MS. DIANE REHMThanks for joining us. I'm Diane Rehm. In 1980, Cornelius Dupree was sentenced to 75 years for aggravated robbery, abduction and rape. He was released from a Texas prison last July on parole after 30 years behind bars. Less than two weeks later, DNA test results proved his innocence. Joining me in the studio to talk about the causes of wrongful convictions and the benefits and pitfalls of forensic DNA testing, Shawn Armbrust. She's with the Innocence Project. Good morning to you, Shawn.

MS. SHAWN ARMBRUSTGood morning, Diane. Thanks for having me.

REHMProf. William Thompson joins us from the studios at the University of California at Irvine. Good morning, sir.

PROF. WILLIAM THOMPSONGood morning, Diane.

REHMAnd by phone, John Bradley. He's district attorney of Williamson County in Texas. Good morning, sir.

MR. JOHN BRADLEYGood morning, Diane.

REHMThanks for joining us. And if you'd like to join as well, call us on 800-433-8850. We've already had numerous messages posted on Facebook. If you'd like to join us there or send us a tweet, feel free. Shawn Armbrust, let me start with you. Tell us about this latest DNA exoneration in Texas, why it came after Mr. Dupree had been paroled.

ARMBRUSTWell, this wasn't my case, but my understanding is that it was in the works for four years. But sometimes in these cases, it just takes forever to find the evidence. You know, if you can imagine, a lot of these urban evidence warehouses look like the last scene from "Raiders of the Lost Ark," you know? These warehouses that are full of evidence that -- you know, it's kind of like your basement. There's no real reason to clean it out. There's no real reason to keep it organized, and it's tough to find. And that's starting to change, but it can be very time-consuming. And I think some of that happened with Mr. Dupree's case.

REHMJohn Bradley, what are the most common reasons for wrongful conviction?

BRADLEYWell, I think, statistically, they have shown that it is the misidentification of an individual from an eyewitness.

REHMSo that -- haven't I read somewhere that eyewitness identification is the least reliable of all causes of conviction?

BRADLEYWell, I've certainly heard some statements about that. I don't think that there is sufficient research yet to say that with a lot of certainty. But I think they also should put those things in context. If you look at the total number of exonerations and compare it to the number of cases that get prosecuted successfully nationally, those are actually very, very small outcomes.

REHMAnd, Prof. Thompson, what about invalidated or improper forensic evidence? How often does that happen?

THOMPSONWell, that's been another major factor in the exoneration cases that we know about. Those cases have opened up a marvelous window of opportunity for academics, like myself, to examine the system and see what can go wrong. And as mentioned, the most common error seems to be in eyewitness misidentification. But problems with forensic science is close behind. There have also been problems with false confessions, dishonest informants, and police and prosecutorial misconduct factor in also.

REHMWhat about those false confessions? Where do they come from?

THOMPSONWell, people who study false confessions say that, in most cases, people, after lengthy interrogation, are led to believe that confessing falsely is their best option, that a false confession is their only way out, so to speak, and that they all receive leniency from the false confession.

REHMIsn't there also some indication that those who make false confessions are 18 years old or even younger, and some are developmentally disabled?

THOMPSONThere -- yeah, disproportionately false confessions seemed to be found in young people and people with limited mental abilities, but not exclusively. The Norfolk Four case is an interesting example of where some Navy personnel, after lengthy interrogations, were convinced to confess to crimes that they clearly did not commit.

REHMHmm. When did DNA testing actually become available for use in criminal trials?

THOMPSONThe first criminal cases in the United States involving DNA testing were in the late 1980s -- 1987 and 1988. DNA testing started to be used more widely to retest post-conviction cases in the early '90s, and the -- it was spearheaded by the Innocence Project in New York, and particularly by Barry Scheck and Peter Neufeld, the directors of that project. So they -- we should -- they deserve thanks for having started what is actually a movement now to reexamine a lot of cases in criminal justice.

