Open PDF in Browser: Russell D. Covey,* Manufacturing False Convictions: Lies and the Corrupt Use of Jailhouse Informants
Through the combined efforts of the innocence bar, conviction integrity units, and an expanding cohort of multidisciplinary scholars, new data about the mechanisms underlying wrongful convictions is rapidly accumulating. Jailhouse informants—jail inmates who testify against fellow prisoners in exchange for charge and sentence reductions and other rewards—have long been identified as significant contributors to false convictions. The new data demonstrates with fresh clarity the extent to which reliance on incentivized jailhouse (and other) witnesses poses severe hazards to the integrity of the criminal legal system. Indeed, the new data reveals overwhelming evidence of the link between the use of jailhouse informants and patterns of corruption and misconduct.
This Study draws on recent exoneration data, taken from the National Registry of Exonerations (“NRE”), consisting of cases in which innocent individuals were wrongfully convicted based, in whole or in part, on false testimony from jailhouse informants. This Study examines these cases to identify commonalities and patterns. Its primary finding, and one confirming prior research, is that there is a powerful link between law enforcement misconduct and the use of jailhouse informants and other incentivized witnesses. The Study also provides further insight into the specific pathways by which this misconduct tends to manifest, revealing how false jailhouse informant testimony is induced, generated, and insulated from close inspection or challenge. Another important finding of the Study is a significant correlation between the incidence of false confessions and false testimony from jailhouse informants. This correlation helps to further build the profile of what likely wrongful convictions look like, and can be used by concerned prosecutors, judges, and innocence lawyers to help identify cases that call for closer scrutiny. It also provides the basis for reform measures to further limit the damage done to the truth‑seeking process through the use of jailhouse informants and other incentivized witnesses.
Introduction
There is a growing understanding that jailhouse informant testimony is deeply unreliable and a common cause of wrongful convictions. As part of a wide‑ranging review of the causes of wrongful convictions in Texas, for example, a Texas commission in 2015 and 2016 conducted hearings probing practices and procedures that have contributed to wrongful convictions and tend to undermine the fairness of the legal process.[1] During the hearings, the commission heard testimony from witnesses who discussed a range of issues related to the use of jailhouse informants, including the reliability of informant testimony, the potential for coercion and manipulation, and the need for greater transparency and oversight.[2] One of the key areas of agreement was the need for a standardized process for handling informants, including requirements for recording and disclosing informant testimony and providing additional safeguards to protect defendants’ rights. Some witnesses also called for increased training and resources for defense attorneys to better challenge informant testimony in court.[3]
The commission’s report and recommendations led to the passage of Texas House Bill 34, which requires prosecutors to disclose specific information about the jailhouse informants a prosecutor plans to use as witnesses. This information includes the informant’s criminal history, the nature of any deals made or promises extended in exchange for the testimony, and prior cases in which the informant provided information for the State.[4] The law also requires that prosecutors keep a record of all deals made with informants[5] and allows for more expansive impeachment of jailhouse informants based on prior charges, which may have been dropped in exchange for a benefit.[6]
Texas is not the only state reassessing legal rules surrounding jailhouse informants.[7] In October 2020, following a probe into a wrongful conviction case in which the conviction was primarily based on the testimony of two jailhouse informants, New Jersey’s attorney general ordered significant changes to the way prosecutors would be permitted to use jailhouse informants.[8] Under the new mandate, prosecutors must now clear the use of a proposed informant with supervisors and maintain a database of approved informants.[9] The New Jersey Supreme Court also made “amendments to discovery requirements” to further increase transparency regarding jailhouse informant usage and testimony.[10] Other states, including Connecticut, Illinois, Maryland, Minnesota, Nebraska, Oklahoma, and Florida, have also instituted tracking requirements and enhanced regulation of jailhouse informants.[11] Connecticut and Illinois also require mandatory pretrial reliability hearings.[12]
The movement among these states and others to raise more substantial guardrails on the use of jailhouse informant testimony is based on an ever‑deepening reservoir of data that underscores the dangers that accompany reliance on such witnesses. Since 2020, more than fifty people have been exonerated after having been convicted, in whole or in part, with testimony provided by jailhouse informants.[13] A close review of this data highlights the many ways in which the practice undermines the reliability of the criminal process.[14] Indeed, these cases make clear that the use of jailhouse informants is a red flag concerning both the quality of evidence and the procedural integrity of the prosecution’s case.[15] The data suggests that cases involving jailhouse informants are highly likely to include false and prejudicial testimony from informants.[16] As the facts from these cases also make clear, there is a persistent correlation between the use of jailhouse informants and misconduct by police, prosecutors, corrections officials, and others—including the informants themselves who take ready advantage of the incentives available to obtain benefits at the expense of innocent people.[17]
This Article reviews these cases to identify, at a granular level, the mechanisms that link jailhouse informant use with wrongful convictions. The data suggest that several of the recent reform measures described above are likely to reduce the incidence of egregious miscarriages of justice but, without a fundamental recalibration of the incentives that undermine the reliability of informant testimony, are also ultimately inadequate to stem the tide of wrongful convictions caused by the use of jailhouse informants.[18]
Part I briefly describes the dataset used in this Article. Part II examines several cases featuring jailhouse informants that exhibit common patterns of corruption and misconduct. These cases suggest a link between jailhouse informant usage and other types of misconduct that flourish within certain “cultures of corruption,” including police, prosecutors, and sometimes more peripheral actors as well. The data studied also indicate that at least two other factors—inadequate legal counsel and high‑profile crimes—encourage police and prosecutors to use unreliable jailhouse informants and thereby produce wrongful convictions.
