Lessons Learned From Wrongful Convictions

by Jack E. Call, Professor of Criminal Justice, Radford University, E-mail: jcall@radford.edu

The advent of sophisticated DNA testing technology in the 1990’s resulted in the reexamination of hundreds of cases in which evidence capable of being tested for DNA had been left on a victim or at the crime scene.  As a result of this new technology and other efforts on behalf of persons thought to have been wrongfully convicted, more than 3,000 persons have been exonerated.[1]

               Recently, I taught a 6-class course on wrongful convictions for the Lifelong Learning Institute (LLI) at Virginia Tech.[2]  In this article, I will describe the structure of this LLI course and discuss some of the lessons learned from these resources.  Footnotes throughout the article provide extensive references to useful resources on wrongful convictions

The Spring 2022 Course on Wrongful Convictions

               Students in LLI courses at Virginia Tech are highly-motivated learners, but they also live busy lives, in spite of their ages.  (All LLI members are fifty or older).  Therefore, the challenge was to require them to do a reasonable amount of work, while adequately covering the course topic.  Toward that end, I decided to have them watch videos on Netflix discussing wrongful conviction cases. 

               The Central Park Five.  The first video was a Netflix docudrama on the so-called Central Park 5 entitled “When They See Us.”  The Central Park 5 were five teenagers who were convicted of sexually assaulting and savagely beating a young woman, Trisha Meili, who was jogging in New York City’s Central Park one evening in 1989.  Under intensive police questioning, all five of the accused teenagers admitted to being involved in the assault, although each of them accused someone else of being the actual rapist.  Each confession contained an account of the facts that varied substantially from the account of the facts given in each of the other four confessions.  The accounts varied as to who was directly involved in the assault, where the assault took place, what the victim was wearing at the time of the assault, and the weapon used to carry out the assault.

               Because this program was a docudrama, actors portrayed the defendants, the investigating police officers, the prosecutors, the defense attorneys, family members of the defendants, and key witnesses.  The decision to dramatize critical events in the case raises interesting questions about whether the actions and motivations of people who played a key role in the case are portrayed accurately.  In fact, the key supervisor in the prosecutor’s office, Linda Fairstein (who did not prosecute the case in court), has sued Netflix and the show’s director and co-writer, Ava DuVernay, for defamation of character.[3]

               Eventually, another man, Matias Reyes, who was serving a lengthy prison sentence for other sexual assaults, confessed to prison officials that he was the one who sexually assaulted Ms. Meili.  DNA tests confirmed that Reyes was the only source of DNA evidence found on Ms. Miele and at the crime scene.  This new information prompted an extensive re-investigation of the case, resulting in a request to the courts from well-respected Manhattan District Attorney, Robert Morgenthau, to dismiss the charges against the Central Park 5.

               Students in the class also watched a PBS documentary about the case (produced by Ken Burns, David McMahon, and Sarah Burns[4]).  This documentary is much shorter than “When They See Us.”  It provides a useful companion piece to the Netflix docudrama because it provides the viewer with a journalistic perspective, as well as an opportunity to see the real five Central Park defendants and hear their own perspectives on the case in their own words.  In addition, it gives viewers an opportunity (albeit somewhat limited) for assessing whether “When They See Us” presents an accurate portrayal of the five defendants.[5]

               The case raises serious questions about the manner in which police detectives interrogated the five suspects.  It is clear that the suspects were subjected to extremely intense interrogation that placed them under great pressure.  Four of the five suspects were juveniles.  Psychological assessments of the fifth subject indicated that he had below-average intelligence .  It appears that a couple of detectives may have intentionally persuaded family members of one of the boys to leave the interrogation room so questioning could continue in their absence.  At least one of the detectives ignored information about Matias Reyes that arose early in the investigation (and should have made Reyes a person of great interest before the investigation had proceeded very far).

               “The Innocent Man” Cases.  The class also watched another Netflix production entitled “The Innocent Man.”  The program consists of six episodes, based on the only non-fiction book written by John Grisham (also entitled The Innocent Man).[6]  The program focuses on two murder cases, both of which took place in Ada, Oklahoma.  The first case concerned the murder of a young Ada woman, Debbie Carter, who was raped and killed in 1982 in her small apartment after returning home from an evening in a local bar, The Coachlight.  Ron Williamson and Dennis Fritz were convicted of her murder and sentenced to death and life without parole respectively. 

               Williamson was convicted largely on the basis of the analysis of hair samples found at the crime scene, a “confession” made to a jailhouse informant, and a dream Williamson reported to the police in which he strangled the victim.  The dream was treated as a confession.  Fritz was convicted primarily on the basis of an analysis of hair found on Debbie Carter, testimony of a jailhouse informant who said Fritz had confessed to him (the same jailhouse informant who testified against Williamson), and Fritz’s association with Williamson.  Grisham suggests that the prosecution was largely pursuing a guilt-by-association theory of the case.