REHMWilliam Thompson, he's professor of criminology, law and societal psychology and social behavior at the University of California at Irvine. John Bradley is district attorney of Williamson County, Texas. Shawn Armbrust is executive director of The Mid-Atlantic Innocence Project. Do join us, 800-433-8850. Send us your e-mail, to drshow@wamu.org. Shawn, how commonplace do you believe DNA convictions or exonerations could be?

ARMBRUSTWell, I think, you know, the total number of DNA exonerations, as John said, is 265 to date. And that may seem like a small number when you factor in all of the criminal convictions we have, but it's not that small when you look closer at the DNA exonerations themselves. You know, to have a DNA exoneration, you have to have a case where DNA actually can prove innocence.

REHMRight.

ARMBRUSTAnd I know, you know, we all watch "CSI." We want DNA to be there in every case, but it's not. So it's mostly rape and sexual assault cases, which are a really tiny percentage of criminal conviction. I think it's 1 to 1.5 percent. And then from there, the evidence actually has to be available. And we find that it's destroyed in about 75 percent of all the cases where we actually attempt to do DNA testing. So that 265 number really is the tip of the iceberg. And the problems we see in the cases, in those 265 cases, their problems were seen in other cases, too. We just don't have DNA to prove it.

REHMJohn Bradley, what about -- how many states actually require the preservation of DNA? Do we know that?

BRADLEYI don't know what the statistical number of states that have requirements for preservation of biological material. Texas has a statute that requires the preservation of biological material for quite a long period of time. But most of these cases -- in fact, almost all of these cases that we're talking about are cases that were prosecuted and evidence was collected before DNA testing was standard in the criminal justice system. This case from Texas that you mentioned at the top of your show, for example, was back in the 1970s when people were not even thinking in these terms, so it's actually a credit to Texas that the evidence was preserved and still available for testing. What I think you will see happen is there will be a diminishing number of cases that you could even talk about because all of this testing is now being done before trial.

REHMShawn.

ARMBRUSTWell, I think that's why it's important to sort of take these cases and learn from them now, so that we can apply the lessons because, you know, they are -- they do tend to be older prosecutions. But some of the things we see in those prosecutions are things we're still seeing today. For example, some of the eyewitness procedures that can sometimes lead to misidentification, we still see them in a lot of cases today. So it really is important to take these cases and kind of use this window of opportunity to try to make this system more accurate.

REHMBill Thompson, then the question becomes if you do have that DNA material available, to what extent are prosecutors willing to take a look at it after a conviction has taken place?

THOMPSONWell, there's been resistance to post-conviction testing in a few places. In my experience, prosecutors, by and large, are trusting in DNA evidence and are willing to take a look at it if they're convinced that the item being tested is probative for the case. There's often an argument about whether a particular item from the crime scene that fails to match the inmate is really exculpatory or not. And, of course, that has to be argued out by people like Shawn to convince a court that the exculpatory evidence really is exculpatory. In my experience, the system has been quite skeptical about DNA evidence in the cases, like that of Mr. Dupree, where there's an exoneration -- are proved -- where the person is proved innocent six ways from Sunday. And when that happens, most prosecutors acknowledge the truth and get behind the process.

REHMWhat about bad labs, Shawn?

ARMBRUSTWell, I think -- I mean, that has been something we've seen as a problem. You know, one of the things DNA has taught us is that sometimes forensic science doesn't work as well as we want it to. And, you know, while DNA has mostly taught us about things like hair analysis and bite mark analysis, it also has taught us that we should be skeptical about everything and sort of be very careful about what we use.

REHMShawn Armbrust, she is executive director of The Mid-Atlantic Innocence Project. Short break. We'll be right back.

REHMAnd just before the break, as we talk about those who have been wrongfully convicted and then found innocent by use of DNA evidence, we were talking about some of the problems specific to DNA testing. Bill Thompson, what about Houston? What happened to the Houston Police crime lab?