Part III examines the mechanisms that generate false jailhouse informant testimony and lead to wrongful convictions, including the kinds of deals and rewards used to induce informants to provide false testimony, the disclosure failures that insulate the witnesses from closer scrutiny, the routine assurances provided by jailhouse informants and prosecutors about the reasons for their testimony and its reliability to win jurors’ trust, and the dynamics identified by researchers that help explain why informant testimony is so persuasive to jurors and, when false, can so easily lead to wrongful convictions.
Part IV identifies several common features of false jailhouse informant testimony that should trigger careful scrutiny both of the reliability of information provided by jailhouse informants and the investigative tactics of the police and prosecutors responsible for proffering their testimony. These include witness recantations, serial snitches, and jailhouse informants who emerge at the eleventh hour or as new witnesses whose evidence purportedly bolsters a prosecution case that has already failed to secure a conviction at a prior trial. Part IV also discusses the correlation observed in the data between false confessions and false jailhouse informant testimony. This Article hypothesizes that the presence of a contested confession should trigger heightened concerns about the reliability of any jailhouse informant recruited to testify in that case and argues that use of jailhouse informants should trigger heightened scrutiny of the circumstances and methods used to obtain any purported confession. Given the high incidence of misconduct documented in the data, Part IV also identifies past allegations and findings of misconduct by police and prosecutors as a significant reason to doubt the veracity of a jailhouse informant brought forward by those officials.
Part V identifies policy reforms that might address these findings. Proposed reforms include the creation of police and informant databases, careful documentation of state interaction with informants, and accompanying enhanced discovery of those records. This Article also recommends new restrictions on the use of informant evidence and changes to the post‑conviction review process to facilitate challenges to convictions that are the product of any incentivized testimony, should those witnesses subsequently recant. This is followed by a brief conclusion.
I. The Data
The National Registry of Exonerations (“NRE” or “Registry”) maintains a regularly updated database of cases in which there has been a formal exoneration following a wrongful conviction.[19] According to NRE data, between January 1, 2020, and March 1, 2024, there were fifty‑four exonerations of people who were wrongfully convicted of crimes based on evidence including testimony from one or more jailhouse informants.[20] Among all cases in the Registry, the NRE counts 241 jailhouse informant‑related wrongful convictions leading to exonerations.[21] The fifty‑four post‑2020 jailhouse informant‑related exonerations thus represent more than 20 percent of all such known exonerations, indicating that the rate of discovery of wrongful convictions based on false jailhouse informant testimony is growing.[22] The increase does not necessarily mean that the use of jailhouse informants is causing more wrongful convictions today than it did in past years. Many of the exonerations stem from convictions obtained decades ago. The upsurge in exonerations may simply reflect growing awareness of the unreliability of the evidence used in those older cases, as well as a general increase in the rate of growth of the exoneration database itself.[23]
Of the fifty‑four jailhouse informant exonerations examined here, the NRE has tagged forty‑five cases involving some type of official misconduct—defined by the NRE as cases in which “[p]olice, prosecutors, or other government officials significantly abused their authority or the judicial process in a manner that contributed to the exoneree’s conviction.”[24] In other words, documented misconduct occurred in 83 percent of the wrongful convictions in which jailhouse informants played a role. The data demonstrates a powerful connection between the use of jailhouse informants and misconduct by law enforcement officials. All but four of the defendants in the dataset were convicted of murder, and all but one of the cases involved allegations of some type of violent crime, albeit in five of the cases no crime actually occurred.[25]
An in‑depth review of these cases reveals significant patterns that raise troubling concerns about the continued use of jailhouse informants in criminal cases. This Article examines those patterns to assess whether the measures instituted in reform‑minded states to enhance regulation of State use of jailhouse informants are sufficient to reduce the steady flow of wrongful convictions caused by the practice.