               Two years after Debbie Carter’s murder, Denice Haraway disappeared from McAnally’s, the convenience store in Ada where she worked.  Tommy Ward and Karl Fontenot became suspects in the case.  After very persistent police interrogation, they both confessed to killing Denice Haraway.  Their confessions were significantly inconsistent with each other, and neither confession included accurate information about where Denice’s body could be found.  The confessions suggested that her body could be found at places that proved to be inaccurate.  They also indicated that Denice had been stabbed to death.  (Denice’s body was eventually found after the convictions of Ward and Fontenot for her murder.  The body was in a place that neither confession had specified.  Examination of her body suggested that she had been shot in the head – not stabbed).[7]

               Curiously, both men implicated Odell Titsworth as the person who actually killed Denice.  Even the police later concluded that Titsworth could not have been involved in the abduction, sexual assault, and killing of Denice.  A police officer had been involved in a forceful encounter with Titsworth two days before Denice’s disappearance that had resulted in the breaking of Titsworth’s arm.  Titsworth’s doctor indicated that there was no way he could have done the physical acts attributed to him by Ward and Fontenot.[8]

               Ward and Fontenot were convicted of abducting and killing Denice based largely on their confessions and the testimony of two witnesses that someone resembling them was seen near McAnally’s the evening of her disappearance.  There was also testimony from the same jailhouse informant used in the Williamson trial who indicated that she had overheard Ward and Fontenot make incriminating statements.

               At the time of the Williamson and Fritz trials, DNA analysis was not available.  Their lawyers later succeeded in obtaining DNA analysis that resulted in demonstrating that Williamson and Fritz had not raped Debbie Carter.  When the DNA profile was run through a database, it implicated Glen Gore as the rapist.  He had been observed in the parking lot of The Coachlight in an apparent argument with Debbie Carter on the night of Debbie’s death.  Unfortunately, the police failed to pursue this information during  their investigation.  Williamson and Fritz were released from prison after serving eleven years in prison.[9]

               Karl Fontenot’s conviction was overturned by a federal District Court judge in a habeas corpus proceeding.  The court held that the state had violated its constitutional duty to turn over to the defense large amounts of exculpatory evidence that, if heard by a jury, would likely have resulted in a verdict of not guilty.  Fontenot was released on bond in December of 2019 after serving 35 years in prison.  The 10th Circuit Court of Appeals upheld the ruling of the District Court in a divided decision.  The state of Oklahoma has asked the U. S. Supreme Court to overturn the decision of the court of appeals.  As of May 2022, the Supreme Court had not decided whether to hear the case.  If the Court decides not to hear the case, the state could possibly try Fontenot again.

               Tommy Ward’s conviction was also overturned by a state court judge in December 2020.  Similar to Fontenot’s case, the ruling of the state court judge was based on the failure of the prosecution to provide the defense exculpatory evidence that the defense had specifically requested.  The judge also dismissed the charges against Ward on the basis that he could not receive a fair trial at this late date.  The state has appealed that decision, and Ward remains in prison, pending a decision by the Oklahoma Court of Criminal Appeals.  As of May 2022, that decision was still pending.  If the state court judge’s decision is upheld, Ward may not be tried again.

               “The Innocence Files” Cases.  This 9-episode Netflix original series was produced in conjunction with The Innocence Project.  The Innocence Project is dedicated to investigating and overturning convictions of persons that the group believes were wrongfully convicted.  The series tracks the cases of nine wrongfully convicted persons.  The first three episodes focus on the cases of Kennedy Brewer, Levon Brooks, and Keith Harward.  Brooks and Brewer were convicted of killing of killing two young girls in Mississippi.  Harward was convicted of the rape of a woman in Newport News.  All three convictions would not have been possible without testimony from forensic odontologists, who claimed that bite marks on the victims came from the teeth of the defendants.[10]  One odontologist, Dr. Michael West, testified that Brewer and Brooks were the source of marks (that he testified to be bite marks) found on the victim’s bodies “indeed, and without a doubt.”  Forensic odontologists in the Harward case testified that Harward was the source of bite marks found on the rape victim “within a reasonable medical certainty.”  Subsequent DNA analysis determined that Brewer, Brooks, and Harward did not commit these crimes and identified the two men responsible for the three crimes.  Kennedy Brewer spent three years in jail and thirteen years in prison for a crime he did not commit.  Levon Brooks spent sixteen years in prison, while Keith Harward spent thirty-three years in prison.