THOMPSONWell, there was an enormous scandal in Houston. It turned out that lab had been using inadequate scientific procedures, misrepresenting its findings and, in some cases, hiding exculpatory evidence, and that these practices had gone on for well over a decade before they were exposed in around 2002 and 2003 by a journalistic investigation in which I participated a bit. I think the problem in Houston was not just a breakdown in the management of the crime lab. I think it was a breakdown in the entire criminal justice system. This bad evidence was being produced by the lab and routinely introduced in court cases, and it wasn't being detected as bad evidence by the prosecutors or defense lawyers. And it resulted in some false convictions as well. One of the false convictions was based on bad DNA evidence, ironically. So...

REHMWhat do you mean bad DNA evidence?

THOMPSONWell, in that case, the DNA tests were misinterpreted by the analysts. There was -- you know, when I looked at the tests four years later, it was clear that there was an interpretation of the results under which the defendant was, in fact, innocent, that the analysts had either missed or ignored. The analysts were also routinely overstating the statistical significance of their findings.

REHMAnd isn't it true that one FBI analyst actually dumped a portion of the samples that might have contained contaminating DNA before sending the samples through that computer-generated genetic analyzer that typed the DNA?

THOMPSONYes. That's been another problem that comes up from time to time in labs, is the labs are supposed to run control samples to detect problems like contamination. What was happening at the FBI lab -- my interpretation was that the analysts were under pressure to produce results quickly. And one of the analysts, when she was finding contamination in her work, rather than reporting it accurately, simply dumped those samples and proceeded. The same thing was happening in Houston. They dealt with their contamination problem by simply not running the controls that were designed to detect contamination.

THOMPSONAnd so corner-cutting in laboratories and uneven quality of work in labs has been a continuing problem, and it's a problem that's cropped up in a number of the DNA exoneration cases. And so it's -- you know, in addition to looking at eyewitness identification, we need to look at better oversight of forensic science as a possible solution to these problems.

REHMShawn, do you have any idea -- do we have any idea how many people have actually gone to their death, that could have been exonerated by DNA evidence?

ARMBRUSTWe don't know how many have. We know that there's -- there have been a couple of cases where they've done posthumous DNA testing -- at least one case in Virginia where that happened, and it did prove that the person was guilty. There have been -- there's been one high-profile case in Texas where arson evidence has shown that an innocent person, potentially, was executed there. So it's not DNA evidence, but it is scientific evidence.

REHMShawn Armbrust...

ATTY. JOHN BRADLEYI would have to disagree with that last one. I don't think that that's a conclusion that has been reached. There was substantial independent evidence in that case, including the defendant's confession that showed his guilt.

REHMJohn Bradley, he is district attorney at Williamson County, Texas. Joining us now by phone from his office in Richmond, Va., is Virginia Atty. Gen. Ken Cuccinelli. Thanks for joining us, sir.

ATTY. GEN. KEN CUCCINELLIGood morning.

REHMTell me where your office stands on the use of DNA testing for purposes of clarification or even exoneration.

CUCCINELLIWell, we have -- in Virginia, we had a forensic tech who, apparently, had hoarded a good bit of biological evidence and -- over the years. And it surfaced at some point while I was in the State Senate -- which is where I was before I became attorney general -- and there was some discussion about whether or not that should be tested and be made available, should the state fight it. I, certainly, have always taken the position, be it in the general assembly or now as attorney general, that if we have available information that may shed light on the accuracy of our criminal justice process, that we ought to use it. And it doesn't -- you know, it doesn't dictate outcomes, but there's no reason not to do that.

CUCCINELLIAnd I have plenty of experience with some of our prosecutors who believe that once a case is closed, you ought to resist reopening it, much less -- and, obviously, what I'm talking about is -- would be the equivalent of reopening the case. So we're very open to that and have a great concern, frankly, just to be correct. And one of the first votes I had in the State Senate years ago, breaking with my party, was to break a tie on a writ of actual innocence where we expanded the available opportunities to prove one self innocent after a conviction. So this is an area that I've spent a little personal time and political capital to protect and feel very strongly about it.