II. Misconduct And Cultures Of Corruption
A. Corrupt Cops and Jailhouse Informants
By all indications, the use of false jailhouse informant testimony rarely happens by innocent mistake. As noted above, official misconduct by police and prosecutors has been found in 83 percent of the post‑2020 jailhouse informant exoneration cases. The high incidence of misconduct in these recent cases is consistent with historical rates. Among exonerations involving jailhouse informant testimony dating back to 1989, known official misconduct was present 76 percent of the time.[26] In contrast, with respect to exonerations stemming from all causes, the NRE flagged official misconduct as a contributing factor in only 60 percent of recorded exonerations. This suggests that while official misconduct of some sort is a prevalent feature of wrongful convictions in general, it is even more prevalent in cases involving jailhouse informants.[27] Indeed, in some instances the same police detectives whose misconduct contributed to the wrongful convictions studied here were found responsible for dozens of other wrongful convictions as well.[28] In short, there is a continuing and well‑documented link between official misconduct by police and prosecutors and the use of false testimony from jailhouse informants.[29]
The more recent exonerations illustrate several types of misconduct that contribute to wrongful convictions. One such type of misconduct—the concerted effort to manufacture false evidence—is present in many of the cases reviewed. Consider, for example, a New York case in which three men, Thomas Malik, Vincent Ellerbe, and James Irons, were exonerated in July 2022 after serving twenty‑six years in prison for a crime they did not commit.[30] The three men had been charged and convicted in New York for second‑degree murder in the 1995 killing of Harry Kaufman, a subway token‑booth employee who died after assailants firebombed the booth in which Kaufman had been working.[31] Not surprisingly, the incident received an enormous amount of media attention.[32] After months of fruitless investigation, New York Detective Louis Scarcella and his partner, Detective Stephen Chmil, took Irons into custody and induced him to falsely confess to attempting to rob the token booth.[33] Irons then (according to the detectives) implicated Malik and Ellerbe.[34] Detectives Scarcella and Chmil arrested Malik and Ellerbe and promptly elicited confessions from both of them as well—confessions that would later also prove false.[35] Weeks later, the detectives identified a jailhouse informant named Shabazz who claimed that Malik had also confessed to him.[36] Shabazz testified at Malik’s trial and Malik was convicted. Irons and Ellerbe were also subsequently convicted in separate trials.
However, the convictions began to unravel after the 2013 exoneration of David Ranta, another man whose wrongful conviction had come to light. Ranta was convicted of a Brooklyn murder that had also been investigated by Detective Scarcella.[37] The New York Times conducted an investigation and published an article shortly after the Ranta exoneration and uncovered substantial misconduct by Detective Scarcella.[38] According to The New York Times reporters, investigators discovered that Scarcella and his partner “removed violent criminals from jail to let them smoke crack cocaine and visit prostitutes in exchange for incriminating Mr. Ranta.”[39] Further investigation uncovered evidence that Scarcella had engaged in a wide range of misconduct, including: “[F]abricating evidence, coercing witnesses, and concealing evidence of defendants’ innocence.”[40] For example, according to The New York Times, Scarcella had used a prostitute and drug addict named Teresa Gomez as a supposed eyewitness in six different cases, several of which seem to have led to wrongful convictions. Ms. Gomez claimed to have been present, and improbably, to have witnessed the actual killing in several of these cases. Lawyers came to refer to her as ‘‘Louie’s go‑to witness.’’[41] A former supervisor of Scarcella speculated that Ms. Gomez was one of several witnesses paid one hundred dollars each time they testified for the State.
The New York Times’ investigation prompted the Brooklyn District Attorney’s Office Conviction Review Unit (“CRU”) to investigate approximately seventy murder cases that were connected to Detective Scarcella, including those of Malik, Irons, and Ellerbe. The CRU ultimately released a report in 2022 finding “substantial official misconduct and new evidence of innocence” in the cases against the three men (among others).[42] Confessions expert Saul Kassin reviewed the confessions of the men and stated that they were “comparable to some of the worst wrongful convictions I have seen.”[43] The CRU reviewed the jailhouse informant Shabazz’s testimony in the Malik trial and concluded that details of Shabazz’s testimony seem to have been obtained from newspaper reports about the case and were not consistent with what Malik had allegedly told the police. All three men’s convictions were ultimately vacated in 2022.[44]
Similar collaboration between corrupt cops and prosecutors occurred in the Maryland wrongful conviction of Keith Davis.[45] The string of events that resulted in Davis’s wrongful conviction began when a security guard at the Pimlico Race Course was shot and killed as he was arriving for work. While investigating an armed robbery of a cab driver that occurred later that morning, police identified Davis as a possible suspect, firing at him thirty‑three times and hitting him with several shots, including in the face.[46] These events took place shortly after another Black man, Freddie Gray, was killed in police custody, sparking national media attention and major street protests in Baltimore.[47] The officers charged Davis for the robbery of the cab driver. Davis went to trial on the charge and was acquitted. A week later, prosecutors charged Davis for the Pimlico shooting.[48] At trial, Davis contended that police planted a gun and attributed it to Davis as part of a cover‑up, likely motivated by a desire not to further stir popular unrest in the wake of the Gray incident.