               Episodes four through six focus on the wrongful convictions of Franky Carrillo and Thomas Haynesworth and the role that mistaken eyewitness identifications play in many wrongful conviction cases.  Carrillo was convicted of killing the father of a boy who was talking with some friends on the lawn of his home in a suburb of Los Angeles.  The father had come out of the house to ask the boys, who were members of a black gang called N-Hood, to be quieter.  A car drove by the house twice.  The second time, a person in the passenger seat fired shots that struck and killed the father.  Six of the group of boys on the lawn identified Carrillo as the shooter.  There were very serious questions about the manner in which the sheriff’s deputies conducted the photo arrays that resulted in Carrillo’s identification.  At a re-trial of Carrillo, the trial judge agreed to visit the crime scene, where a “drive by” of the house was conducted under the same physical conditions that existed on the night of the murder.  Even representatives of the prosecution agreed that it would have been impossible for the boys on the lawn to have gotten a good look at any of the occupants in a car driving by the house.  Carrillo spend sixteen years in prison; his father, who raised Franky alone, died while Carrillo was in prison.

               The Thomas Haynesworth case began with five assaults on white women by a black male within a mile of each other in Richmond, Virginia, during a 4-week period in January and February of 1984. The first victim was Janet, who was attacked by a black male with a knife.  Janet was very conscious of the need to pay close attention to details of her assailant’s appearance, including looking at his face as much as possible and looking for ways to assess his height.  Initially, Janet failed to pick out a picture of Haynesworth in a police photo book.  She eventually picked him out after she had been told that the police had arrested someone for her rape.  Haynesworth was also convicted of raping two of the other victims in the series of rapes, based primarily on identifications of the victims, who were very confident about their identifications.  Eventually, DNA analysis implicated another person as the rapist.  Haynesworth spend twenty-seven years in prison before his exoneration.

               The final three episodes focus on governmental misconduct.  The first case involved the conviction of Chester Hollman in Philadelphia in 1993 for a robbery/murder.  A witness had seen two people flee the crime scene and get in a white Chevrolet Blazer with a license plate beginning with “YZA” and containing two other people.  The witness indicated that the driver of the Blazer was a woman.  Hollman was arrested not long after the crime, driving a white Chevrolet Blazer toward the crime scene with only a single passenger.  The passenger became one of two primary witnesses against Hollman at his trial.  After Hollman’s conviction, the passenger recanted her testimony, saying that the police threatened her with prosecution for murder if she did not testify against Hollman.  The other primary witness against Hollman, Andre Dawkins, later recanted his testimony as well, indicating that he was pressured by the police and threatened with severe action against him on a couple of outstanding warrants.  Dawkins had a lengthy criminal record, not all of which was share with the defense.  There was also a statement signed by Hollman that was introduced at his trial.  It included a statement that he “told the bitch [referring to the passenger in Hollman’s Blazer] to keep her mouth shut.”  The practice of Philadelphia police at the time was not to record interrogations.  Instead, they wrote a summary of an interrogation afterwards and had the suspect sign it.  Hollman indicated that when he signed the statement there was some empty space above his signature and the quoted statement above was not in the statement when he signed it.

               There were serious issues about Hollman’s guilt.  As indicated above, there was significant exculpatory evidence held by the police that was not shared with the defense (as required by the Constitution).  In addition, the police had received a tip early in their investigation that they should investigate a woman named Denise Combs.  The police interviewed Combs and discovered that she had rented a white Chevrolet Blazer the day of the crime from Alamo car rentals.  The license plate on the rental began with “YZA.”  (It turned out Alamo had a whole fleet of white Chevy Blazers with license plates beginning with “YZA,” including the Blazer Hollman had rented).  The information about Combs was never shared with the defense. 

In 2019, the Innocence Project retrieved a box of evidence in the Hollman case that had been preserved.  It was discovered that there were nail clippings from the crime scene that contained DNA material.  DNA analysis revealed that it was not Chester’s.  The Philadelphia District Attorney’s office submitted a statement indicating that the District Attorney thought it was likely that Hollman was innocent.  In July 2019, a judge agreed that Hollman was probably innocent and ordered him released after serving 27 years in prison.[11]

The second case in this third segment of “The Innocence Files” involved the conviction of Dewayne Brown for killing a Houston police officer while fleeing from the robbery of a check cashing business.  A clerk in the store was killed as well.  There were three robbers.  Two of them were caught by the police and accused each other of killing the female clerk, but they both accused Dewayne of being the third robber and of killing the police officer.  The government’s case was based on testimony from one of the robbers, who cut a deal with the prosecution, and testimony from Dewayne’s girlfriend, Ericka, who was threatened with a perjury prosecution (based on her grand jury testimony) if she failed to implicate Dewayne in the murders.

Dwayne’s defense was based on his contention that he had seen a television report about the murders shortly after their occurrence and had called Ericka about what he had seen.  The timing of this call from Ericka’s home phone would have made it impossible for him to be involved in the murders.  However, the phone company said it no longer had records of calls from Ericka’s landline phone.  Ericka refused to confirm at trial Dewayne’s testimony about calling her not long after the crimes had occurred.  Dewayne was convicted and sentenced to death.