REHMThere has been an effort in some states to establish, "a set of best practices" in regard to DNA testing. What about in Virginia, Mr. Cuccinelli?

CUCCINELLIWe have moved in that direction as well and, frankly, expanded not merely on a protective basis, but on what I would call an aggressive basis the use of DNA. For instance, Virginia pioneered the use of DNA data bank among our convicted felons, which we have used to break otherwise cold cases in Virginia. And it has given us some convictions -- frankly, helped us solve cases that we were at a dead-end on. Other states have quickly followed that lead. Part of that effort entails establishing a methodology to manage and utilize your scientific evidence, your DNA evidence, and we've done that in parallel. And I would say that we have tweaked that since we began. I don't -- I'm not here to say that we got it right, right off the bat.

CUCCINELLIBut I think we've improved our system rather dramatically over the years and have been in a position to make use of that evidence, both for some exonerations and for some convictions that we otherwise wouldn't have gotten.

REHMWe've had an e-mail from Denise, who lives in Sterling, Va., who says, "Many people don't realize Virginia has one of the strictest rules for entering new evidence once a conviction has been made. For example, there's been a three-week rule where no new evidence can be admitted after conviction. That's been changed somewhat to account for advances in DNA testing, but I believe it is still one of the most restrictive states." She goes on to say, "There was a famous case a few years ago where the state refused to do a DNA test on a man on death row. The man was put to death before DNA testing was done. The test did not prove to be exculpatory, but the intransigence of the Virginia justice system was certainly highlighted." Your reaction, Mr. Cuccinelli.

CUCCINELLIWell, we -- that's called the 21-day rule. She is correct in characterizing that as one of the toughest finality rules in the country. But beyond the 21-day period, we have several opportunities for writs of actual innocence -- one, basically biological evidence-based and non-biological evidence-based, and the burden and the -- just taking the non-biological. Just so your listeners know. The burden there is to establish on the -- by the defendant and now convict, by clearing convincing evidence.

CUCCINELLIThey have to bring forward new evidence that wasn't otherwise available, that no rational trier of fact, a juror, could have found proof beyond reasonable -- proof of guilt beyond reasonable doubt in the underlying case in light of the new evidence. That is after the 21-day periods, and that's the non-biological evidence that -- we have an equivalent statute for biological evidence, which was the vote I was telling you about. That was the bill in the courts that we had to -- took some doing to get out, but it was...

REHMI'll bet. I'll bet.

CUCCINELLIYou know, it was a good bipartisan effort and opinion.

REHMSure.

CUCCINELLIAnd there are a lot of us very committed to the process being as close to dead-on accurate as it can be. It's always made up of human beings, and we're always aware of the potential for error to creep in. It's where part of my caution comes from. You know, you don't hear me talking along the lines of what your e-mailer described as being someone who had blocked even the attempt to review DNA evidence. That is not an approach that I support as attorney general, and -- especially in the death row case. There -- we can correct most any mistakes, except that one.

REHMAll right. And what is your reaction, finally, to the comments of Supreme Court Justice Antonin Scalia, who dismissed the idea that an innocent person may ever have been executed in America? He wrote that in a 2005 case, Kansas v. Marsh.

CUCCINELLIWell, I haven't read that particular case. I've read plenty of others by Justice Scalia, but not that particular one. And my general impression -- and, obviously, I have a bit of a Virginia bias. I'll concede that. But I've been through our statutes rather thoroughly. As a member of the Courts Committee, we deal with that penalty issues every single year. And now, as attorney general, we're involved in every single appeal in Virginia. I believe that we have established about as good a process as you can in Virginia. And I will never claim perfection, but I think that, in terms of having a reliable consistent process that won't get you wild-eyed results, we've done that in Virginia.