The first prosecution ended in a mistrial, and efforts to retry Davis began. Only weeks prior to the start of the retrial, the State disclosed the existence of a jailhouse informant who claimed Davis confessed to him. With the help of this new witness, Davis was convicted. The conviction, however, was thrown out for procedural errors, and Davis was then tried (unsuccessfully) twice more. Unwilling to let the case go, the Baltimore City State’s Attorney was preparing for yet another trial against Davis on new charges when she lost her bid for re‑election. The new State’s Attorney initiated a review of the case against Davis and eventually dropped charges against him,[49] explaining the “dismissal [was] about the prosecutorial missteps of my predecessor in her pursuit of a conviction at all costs.”[50] He blamed Davis’s extended saga on “prosecutorial vindictiveness.”[51]
B. The Link Between Informant Use and False Confessions
As the Malik case discussed above illustrates, there is a significant correlation in the dataset between jailhouse informant usage and false confessions. Of the nearly 3,500 exonerations recorded in the NRE, researchers have identified false confessions as a contributing factor in 439 cases, or 12.6 percent of the total. Earlier studies consistently have found false confessions to have been a factor in a similar percentage of cases.[52] In the jailhouse informant dataset here, in contrast, false confessions were documented in 30 percent of the cases—more than double the normal rate.[53] This suggests that cops and prosecutors who utilize informants are more likely to also coerce or manufacture false confessions from defendants. Detective Scarcella’s methods—which generated numerous wrongful convictions—certainly fit this description[54]
Another instance of the false confession‑jailhouse informant link can be seen in the prosecution of a Pennsylvania man named Walter Ogrod.[55] The chain of events that ultimately led to Ogrod’s wrongful conviction started in 1988, when the naked body of a four‑year‑old girl was found in a cardboard box near her home in north Philadelphia.[56] The case drew “massive media attention,” but leads in the case did not pan out, and four years later the investigation remained open, at which time it was assigned to Philadelphia Detectives Martin Devlin and Paul Worrell.[57] The detectives were well known for their success in “getting confessions,” a skill later revealed to stem from their willingness to employ coercive interrogation methods.[58]
After failing to make headway in the case, the detectives turned their attention to Ogrod, who was intellectually disabled and lived across the street from the victim. After an alleged eighteen‑hour interrogation, the detectives got Ogrod to confess to the murder. The State’s case against Ogrod, which centered on the confession, was weak. No physical evidence linked Ogrod to the crime, and Ogrod recanted the confession shortly after the interrogation. After nine hours of deliberation, the jury returned an acquittal verdict, but when the verdict was announced, one of the jurors protested that they did not agree with the verdict. Thus, instead of an acquittal, the case ended in a mistrial.
Ogrod was retried two years later. But this time, the State’s case was supplemented with two jailhouse informants who claimed that Ogrod had confessed to them. With this new evidence, prosecutors won a conviction. The following day, the jury took only ninety minutes to sentence Walter Ogrod to death. Years later, an investigation would show that one of the jailhouse informants who testified at Ogrod’s second trial was a serial informant who had previously worked with the prosecutor. Investigators also found that a substantial amount of exculpatory evidence had been withheld. DNA tests ultimately excluded Ogrod as the perpetrator, and Ogrod was exonerated in 2020.[59] Meanwhile, in 2021 Detective Devlin was indicted along with two other detectives “on charges of perjury and false swearing” for faulty investigative practices that led to a wrongful conviction in yet another case.[60] In a statement released by the district attorney’s office in conjunction with Ogrod’s exoneration, the district attorney stated that Detective Worrell was “known to the court for a pattern and practice of eliciting false confessions dating back to at least 1992, and has been implicated in other wrongful convictions that have since been vacated.”[61]
Similar patterns of police and prosecutorial corruption are highlighted in the Wayne County, Michigan prosecution of a Detroit man named Bernard Howard.[62] In that case, prosecutors put on a jailhouse informant named Joe Twilley to testify that Howard had confessed involvement in a triple murder. Twilley was a serial snitch who asserted during cross‑examination that he had a special ability to get inmates to trust him. Twilley estimated that forty or fifty inmates had confessed their crimes to him.[63] Twilley also helped prosecutors get a conviction against a man named Ramon Ward. Like Howard, Ward would later be exonerated.[64] The subsequent investigation of both Howard’s and Ward’s wrongful convictions revealed ample evidence of misconduct by police detectives and prosecutors.