Years later, during the appeals process, the phone records materialized.  Breck McDaniel (referred to as the police phone expert) said he was spring cleaning his garage and found a box of documents with “Brown” written on it.  It contained landline records that confirmed that Dewayne had called Ericka from her home the morning of the robbery-murders.  The government was legally required to share that evidence with the defense prior to trial.  The defense also discovered an email from Breck McDaniel to Dan Rizzo, the prosecutor in the case, the day after Ericka’s trial testimony indicating that he had a phone record that did not support Ericka’s testimony that Dewayne had not called her.

The final case concerned Kenneth Wyniemko, convicted in Macomb County, Michigan, of rape and robbery.  The reasons for suspecting Wyniemko intially were his identification at a lineup (Ken says he was the only man in the lineup with a mustache), latex gloves found in Ken’s home (the assailant wore latex gloves), and the fact that the victim and her husband were customers at the bowling alley where Ken worked part-time.  This evidence, evidence from a jailhouse snitch, and testimony from Ken’s ex-girlfriend combined to convict Wyniemko.  There were serious questions about the manner in which the lineup identification was conducted.  There was also evidence suggesting strongly that the police had provided the jailhouse snitch the information he needed to provide evidence against Wyniemko.  In 2002, Wyniemko was released after DNA analysis (unavailable at the time of his trial) excluded Wyniemko as the source of semen found on the victim’s underwear.

At the last minute, the class also decided to watch and discuss a PBS Frontline documentary about the Todd Willingham case in Texas.[12]  Todd Willingham was accused of setting his house on fire while his wife was out shopping.  The fire killed their three young children – one-year-old twin girls and a 2-year-old daughter.  Willingham was convicted largely on the basis of expert testimony by arson investigators suggesting that the fire had been set intentionally.  In 2004, Willingham was executed.  The execution was carried out in spite of an investigative report by Dr. Gerald Hurst, an arson researcher and acknowledged expert on arson, who concluded that the fire at the Willingham house had started accidentally (probably by a faulty electric space heater).[13]

Causes of Wrongful Convictions

               As demonstrated by the cases discussed in the videos above, there are several factors that contribute to wrongful convictions – false confessions, mistaken identifications by witnesses, false testimony at trial (especially from jailhouse informants), government misconduct, and faulty forensic evidence.  The attached table reports the frequency of these factors in four important wrongful conviction databases.  (The percentages total more than a hundred percent because typically more than one factor contributed to a wrongful conviction case).  The great disparity in the percentages can probably be attributed to the subjective nature of determining whether a particular factor was present in a given case and the difference in the cases included in the datasets.  The National Registry of Exonerations is the most thorough database and therefore is probably the most accurate one.

               It should also be noted that even the National Registry of Exonerations does not include all recent cases of wrongful convictions.  The majority of wrongful convictions were established by DNA analysis.  DNA analysis is most helpful in sexual assault cases.  However, in the majority of criminal cases, there is no evidence that can be submitted to DNA analysis.  Therefore, it seems fair to conclude that there are many wrongful convictions that have not been identified.

Possible Remedies to Wrongful Convictions

               It also seems fair to say that most students of the criminal justice system recognize wrongful convictions as a significant problem that demands attention.  The following discussion does not pretend to be an exhaustive list of possible remedies, but it attempts to provide some ideas that should at least be part of the discussion.  Hopefully, at least some of them will be adopted in the near future.

               Faulty forensics.  The problem of faulty forensics was first brought to national attention by a study conducted by the National Research Council and published in 2009.[14]  The National Research Council is the operating arm of the United States National Academies of Sciences, Engineering, and Medicine and is a highly respected scientific group that is tasked by Congress from time to time to investigate matters of great public importance.  This study examined a wide range of forensic practices, concluding that only DNA analysis was based on a firm scientific footing.

               A recent book, Autopsy of a Crime Lab, provides a more current examination of the issues surrounding forensic practices.[15]  Although the book is written by a law professor, it is written for the general public and is “must reading” for anyone concerned about the state of forensic practices in this country.

               The problems associated with forensics practices are varied and complicated.  They do not lend themselves to a few simple solutions.  There are three “big picture” ideas that come readily to mind, however.  First, the basis for forensic practices needs to be studied scientifically.  For example, when Dr. Gerald Hurst began conducting controlled experiments about fires, he discovered that much of the intuitive logic that arson investigators engaged in was flawed.  Academicians should be encouraged to conduct this kind of research. 

Second, most of the forensic specialties have their own professional organizations.  Those organizations should play an important role in encouraging and even sponsoring this needed research.  They need to recognize that there are substantial issues of justice and fairness at stake, and they should assume a leading role in establishing a scientific basis for their specialty.  They should also actively pursue rigorous methods for testing the capabilities of forensic examiners.