CUCCINELLII would be very -- I would not make the kind of a statement along the lines that there's never been a mistake in a death penalty case where someone's been executed. I would, frankly, be rather surprised if that were true, just as a statistical matter. I was an engineer before I went to the dark side and went to law school, and I still have that sort of left brain statistical numbers approach. And I just have a hard time imagining that, through all the different states and the federal system, that there hasn't been a mistake made in that kind of a case. You know, one of the things that scientific evidence, forensic evidence does for us is it does provide the opportunity for greater certainty in terms of some of our conclusions.

CUCCINELLITake the other end of the spectrum. You know, witness identification, famously inconsistent in terms of its reliability. That doesn't mean in every case. The problem is that, in some cases, the witnesses can be extremely reliable, accurate, dead-on and in other cases, not. But as a category of evidence, it is one you can look at and say, hmm, this is -- how much can I trust conclusions that rest solely on this form of evidence? And the forensic advances that have been made in the last 10, 20, 30, 40 years have really enabled us, I think, in some respects, to be more confident in the certainty of some of our conclusions.

CUCCINELLIThat's not -- doesn't appear in every case. You know, it's not available in every case, but where it is available, it certainly allows us to draw stronger conclusions. And where it's available and wasn't used and suddenly becomes available, I certainly would not expect Virginia, under my leadership -- as long as I have a say in it -- to be blocking people's ability to test that evidence.

REHMVirginia Atty. Gen. Ken Cuccinelli, thank you so much for joining us, sir.

CUCCINELLIAnd thank you for having me. You have a great day and Happy New Year.

REHMHappy New Year to you. And you're listening to the "The Diane Rehm Show." Shawn, your comments.

ARMBRUSTWell, first of all, I agree with the attorney general, that Virginia really has made some wonderful strides. They have eliminated -- they haven't eliminated the 21-day rule, but they've relaxed it. And, you know, I've had pleasant experiences working with his office on DNA cases. They help us expedite them, and that's been great. On the issue of executing someone innocent, you know, I think I would have been inclined to believe, 14 years ago, that it was impossible to execute someone innocent in the United States because there are so many layers. But that's before I started doing this work. I was a 21-year-old journalism student, and I signed up to take a class where we reinvestigated old murder cases.

ARMBRUSTAnd the case I signed up to work on was a guy who had been on death row for 17 years, had lost all of his appeals and had been saved at the last second, 50 hours before his scheduled execution, not because of concerns about innocence, but because his IQ had been tested at 51. Innocence wasn't an issue, but it turned out he was innocent. And that had never been able to make its way through the courts. And so, I think, people look at all of these procedures. And they see them, and they think, oh, how could we miss? But a lot of the times, the courts aren't really considering factual issues. They never get to the substance.

REHMHow was it determined that he was innocent?

ARMBRUSTThe real killer confessed on video.

REHMWow.

ARMBRUSTYeah.

REHMSo he could have gone to his death.

ARMBRUSTHe had been fitted for his coffin. His burial suit had been ordered. He had ordered his last meal. He was about as close as you get. And it wasn't innocence that saved his life.

REHMAnd someone who was already imprisoned confessed?

ARMBRUSTNo. No. It was someone who was out. We did an investigation, and we tracked down the person who had been with the real perpetrator when he committed the crime. And she had been afraid to talk before, and she talked to us and sort of -- once he knew that she had sort of put him there, he confessed.

ARMBRUSTShawn Armbrust, she is executive director of The Mid-Atlantic Innocence Project. We're going to take a short break. When we come back, we'll open the phones, 800-433-8850.

REHMAnd we have our first e-mail. This is from Henry who says, "In the year 2000, Frank Lee Smith died in prison after spending 14 years on death row for a crime DNA testing later showed he did not commit. Gov. Jeb Bush was vehemently against DNA testing. His response was, "Things happen." Jerry Frank Townsend spent 28 years in prison, also in Florida, for a crime DNA testing later showed he did not commit. Our justice system is sick." William Thompson, do you want to comment?