The district attorney’s practice of relying on serial informants like Twilley was particularly inexcusable because prosecutors knew that Twilley and other informants were unreliable. Indeed, the deputy chief prosecutor, Robert Agacinski, wrote a memo around the time of Howard’s trial expressly warning prosecutors about using unreliable jailhouse informants and alerting them that police detectives were making promises of leniency to informants that had not been approved by, or even shared with, prosecutors.[65] Agacinski’s memo specifically included a warning about Twilley, describing him as among those who “had claimed to obtain ‘confessions’ in several cases that he was aware of” or in which he had himself “personally prosecuted.”[66] The memo further noted one instance in which police had gone behind a prosecutor’s back after the prosecutor had expressly refused to support a request to reduce Twilley’s sentence on a murder conviction and succeeded in getting the judge to reduce Twilley’s sentence anyway.[67] Given the memo, prosecutors clearly knew that police were fabricating false confessions and inducing jailhouse snitches to fabricate them against others who were in custody.[68]
Years after Howard’s wrongful conviction, a Conviction Integrity Unit (“CIU”) investigation recommended dismissal of charges against Howard, concluding that Howard’s confession “appears to be something that the police stitched together through information from the jail house informants and then presented to Mr. Howard, who, at the time, was only eighteen years old and basically illiterate.”[69] According to Valerie Newman, the head of the CIU, “Detroit homicide detectives coerced Howard into confessing to the murders.”[70] Newman said, “the confession was ‘presented as, you know, sign this so you can get out of here.’”[71]
Several threads tie these cases together. First, in each of the cases there was a concerted effort to manufacture false evidence. That false evidence often took the form of false confessions, as in the New York firebombing case in which Malik, Ellerbe, and Irons were all wrongly convicted on the basis of false confessions. False confessions also factored prominently in the convictions of Walter Ogrod, Bernard Howard, and Ramon Ward. But false evidence can take other forms as well. The same detectives who coerced false confessions from Malik, Ellerbe, and Irons were found in other cases to have induced street criminals and others to generate false testimony. Detectives in the Davis case planted a gun to implicate Davis in a crime he didn’t commit. In all these cases, jailhouse informants were used to generate even more false evidence against the defendants. Second, these cases are illustrative of the often collaborative nature of the misconduct. In theory, prosecutors should serve as a check on overzealous cops, but as these cases show, all too often prosecutors choose to join forces with unscrupulous cops to convict innocent people. The prosecutors in Detroit knew that police detectives were coercing false confessions and using untrustworthy serial snitches and prosecuted those defendants anyway. The Baltimore City State’s Attorney pursued what was later acknowledged to be a vindictive prosecution against Keith Davis, seemingly to help shield the officers who used excessive force against him from accountability. And, it seems highly unlikely that New York prosecutors were unaware that Detective Scarcella—who had a reputation for producing dubious witnesses and obtaining confessions—and his colleagues were taking shortcuts in their investigations.
C. It’s Not Just Cops and Prosecutors
The culture of corruption engulfing the use of jailhouse informants can extend beyond unscrupulous police detectives and prosecutors to include other, more peripheral players. Some cases in the dataset involve correctional officials who collaborated with prosecutors to generate false informant testimony, or who on occasion themselves gave false testimony.[72] In Jackie Wilson’s case, for example, two Cook County correctional officers “testified that Jackie made statements while awaiting his retrial.”[73] One correctional officer testified, almost certainly falsely, that Wilson said, “[y]ou should have killed us when (you) had the chance, killed me when you had the chance because I already killed two Chicago police officers.”[74] The other correctional officer testified that Wilson threatened to kill a guard “just like I did the other two policemen, way back whenever it was.”[75]
In scandals that have led to national media attention, prosecutors have also been caught running elaborate “snitch tank” systems with the cooperation of numerous law enforcement and corrections officials.[76] One such major scandal emerged in Orange County, California around the use of such a system to secretly gather evidence against jail inmates. Paul Smith, who was exonerated in 2021, was one of the victims of the Orange County scheme[77] in which at least fifty‑seven felony trials were compromised.[78] The convictions of twenty‑one other defendants were vacated after it was revealed that law enforcement officials were using informants to elicit incriminating statements after charges had been filed in violation of the Sixth Amendment and concealing information about the inducements used to reward the informants.[79] The United States Department of Justice released a report in 2022 documenting rampant misconduct in the use of jailhouse informants by Orange County prosecutors, the Orange County Sheriff’s Department, and corrections officials.[80] Similar systems also seem to have been operating in other states as well.[81]
While this data is anecdotal, it shows that a certain type of us‑versus‑them mentality, one that privileges making cases and getting convictions against targeted individuals, can take over a unit, an office, or operate at even larger scales to engulf entire criminal enforcement systems. Documented instances of this occurring suggest the need for systemic safeguards and safety valves to short‑circuit the cascading failures that eventuate in large numbers of unreliable prosecutions and wrongful convictions.