Third, courts have been remiss in their responsibility to ensure that only reliable forensic evidence is admitted in criminal trials.  Garrett refers to this as the courts’ “gatekeeping responsibility.”  In Autopsy of a Crime Lab, he has an interesting chapter that details how courts have failed to live up to this responsibility.  The judicial community needs to recognize this failure and engage in a much more rigorous examination of the reliability of many forensic practices.  In many instances, courts should not be permitting the introduction of forensic evidence that is often being submitted to juries for their consideration.

It would also be helpful to eliminate government-operated forensic laboratories.  There is a great deal of evidence to suggest that these labs often operate with a pro-prosecution bias.  Law enforcement agencies could be required to utilize independent laboratories.  Those labs should be audited periodically to ensure that they do not have a bias, in one direction or another.  At a minimum, it should be standard practice that samples sent to a lab by law enforcement should not be labeled with the name of a suspect or otherwise indicate in what manner the prosecution hopes to utilize the results of a sample.

False Confessions.  The public is understandably reluctant to think to think that a person would confess to a crime the person did not commit.  I have read much of the literature on false confessions, and I still experience difficulty in understanding why an innocent person would confess to a crime they did not commit.  However, there is a substantial body of research that establishes beyond serious question that people sometimes confess to crimes they did not commit.  The media could assist here by educating the public about this phenomenon.  Law enforcement officers need to accept this fact as well and be careful not to engage in tactics that may prompt a suspect to confess falsely.  It might also be wise for judges to be more willing to permit defense attorneys to introduce expert testimony about false confessions in cases where the defense is arguing that the defendant’s confession was false.  Perhaps there is a judicial concern that juries will be unduly influenced by such testimony, but such a concern seems misplaced.  At a minimum, in these cases juries should receive some instruction that makes them aware that research has established that it is not uncommon for innocent persons to confess to crimes they did not commit.

At one time, law enforcement officers were nearly always trained to use the so-called Reid technique in their interrogation of suspects.  This technique has been subjected to considerable criticism.  There are newer, less confrontational interrogation techniques that should form the basis for police interrogations.[16]  It would also help if all police interrogations were required to be recorded, even videorecorded.  This would avoid conflicts in testimony about what the police actually did and permit juries to observe directly the manner in which a confession was obtained.

Eyewitness identifications.  The primary concern with eyewitness identifications is that lineups and photo arrays are too frequently conducted in a suggestive manner.  There are constitutional restrictions that apply to these procedures,[17] but the protections they provide are weak.  This issue is significant enough that the U.S. Court of Appeals for the 3rd Circuit empaneled a task force to study it.  Their recommendations included the following:

  1. Make all lineups and photo arrays “double blind.”  Neither the eyewitness nor the person administering the identification procedure should know who the suspect is.  This prevents the administrator from unintentionally “slanting” the procedure in such a way as to encourage the eyewitness to identify a suspect.
  2. All fillers (non-suspects) should fit the description provided by the witness and look similar to the suspect.  There should be at least 5 fillers.
  3. Show-ups (a single suspect presented to the eyewitness for possible identification) should be used only when there is a very good reason not to conduct a lineup.
  4. Instructions for conducting lineups and photo arrays should be written and provided to the witness prior to the identification procedure.
  5. A written, verbatim statement of the witness’s confidence in the accuracy of an identification should be taken down by the administrator immediately following the procedure.
  6. Witnesses should be kept separate from each other during the procedure so one witness may not influence another witness.
  7. Law enforcement agencies should use composites and sketches of a perpetrator rarely and with great caution.
  8. All lineups, photo arrays, and show-ups should be video and audio recorded.
  9. No one should confirm that a witness has selected the police suspect.  Such reinforcement tends to improperly increase the witness’ confidence in their identification.

Even when lineups and photo arrays are conducted properly, eyewitness identifications are notoriously unreliable.  A large body of research has made this conclusion abundantly clear.  At a minimum, juries should be given an instruction to this effect in cases where an eyewitness identification is an important part of the prosecution case.  In cases where the eyewitness identification is a critical part of the prosecution case, judges should permit the defense to introduce expert witnesses to testify about the research on eyewitness identifications.

False testimony by prosecution witnesses.  The primary issue in this category is the use of jailhouse informants.  The frequency with which jailhouse informants give testimony that turns out to be untrue but important to the prosecution case is a significant problem.  Part of the problem is that prosecutors are simply too willing to utilize this kind of testimony.  Sometimes it appears that prosecutors have encouraged these informants to testify falsely (see the following section on government misconduct).  In other instances, it appears that informants have volunteered false testimony in the hope that prosecutors will give them a break in their treatment of the charges against them.[18]  Judges seem all too willing to simply let these informants testify and then leave it to the jury to decide whether the testimony should be believed.  Juries then seem all too willing to believe this testimony.  (And then later developments make it apparent that the testimony of the informant was false). 