THOMPSONWell, I agree there are things that can be improved about our justice system. And, I think, we ought to take this opportunity, having seen these cases through DNA testing, to talk about what we can do to make the justice system work better. We know, for example, that eyewitnesses often make mistakes, but research studies have shown there are ways to improve eyewitness identification. It will never be perfect, but we should use the procedures that make it as accurate as possible. Studies show, for example, that eyewitness identifications from lineups are improved in accuracy if the people conducting the lineups are blind to the identity of the suspect, so they can't inadvertently cue the witness to pick the right person.

THOMPSONIt's also been shown that conducting lineups in a sequential, rather than simultaneous, manner improves accuracy. And so these are changes in procedures that are being used right now that could improve the accuracy of our system, and we should be talking about doing them. With regard to interrogations, we know there are false confessions sometimes. Experts in the area have recommended, as a result, that we should routinely tape record all interrogations of suspects, which would help determine whether the suspect was being pressured or was being inadvertently fed information that would appear to verify the confession. So...

REHMAnd joining us now from his office in Dallas, Texas, is District Attorney Craig Watkins. Good morning, sir. Thanks for joining us.

MR. CRAIG WATKINSHow are you doing? Thanks for having me.

REHMI'm fine. Thanks. Tell us about your office's role in the recent exoneration of Cornelius Dupree.

WATKINSWell, you know, when we initially came in as a district attorney of Dallas County and the new administration back in 2007, we accurately pursued these claims of innocence. In fact, we incorporated within our office a unit. It's called the Conviction Integrity Unit, and the primary responsibility of that unit is to respond to and give a legitimate look at all the claims of innocence. And, unfortunately for our system, the unit has a place. And it's shown that we have flaws in the system, and we've been in the position to exonerate 25 individuals.

REHMTwenty-five. Now, are more district attorney's offices across the country creating these kinds of units?

WATKINSNow, the only other one that I know of is Cy Vance in Manhattan. He's got a unit there, and they are actively pursuing them aggressively, like we have in Dallas County. In fact, I think he's devoted at least five or six prosecutors to the unit here in Dallas because of budgetary issues. We only have two prosecutors, a secretary, and an investigator. And we were fortunate enough to come up and review the cases and have them tested -- of the claims -- and been very successful in our approach. But we would hope that all other DAs adopt the model, you know. The responsibility -- the way we see it -- of a district attorney is to seek justice, not convictions. And...

REHMAnd, indeed, hasn't there been -- haven't there been suggestions for a national commission for prosecutorial standards?

WATKINSThere has, and -- but, unfortunately, it hadn't gone too far.

REHMHmm.

WATKINSThe way I see it, for this to work and for our justice system to regain its credibility, this conversation or this message about the flaws in the system will have to come from district attorneys, elected DAs throughout our country. And we have to police ourselves. And in order for us to make all of this work, we need to do that. And other DAs throughout the country need to adopt some of the policies...

REHMHmm.

WATKINS...and procedures that we put in place in Dallas County. Now, you know, this, in Dallas County, is an elected position. And when you are -- when it's political, you have to deal with the political ramifications of being an elected DA.

REHMSure.

WATKINSAnd for the longest time, the mantra has been to be tough on crime. Well, I think, overall -- if you just look at the overall system and even you couple that with exonerations and just overall, tough on crime had worked for our country. And so, I think, this is an opportunity for elected DAs, from a political standpoint, to gain credibility with their constituents to say why they should be elected. Because not only are they pursuing convictions against the truly guilty, they are also pursuing legitimate claims of innocence. And if you look at this from a political standpoint -- Conservative, Democratic or Republican standpoint -- at the end of the day, I think the average citizen hates to see a person...

REHMSure.

WATKINS...that has been convicted for something they didn't do.

REHMWell, now, let me ask you. You are the first African-American DA in all of Texas.

WATKINSYes.

REHMWhat is your belief about the role of race in these DNA exonerations?