D. The Defense Bar is Not Free of Blame
It is not only law enforcement officials whose conduct should be scrutinized. In many cases, blame also falls on the shoulders of incompetent, lazy, or overburdened defense lawyers.[82] In several of the exoneration cases studied here, defense counsel failed to conduct a basic fact investigation, failed to hire appropriate experts, failed to obtain discovery or review the discovery that had been made available to them, and failed to mount any sort of competent defense despite often having numerous available avenues.[83] Frequently, defense counsel failed to conduct an even minimally competent effort to impeach the jailhouse informant.[84] Competent defense counsel might well have prevented at least some of these wrongful convictions from occurring.
Indeed, the dataset cases exhibit exceptionally high rates of inadequate legal representation.[85] Whereas less than a third of the cases in the Registry involve documented inadequate legal defense, half of the cases in the jailhouse informant dataset were so identified. While we can only speculate about the reasons for this correlation, there is little doubt that competent defense counsel provides a critical check on the abuse of unreliable jailhouse informant testimony. A good lawyer is far more likely to successfully obtain and review discovery, to independently investigate an informant’s background and history of cooperation with state agents, and to inform juries regarding the high level of unreliability inherent in incentivized testimony. Poor lawyering, as so often occurred in these cases, is far more likely to leave false informant testimony unchallenged and as a result, to lead to wrongful convictions of the innocent.
E. High Profile Pressures Lead to Misuse of Informants
The general public is often shocked to learn that police, prosecutors, and other public officials have conspired to secure the convictions of innocent people. Part of the explanation for why law enforcement officials resort to such tactics is the often‑extraordinary public pressure to get convictions.[86] Politics, heavy media coverage, and particularly brutal facts or sympathetic victims always put intense pressure on police and prosecutors to make arrests and obtain convictions.[87] This dynamic was apparent in many of the cases in the dataset where, like in the Ogrod case, media scrutiny was especially great.[88]
This was the case in the wrongful conviction of Chicago man, James Harris III.[89] The victim in the Harris case was from a prominent family residing in Chicago’s “Gold Coast,” a wealthy part of the city unaccustomed to such shocking crimes.
As Harris’s lawyers argued during opening arguments (unpersuasively to the jury):
What you have is a crime that will not be tolerated, a crime that must be avenged . . . and what develops is an atmosphere of pressure, of panic, and desperation on the part of the community, on the part of the police. And the evidence will show that what results is a desperate prosecution such as this.[90]
Harris’s lawyers were proven correct when Harris was exonerated in 2023.
A similar media frenzy contributed to the pressure on prosecutors to obtain a conviction after a high‑profile rape‑murder in Hawaii.[91] Even more intense media pressure followed the firebombing, discussed above, of a token booth in New York City which led to the wrongful conviction of Thomas Malik and his codefendants, James Irons and Vincent Ellerbe.[92] In these and other cases, the decision by law enforcement officials to turn to jailhouse informants to shore up what might otherwise be weak cases because of perceived pressure to get convictions proved a key factor in the processes that led to the wrongful convictions.
III. Making False Jailhouse Informant Testimony
False confessions induced or coerced by cops share an important evidentiary similarity to false jailhouse informant testimony because jailhouse informant testimony almost always consists of a claim by the informant that the defendant confessed or made an incriminating admission to a fellow inmate. Though jailhouse informants typically carry far less credibility with jurors than testifying police officers, nonetheless when an informant testifies that the defendant confessed, jurors often believe it even when such belief is unwarranted.
Why might this be the case? Of course, a dataset of exoneration cases cannot provide direct evidence on which to base general assessments about how persuasive or compelling jailhouse informant testimony is. Nonetheless, there is ample reason to believe that jurors do find such testimony compelling. Two primary explanations emerge from the research and data.