Part of the solution may be for judges to hold hearings on the admissibility of such testimony, when requested to do so by the defense, so the judge may determine whether the testimony is reliable enough to permit a jury to hear it.  At a minimum, juries should be instructed that informant testimony should be considered by them with a great deal of skepticism.  There is a legal requirement that the defense be informed of any benefit extended to the informant in return for the testimony (see the discussion of the Brady case below).  The cases indicate that prosecutors too often neglect to provide this information to the defense.  What’s more, these informants often understand the system well enough to know (or at least expect) that they will receive some benefit from their testimony, even when no explicit deal has been struck.

Government misconduct.  The two most likely sources of government misconduct are law enforcement officers and prosecutors.[19]  Sometimes the improper police conduct brings about one of the other factors discussed earlier.  For example, during the interrogation process, police may use impermissible tactics, such as undue pressure, to obtain false confessions.  Or the police may conduct a pretrial identification procedure in a suggestive manner. 

A very significant problem in wrongful conviction cases is the presence of confirmation bias.  An investigator may bring a bias (often an unconscious bias) to a case.  This contributes to the development of a theory of the case that leads to tunnel vision.  The investigator (again, often unconsciously) discounts evidence that is inconsistent with the theory that has developed and gives undue weight to evidence consistent with that theory. 

Improved training can do much to alleviate these problems.  Police departments should recognize that they need to learn from the lessons that the wrongful conviction cases present to them.  Trainers should be chosen who demonstrate a willingness to accept these lessons and pass them on to colleagues, especially younger colleagues who are less likely to have developed bad habits.  Supervisors should keep on eye on their investigators for signs that these habits have creeped into their practices.

Some prosecutors have also developed bad habits, as the wrongful conviction cases reveal.  Prosecutors are part of the problem with jailhouse informants, as discussed earlier.  Some prosecutors seem all too willing to utilize highly questionable testimony from informants.  In extreme cases, it appears that prosecutors (or the police) have pressured informants to give testimony that prosecutors and police know is false.

It also appears that too many prosecutors are not taking seriously their constitutional obligation to turn over to the defense exculpatory evidence in the possession of the government, as required by Brady v. Maryland.[20]  There are at least three aspects to this issue.  Some cases suggest that some police departments are aware of the prosecutor’s obligation and maintain two investigative files – one that they share with the prosecutor and one that they keep to themselves.  The police ”bury” exculpatory evidence in these latter files so the prosecutor is unaware of their existence and therefore cannot turn the evidence over to the defense.  Of course, this tactic does not relieve the government of its obligation to share this exculpatory evidence.  Obviously, police administrators and prosecutors should make it clear to investigators that this practice will not be tolerated.

A second aspect of this issue is that too many prosecutors seem not to take this constitutional obligation as seriously as they should.  Brady only requires the prosecutors share exculpatory evidence with the defense that is “material” to the case.  Some prosecutors take a very narrow view of what “material” means.

One of the reasons that some prosecutors take this narrow view of materiality is related to the third aspect of this issue.  Courts have far too frequently also taken a narrow view of what “material” means.  The reasons for this are not entirely clear.  It may simply be that appellate courts are reluctant to overturn convictions in cases where they believe that evidence of guilt was very strong.  Nevertheless, prosecutors in some jurisdictions know that when they take a narrow view of materiality, there is not much chance that their appellate courts will overturn a conviction on grounds of a Brady violation.  Appellate courts need to take a stricter approach to materiality and be more willing to overturn convictions on this basis.  When these convictions are overturned, defendants can still be re-tried, but prosecutors will get the message that they need to take Brady more seriously.

From these issues of problematic use of jailhouse snitches and lack of adherence to the Brady rule, it is apparent that some prosecutors have lost sight of their primary role.  The American Bar Association’s Standards for the Prosecution Function state that “the primary duty of the prosecutor is to seek justice within the bounds of the law, not merely to convict.”  The Supreme Court recognized this principle in Brady, drawing on its earlier decision in Berger v. U.S., where the Court said that the prosecutor “is the representative not of an ordinary party to a controversy, but of a sovereignty . . . whose interest . . . in a criminal prosecution is not that it shall win a case, but that justice shall be done.”[21]  The solution to this problem may lie, in part at least, with law schools.  Nearly all law schools require courses on criminal law, constitutional rights of criminal defendants, and professional responsibility.  Professors teaching these courses should place heavy emphasis on the legal and ethical responsibility of prosecutors to seek justice first and convictions second.

