody came forward and said I’m the one who committed this crime, and it was his DNA, that they were exonerated. Even in cases where you think that something is pretty dispositive, at the very least reasonable doubt doesn’t always work.
Corey: I think a lot of factors played into this one. Again, high profile case police thought they had to solve and they had potential perpetrators who were not very high value socially.
Barbara: That’s absolutely right. I remember because I grew up in upstate New York and it’s a pretty different world from New York City. But we’d get the papers and the media from there, and that case was just the biggest story, it was such a horrific attack on this woman, nobody expected her to live or recover, she did thankfully. And you had confessions, right? You had confessions from these boys. We know more now, we know how somebody might confess to something they didn’t do and we know that the kinds of people who are likely to confess to something that they didn’t do, a big factor is youth. They were young, maybe as young as 14.
Corey: I think that’s about right.
Corey: I think that’s one of the first forays of Donald Trump into politics, when he took a two page ad out in the New York times.
Barbara: There are some players from that who are absolutely convinced that, okay, Matias Reyes the person who confessed to doing it and it was his DNA, they think well they must have participated as well. Even though their statements didn’t really add up together.
Corey: They didn’t know where she was, actually. They had no idea where the crime was committed.
Barbara: And the drag marks. It’s horrible but the things that you have to think about in assessing the credibility of a confession. But just the way the vegetation was at the time it didn’t look like a group of six boys. It would have had to have been six people with him, or at least five if he came along later, which is one of the theories. That the boys attacked her and then this guy just kind of stumbled across her then attacked her again.
Corey: This case although it doesn’t fit the technical definition of a group conviction, feels like some of the other cases discussed that have racially biased angles. We’ve seen cases over the past thirty years in the case of drug deals where a larger number of people are convicted of drug dealing and they’re almost always black or latino. I’m not really aware recently of large numbers of white kids being convicted falsely of rape, so this feels very similar to these kinds of drug cases. There’s a couple large cases, there’s one in Texas where an informant got about 39 people convicted and there are similar cases in LA.
Barbara: And Baltimore, Philadelphia, there are some group exonerations. Glad you brought that up. So the national registry of exonerations has thus far focused on individual exonerations and part of our definition of an exoneration is it requires some sort of individualized revisiting of the evidence. The problems with these mass exonerations is that it’s often sort of a domino falls and then a lot of other dominoes fall. I think about the Massachusetts cases where the lab tech was dry labbing which means that they said they were doing the tests on the drugs but they weren’t and then huge swathes of convictions are vacated. Those aren’t an individual revisiting of the evidence in a particular person’s case so we’ve never really known quite how to fit them in. So they’re not in the registry but we are launching a group exoneration part of the website, hopefully in the spring or early summer to look at these. Because individual exonerations are really focused on the individual defendant or a few defendants where the facts are really revisited.
Barbara: But we also care about these more systemic corruption cases that lead to the exoneration of a number of defendants, and there the focus of inquiry is usually on some official actor. Usually a police officer who was discovered to have been corrupt, planting evidence or fabricating lab results or something like that. We are going to be devoting some resources to that in the near future.
Corey: At this stage do you have any estimate on the magnitude of this problem? Is it far smaller, if you take some of these group exonerations?
Barbara: Some might be a few dozen defendants. I don’t know off the top of my head how many were in Texas, which is what I think you were just referring to. Sometimes we don’t even have an accurate count of the number of defendants whose cases were cleared. I think Ramparts is a good example. In Philadelphia there have been, I believe over 1000 convictions have been vacated based on the discrediting of a particular arresting officer. Many thousands in Massachusetts based on the lab scandals there, there’s actually two separate lab scandals. One was a case of drylabbing.
Corey: So dry labbing, it wasn’t labbing at all.
Barbara: No, nope. It’s just faking it. Then another was a lab analyst who was dipping into the samples that she was supposed to be testing. There often very different in nature. The process of exoneration is, it’s really hard to be exonerated, the whole presumption of innocence you get before you’re convicted, it quite understandably flips once you’ve been convicted. So now once you’ve been convicted there aren’t really good mechanisms to revisit the facts. The apelut process is about correcting the process error or legal error but courts aren’t really interested in revisiting facts so if you discover new evidence, either because the state withheld it or your attorney was ineffective, or just something that wasn’t anybody’s fault, it just wasn’t available to you then. There’s not a lot of mechanisms to get back into court and then there’s a tremendous hurdle that the defendant has to overcome to get relief it’s really high. Which is why we feel that exonerations are a really good proxy for innocence.
Barbara: Of course we know, there have got to be some people in the registry that are actually guilty. Just like we know there are lots of people who are not in the registry who are actually innocent. With group exonerations it’s even harder because if they throw out a bunch of drug convictions because they never tested the drugs, some of those people did have drugs. We just don’t know who.
Corey: I believe in fact there’s actually a new case coming out of New York state because they found the algorithms used to basically used to de tangle different people’s DNA. There’s a new algorithm developed to basically tell different people’s DNA from DNA samples. But it looks like the algorithm may not actually be as accurate as people thought. This is a thing that’s been developing over the past, I think it started about a year ago or so when I first read the papers about this. There’s actually proprietary technology built by the company, it was a black box and the investigators using this system without any idea how it actually functions. So they’re now revisiting those cases and throwing many of them out.
Barbara: I would imagine that in those cases if there’s DNA involved they tend to be more serious cases and so there would probably be an individualized inquiry.
Barbara: I don’t know.
Corey: I think they’re reopening the case, they’re not throwing them out entirely.
Barbara: Right. So assuming the material still exists you can retest it using accepted methods.
Corey: I think we’re about out of time.
Corey: But thanks for coming by.
Barbara: Thanks for having me.
Corey: This has been a really interesting discussion.