First, the evidence provided by jailhouse informants is the defendant’s purported confession, a form of evidence that is singularly persuasive. Prosecutors present statements attributed to the defendant by the jailhouse informant as if they were factually true. If the jury believes the jailhouse informant, and prosecutors strongly urge them to do so, no other evidence is needed. In the Caldwell case, for instance, the prosecutor said in closing arguments that
[Caldwell] told you [that he committed the crime] through the testimony of George Thompson. . . . He told us what he did. . . . His words from his very own mouth. Do you need any better evidence? Can you think of any better direct evidence of what happened in that dark bedroom? That’s from his lips.[93]
Except it was not from Caldwell’s lips. It was from the lips of the informant, and it was false.
The case against Lacino Hamilton was built almost entirely on statements reported by the jailhouse informant during the preliminary hearing.[94] There was no physical or forensic evidence connecting Hamilton to the crime. The main evidence at trial was the jailhouse informant’s prior testimony, read to the jury because the informant died before trial. The jury found the jailhouse informant’s statements to be enough and convicted Hamilton.[95]
Numerous research studies suggest that jurors tend to find jailhouse informant testimony to be highly persuasive evidence primarily because it is a form of confession evidence.[96] One study conducted by Saul Kassin and Katherine Neumann, for instance, found that mock jurors were far more likely to convict suspects when the evidence included a confession than when other types of traditional evidence, such as eyewitness identifications or physical evidence, were presented.[97] Because the testimony proffered by jailhouse informants almost always concerns an alleged confession made by a fellow prisoner, these research findings are especially pertinent here. Research also indicates that jurors generally are not good at assessing how witness incentives or bias might affect the reliability or credibility of witness testimony. As a result, jurors are far more likely to accept a statement made by an incentivized witness, such as a jailhouse informant, without considering the existence and magnitude of the incentives that might have motivated the witness to testify falsely.
The inability of jurors to adequately discount incentivized testimony was part of the findings, for example, of a study conducted by Saul Kassin and Holly Sukel concerning coerced confessions. The Supreme Court has long held that confessions that were induced by force or coercion are inadmissible both because they render confessions involuntarily, and because they are inherently unreliable.[98] In the Kassin and Sukel study, mock jurors read transcripts of a murder trial containing either a coerced confession, a non‑coerced confession, or no confession.[99] While the mock jurors were able to appropriately identify the coerced confessions and accord less evidentiary weight to those confessions, they nonetheless convicted defendants at a higher rate in all cases where there was confession evidence regardless of whether the confession was coerced or not, or whether they had received an instruction from the judge to disregard it.[100] The study thus suggests that when jurors hear confession evidence, even if they know that the confession was coerced, they tend to subconsciously credit the confession as true. There is little doubt that the same mechanisms are at work where incentivized witnesses testify they heard a defendant confess to a crime. Although jurors will acknowledge the source of the confession might be unreliable, the studies suggest they nonetheless tend to credit the alleged confession as true without regard to the unreliability of the source when assessing the guilt or innocence of the alleged confessor.
While primary confessions are those made by a suspect directly to authorities, a secondary confession occurs when a third party reports that another person has confessed to a crime. Jailhouse informant testimony usually involves secondary confessions, and researchers have found that the same dynamics that lead jurors to ignore indicators of unreliability when weighing primary confession evidence apply equally to secondary confessions. For instance, a study by lead author Jeffrey Neuschatz provided mock jurors with trial transcripts containing secondary confession evidence from an accomplice witness, a jailhouse informant, or a member of the community, or with no secondary confession evidence at all. In half of the transcripts, jurors were informed that the cooperating witness providing the secondary confession was given an incentive to testify.[101] The study results revealed that information regarding the existence of witness incentives to testify did not affect the participants’ guilt/innocence determination.[102] In short, mock jurors systematically failed to discount the credibility of an incentivized witness despite being aware of reasons why the witness’s testimony might be biased or false. This research strongly suggests that jurors will consistently treat jailhouse informant testimony as more reliable than it actually is.
While the Neuschatz study found evidence that mock jurors generally treat secondary confessions such as those provided by jailhouse informants as less persuasive than primary confessions, it also found that secondary confession evidence is nonetheless highly persuasive evidence.[103] This is consistent with what we know from the wrongful convictions research. In numerous wrongful conviction cases involving jailhouse informants, there is strong evidence that jurors treated the jailhouse informant’s false report of defendant’s alleged confession “close cases,” that is, in cases where the other inculpatory evidence is weak.[104] False jailhouse informant testimony will thus rarely be harmless or inconsequential.