One last aspect of the role of prosecutors in wrongful convictions bears mentioning.  Some prosecutors have taken note of the increasing recognition of the problem of wrongful convictions and have established conviction integrity units.   These units take a careful look at cases in which serious questions have been raised about the guilt of a person serving a prison sentence.  In some instances, the work of these units has resulted in the prosecutor’s office seeking dismissal of convictions in court.  This is a trend that will hopefully become even more prominent.  It is an important way to fulfill the responsibility of prosecutors to seek justice.  When prosecutors acknowledge errors and seek dismissal of convictions, it sends a strong message to the community that the office of prosecutor is committed to the principle of justice.


               There is no doubt that the public has become increasingly aware that wrongful convictions occur more frequently than most observers of the criminal justice system thought.  It also seems likely that the vast majority of the public would like to see this problem addressed by the system.  The issues discussed in this article are far from comprehensive, but it is hoped that this discussion will stimulate the thinking of thoughtful members of the public and perhaps even prompt some useful changes.  Even incremental change, especially if it persists, would be a move in the right direction.

Frequency of Factors Contributing to Wrongful Convictions




Number of Cases in Source[1]

Percentage of Exonerations Affected by Contributing Factor

Mistaken eyewitness identification

Perjury or false accusation[1]

False confession

Faulty forensic evidence

Government misconduct

National Registry of Exonerations















Garrett, “Judging Innocence”














Garrett – “Convicting the Innocent Redux” (2017)















Innocence Project















Brandon Garrett, “Judging Innocence,” Columbia Law Review, Vol. 108, p. 55 (2008).

Brandon Garrett, “Convicting the Innocent Redux,” in D. Medwed, Ed., Wrongful Convictions and the DNA Revolution: Twenty-Five Years of Freeing the Innocent (Cambridge University Press, 2017)

The National Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/about.aspx

Innocence Project, Emily West and Vanessa Materko, “DNA Exonerations, 1989-2014: Review Of Date And Findings From The First 25 Years,” (2016), http://www.albanylawreview.org/Articles/vol79_3/717%20West%20and%20Meterko.pdf.

[1] Some of these sources continue to add cases to their database.  The numbers in the table were as of February 4, 2022.

[1] Includes informant testimony.

[1] Includes informants only.


[1] See the National Registry of Exonerations (https://www.law.umich.edu/special/exoneration/Pages/about.aspx).

[2] Many institutions of higher education sponsor Lifelong Learning Institutes.  These LLIs offer courses on a broad range of topics for persons over the age of fifty for a reasonable fee.  See https://www.cpe.vt.edu/lifelonglearning/ for information about the LLI at Virginia Tech.

[3] “Ex-Prosecutor Linda Fairstein Can Sue Netflix Over ‘When They See Us’ Portrayal,” https://variety.com/2021/tv/news/linda-fairstein-sue-netflix-ava-duvernay-when-they-see-us-1235037717/.  Fairstein may have a difficult time winning her lawsuit.  She will likely be viewed as a public figure, which will require her to show that the program deliberately misrepresented facts that damaged her reputation or recklessly disregarded the truth of those facts.

[4] Sarah Burns has also written a very informative book about the case (The Central Park Five:  The Untold Story Behind One of New York City’s Most Infamous Crimes, Vintage Books: New York, 2011). 