WATKINSWell, I mean, I think -- I don't think it comes down to my opinion. I think it comes down to statistically that most of the individuals that have been exonerated happen to be African-American. So, I think, that's a conversation that we also need to extend...

REHMSo what are the prosecuting attorneys, what's their reaction about what you are doing and the fact that you're pursuing this DNA evidence?

WATKINSWell, it seems, you know, that sometimes that I'm standing on this island alone, especially in the state of Texas.

BRADLEYCraig, this is John Bradley here. I've been listening for a while, and I hope you remember that you have other elected prosecutors in Texas who have been doing many of these same things. Many of those exoneration cases that you handled were already in the works before you took office. They had DNA applications filed, and the DNA testing was being done. And had you not gone into office, the previous district attorney likely would have handled the case the same way. In addition, those statutes that...

WATKINSWell, you know, actually, I take offense to that. I take offense to that personally because, John Bradley, you sent me e-mails criticizing me in the direction that I have gone with this office. And, secondly, if they were being proactive on this issue, why in the world do we have a bank of over 500 cases where people had submitted these applications by the former district attorney for review, and they were denied? So let's just be honest about this...

BRADLEYWell, I think, (unintelligible) as you see is because in an awful lot of these cases, something that has not been discussed in this talk here is in the vast majority of innocence applications, the defendant is lying and is, in fact, guilty. And we have a duty to the public to protect against the release of clearly guilty and dangerous people. And so there are a lot of prosecutors that, I think, take, perhaps, a little more seriously that screening role than you do. You enjoy the national media. You enjoy the attention that you get. We have a lot of prosecutors who don't seek that. They instead seek justice by reviewing these cases carefully and making sure that a guilty, violent person is not released.

WATKINSWell, obviously, it didn't work. From a philosophical standpoint and the fact that you've used that same language that was used against me in this election cycle, that I've gotten national media attention and this is about Craig Watkins personally, gives me indication that you're less concerned about seeking justice and making maneuvers politically. This issue is not a political issue. This is about seeking justice. And if it were, why over -- since 2001 when we had the law passed, why is it that when we first started testing these cases of the 500 individuals that had submitted requests for DNA tests, that the DA's office in Dallas County -- and I'm sure in your county -- has a policy in place that will deny those tests. And of those 500 cases, of those cases that we determined...

BRADLEYWell, you don't know what you're talking about. You don't know what the policy is in my office. You don't know what the (unintelligible)...

REHMOkay. Hold on here, guys. Hold on here. John Bradley, you would like to respond here. So go right ahead.

WATKINSAll those cases that we decided to look at...

REHMHold on. Wait a minute. Go ahead, John Bradley.

BRADLEYThank you. I appreciate that. I don't -- when Craig Watkins was elected, I was one of the first people to call him and congratulate him. And he's done a lot of good things, but he -- like a couple of the people on this show today -- have substantially exaggerated the claims that are being made. Last year in Texas, we prosecuted statewide about a million cases. And exonerations certainly exist, and they're important to pursue. But what we don't talk about is the vast majority of those claims are false. They're made by people in prison who just want to get out. And there are lawyers who will support and argue for those claims because they want a profit from both the attention and from the money that you can get from compensation on exonerations.

REHMOkay. Now, go ahead, Craig Watkins.

WATKINSYou know, I really do believe that's an archaic mentality. At the end of the day, it's about justice and credibility within our justice system. The fact is, is that innocent people have been convicted in the state of Texas in this country. The fact is that there have been people executed for crimes that they didn't commit. And as an elected DA, I'm embarrassed that you would relegate this conversation down to numbers. We're talking about lives. And in the state of Texas and in this country, it's our responsibility to make sure that justice is served. And so when there is a legitimate claim of innocence, it's a responsibility of the district attorneys...

REHMBut is...

WATKINS...to look at those legitimate claims and not go back...

REHMCraig...

WATKINSCan I finish?

REHMSure.

WATKINSAnd not go back and say statistically that some of these claims or most of these claims have been false. Fine. But at the end of the day, if they are false, it's still our responsibility to look at the evidence, to look at the facts of the case and make a determination as to whether or not this person should have their day in court.

REHMAll right.

WATKINSAnd we shouldn't...

REHMAll right. John Bradley, do you want to respond?

BRADLEYYes. What I continue to see is the elevation of the single exoneration over an entire system and the claim that we have a broken criminal justice system because of these individual cases. I think it's irresponsible. I think that it is an attempt to gather attention to one's self rather than put something in context for the public. I think it's important for the public to understand that if there have been 200-something exonerations, that they have to be balanced against the fact that that is over 30, 40 years of prosecution with hundreds of millions of cases that were successfully completed. It doesn't mean that I'm trying to ignore the exoneration, but I also think it's important that we not use it as a miscontextual (sic) way to attack a system that some people just don't like.

REHMJohn Bradley, he is district attorney of Williamson County, Texas. You're listening to "The Diane Rehm Show." Craig Watkins, do you want to respond?

WATKINSObviously, I do. I just think that in today's criminal justice market, I think things have changed. And I think that it's not just about protecting a flawed jury verdict. It's about doing what's right and what's in the best interest of the citizens that we represent. Now, in Dallas County, as a result of implementing our Conviction Integrity Unit, unfortunately, we found all these individuals who had been wrongfully convicted through DNA. But if you limit it to DNA and not other cases, then that tells us that your mentality is still in the archaic age. DNA is just the tip of the iceberg.

WATKINSThere are cases where there is no DNA. And so when you say that there are very few cases that come about where someone has been exonerated for a crime they didn't commit, obviously, because it only deals with DNA. And there is an issue in our system that needs to be addressed. And to have credibility as a prosecutor, when your job is to seek justice, not to protect a flawed verdict but to seek justice, then the responsibility falls on those elected prosecutors to do that. We have to police ourselves.

WATKINSAnd, unfortunately, what we've seen here in Dallas County, because of DNA, there are flaws within the system. And so we take it a step further. We will, in this office, go to Austin and ask our lawmakers to make reforms necessary to protect the innocent and to convict the guilty. And if you think about it from a conservative standpoint, the fact of the matter is, is when we convict someone that didn't commit the crime, the person that actually committed it will still be out there on the street committing more crimes. So the philosophical approach that is being brought by my colleague is flawed.

REHMAll right. Then...

WATKINSWe have to protect the citizens, and doing that is to make sure that we have the right individual convicted for what they did. The fact that this last exoneration's a person that just spent 30 years of his life in prison, and you want to go back and say that, well, over all of the millions of cases that we've dealt with, works, tells me that there is still a lot of changes that need to be made in this system.

REHMAll right. And, John Bradley, I'm going to give you the last word.

BRADLEYWell, thank you, and I appreciate your invitation to appear on this show. I'm sorry that Mr. Watkins seems to misunderstand my position. I'll repeat myself, that I welcomed his election into office and called him and have continued to support his right to act as an elected official in Dallas as he sees appropriate. What, I think, he misrepresents to the country, and certainly in Texas, is that he has some special quality that allows him to know when people are guilty and innocent, and the rest of us don't. He has a very antagonistic attitude about all of this, and it's not helpful. Because every single prosecutor, as he said does, have a duty to seek justice, and we do that. We don't put fancy titles on it.

BRADLEYWe don't go on national TV shows. We don't have our own reality shows to show that we do that every day, and we don't promote ourselves. What we do is we go to our offices. We teach our prosecutors to be cautious and careful about what they do, and when a genuine case of innocence arises, to use the system correctly in identifying it and release the person.

REHMAll right. And that has to be the last word. John Bradley, he's district attorney of Williamson County, Texas. Craig Watkins is Dallas, Texas, district attorney. Thank you both very much. Also, Shawn Armbrust and William Thompson. Very, very interesting discussion. Thanks for listening, all. I'm Diane Rehm.