[5] Other relevant material about the case and its aftermath: “Linda Fairstein, Once Cheered, Faces Storm After ‘When They See Us,’” Elizabeth Harris and Julia Jacobs, New York Times, June 6, 2019 (https://www.nytimes.com/2019/06/06/arts/television/linda-fairstein-when-they-see-us.html); “The True Story of How a City in Fear Brutalized the Central Park Five,” Jim Dwyer, New York Times, May 30, 2019 (https://www.nytimes.com/2019/05/30/arts/television/when-they-see-us-real-story.html); “Elizabeth Lederer, Prosecutor of Central Park Five, Resigns From Columbia,” Julia Jacobs, New York Times, June 12, 2019 (https://www.nytimes.com/2019/06/12/arts/elizabeth-lederer-central-park-five.html); “Wolf Pack:  The Press and the Central Park Jogger,” Lynnell Hancock, Columbia Journalism Review, January/February 2003; “The Central Park Five, Criminal Justice, and Donald Trump,” Jelani Cobb, The New Yorker, (daily comment), April 19, 2019 (https://www.newyorker.com/news/daily-comment/the-central-park-five-criminal-justice-and-donald-trump); “Netflix and Ava Duvernay Win Dismissal of Defamation Suit,” Gene Maddaus, Variety, March 23, 2020 (https://variety.com/2020/tv/news/netflix-duvernay-john-reid-assocaites-defamation-1203542947/); “Fairstein Resignation Sparks Debate on College Culture, Previous Inaction,” Mack Liderman, Jessica Moss, and Frankie Knuckles, The Miscellany News (Vassar College student newspaper), June 6, 2019 (https://miscellanynews.org/2019/06/06/news/fairstein-resignation-sparks-debate-on-college-culture-previous-inaction/); “Ava DuVernay on the Central Park Five Case and Why She Treated Trump as a ‘Footnote,’” Marlow Stern, The Daily Beast, June 3, 2019 (https://www.thedailybeast.com/netflixs-when-they-see-us-ava-duvernay-on-central-park-five-case-and-why-she-treated-trump-as-a-footnote); “A 2002 Report on the Central Park 5 Convictions Being Overturned,” Karen Kreifeld, amNY, June 5, 2019 (https://www.amny.com/news/central-park-five-1-32018864/); “The Central Park Five:  We Were Just Baby Boys,” Aisha Harris, The New York Times, May 30, 2019 (https://www.nytimes.com/2019/05/30/arts/television/when-they-see-us.html); “Netflix’s False Story of the Central Park Five:  Ava DuVernay’s Miniseries Wrongly Portrays Them as Totally Innocent – and Defames Me in the Process,” Linda Fairstein, The Wall Street Journal, June 10, 2019 (https://www.wsj.com/articles/netflixs-false-story-of-the-central-park-five-11560207823); “President Trump Played a Key Role in the Central Park Five Case.  Here’s the Real History Behind When They See Us,” Olivia Waxman, Time Magazine, May 31, 2019 (https://time.com/5597843/central-park-five-trump-history/); “How the ‘Central Park Five’ Changed the History of American Law,” Elizabeth Hinton, The Atlantic, June 2, 2019 (https://www.theatlantic.com/entertainment/archive/2019/06/when-they-see-us-shows-cases-impact-us-policy/590779/); “Affirmation in Response to Motion to Vacate Judgment of Conviction,” written by Nancy Ryan and submitted to the Supreme Court of the State of New York on December 5, 2002 (https://web.williams.edu/Psychology/Faculty/Kassin/files/CPJ.Morganthau.motion.12.5.02.doc); Litigation Team letter to NYC Corporation Counsel stating the basis for its opposition to proposed settlement of the Central Park 5 case (https://s3.documentcloud.org/documents/4609685/Letter-to-Carter-in-Central-Park-jogger-case.pdf).

[6] John Grisham, The Innocent Man:  Murder and Injustice in a Small Town (Doubleday: New York, 2006).

[7] The Denice Haraway case is discussed in great detail in Robert Mayer, The Dreams of Ada (Broadway Books: New York, 2006).

[8] Why would two innocent suspects implicate the same innocent person in this crime?  There is no clear answer to this question.  One possibility, of course, is that the police interrogators suggested Titsworth’s name to Ward and Fontenot.

[9] Williamson had once come within five days of being executed.

[10] The Brooks and Brewer cases figure prominently in The Cadaver King and the Country Dentist: A True Story of Injustice in the American South, Radley Balko and Carrington (New York: Public Affairs, 2018).

[11] The Philadelphia District Attorney, Larry Krasner, was a former public defender, elected to office in 2017.  He established a Conviction Integrity Unit (as have a few other prosecutors in recent years) to investigate cases of possible wrongful convictions.  A PBS show, Independent Lens, devoted eight episodes to the work of the new Philadelphia District Attorney in 2021.  Those episodes can be viewed at no cost by PBS Passport holders.  Most local PBS stations permit access to PBS Passport with a donation of $60 or more.

[12] https://www.pbs.org/video/frontline-death-by-fire/.

[13] An excellent discussion of this case can be found at David Grann, “Trial By Fire,” The New Yorker, Sept. 7, 2009 (https://www.newyorker.com/magazine/2009/09/07/trial-by-fire).


[14] National Research Council, Strengthening Forensic Science in the United States: A Path Forward, (Washington, D.C.: National Academies Press, 2009)>

[15] Brandon L. Garrett, Autopsy of a Crime Lab: Exposing the Flaws in Forensics (Oakland, Cal.: University of California Press, 2021).

[16] See, for example, the PEACE approach to interrogations, utilized frequently in the United Kingdom. (https://www.college.police.uk/app/investigation/investigative-interviewing/investigative-interviewing).

[17] See, for example, U.S. v. Wade, 388 U.S. 218 (1967) and Stovall v. Denno, 388 U.S. 293 (1967).

[18] See some of the examples discussed in Jack Call, “Judicial Control of Jailhouse Snitches,” Legal Note, The Justice System Journal, Vol. 22, No. 1, pp. 73-83 (2001).

[19] There have been some notorious instances of misconduct by forensic specialists as well.  In this heading of government misconduct involving forensic specialists, the issue does not concern the validity of the techniques being utilized, but the intentional misrepresentation of the results of forensic tests to support a prosecution or the misrepresentation of their qualifications.  See, for example, Garrett, Autopsy of a Crime Lab, pp. 101-103 and 140-142.

[20] 373 U.S. 83 (1963).

[21] Berger v. U.S., 295 U.S. 78 (1935), p. 88.