Anecdotal evidence from the dataset examined here provides further support for the conclusion that jurors often fail to properly discount informant testimony. Indeed, false evidence from jailhouse informants is often the clear difference between an unwinnable case and a slam‑dunk conviction. Consider again the Walter Ogrod case. As described above, Ogrod was on trial for the murder of a four‑year‑old girl, but the evidence linking Ogrod to the crime was weak. No murder weapon was recovered and no physical evidence linked Ogrod to the crime. The first prosecution ended in a mistrial after the jury had agreed to acquit him, but one juror changed their mind during delivery of the verdict. Ogrod was retried months later, this time with the benefit of two newly proffered jailhouse informants. With the addition of this new testimony, Ogrod was convicted, and the next day at the penalty hearing the jury “deliberated less than ninety minutes before unanimously agreeing that Ogrod deserved to die.”[105] The additional testimony by the jailhouse informants appears to have decisively altered the trial outcome.
The same pattern emerged in the Olin Coones case, where again the state initially went to trial against the defendant with very weak evidence and as a result, failed to secure a unanimous verdict. The emergence of a jailhouse informant prior to the second trial, however, tipped the evidentiary balance and Coones was convicted at retrial.[106] The outcome‑altering effect of jailhouse informant testimony was also apparent in the Clifton Caldwell case, albeit there the order of events was effectively reversed.[107] In that case, Caldwell and two codefendants were charged with aggravated rape. A jailhouse informant came forward and claimed that Caldwell admitted to participating in the crime. To avoid Bruton issues,[108] Caldwell’s case was severed from his codefendants and he was tried first. Based in large part on the jailhouse informant’s testimony, the jury convicted Caldwell of aggravated rape and he was sentenced to twenty‑seven to thirty‑five years in prison. Caldwell’s codefendants were tried two weeks later based on the same evidence minus the jailhouse informant. This time, the jury concluded that the evidence was insufficient and acquitted both codefendants.[109] Again, the presence or absence of the informant testimony clearly proved dispositive of outcomes. [110]
There is another reason that use of false jailhouse informant testimony so often leads to false convictions. Recent research demonstrates that presentation of a secondary confession is not only persuasive in itself, it also impacts juror perceptions about the reliability of other evidence in the case. Professor Baylee Jenkins and her co‑authors conducted a study to test the impact of secondary evidence testimony provided by jailhouse informants on assessments of forensic evidence. The study asked mock jurors to evaluate whether a handwriting sample obtained from the crime scene matched a sample provided by the defendant.[111] The study found that jurors informed that a jailhouse informant had reported the defendant’s confession strongly impacted the mock jurors’ perceptions about whether the handwriting samples “matched.” To be precise, knowledge of the secondary confession from the jailhouse informant made jurors more likely to judge the “robbery note” purportedly written by the bank robber as matching a handwriting sample obtained from the defendant, regardless of whether or not the samples were, in fact, produced by the same source.[112] Based on the study results, the researchers concluded that the secondary confession created a “forensic confirmation bias” that affected the judgment of fact finders. This bias caused them to overlook problems with the evidence that they would have been more likely to discredit absent exposure to the secondary confession.
Forensic confirmation bias provides a strong explanation for why the addition of a jailhouse informant tends so often to swing case outcomes. Not only does the informant provide new evidence, but the inclusion of the informant’s testimony alters jurors’ perceptions about the strength or weakness of other evidence the prosecution has also presented.
A. Fabricating Persuasive Confessions
One of the ways in which false confessions—either those obtained directly from the suspect or secondary confessions such as those testified to by jailhouse informants—are made to seem believable is by ensuring that the confessions contain details of the crimes that supposedly only the real perpetrators would know. A critical lesson of jailhouse informant research is that it is remarkably easy for informants to obtain such details. The recent exonerations studied here illustrate a variety of methods by which this is accomplished.
One way to get the information necessary to construct a plausible story is to gain access to a fellow inmate’s legal papers. These papers might include: the charging documents that spell out the factual allegations against the defendant, motions to suppress statements made by the defendant, witness testimony, or other descriptions of the crime found in legal documents, news articles, or personal letters. It became apparent that the jailhouse informant who falsely testified against George Bell, for example, constructed his account on this basis after it was discovered that erroneous details included in his testimony at trial originated in a newspaper story that was in Bell’s legal papers.[113] Similarly, Joshun Edwards and three other men were victimized by a fellow inmate with whom Edwards unwisely shared his legal papers based on the informant’s false promise to come forward with information implicating others if Edwards paid him an undisclosed sum of money. The informant then proceeded to testify at trial against the four men. All of the details in the inmate’s jailhouse informant testimony could be found in the police reports contained in Edwards’s legal papers.[114] One of the informants in Barry Williams’s case, Ernest Cox, also employed this strategy. According to Cox’s brother who was housed in the same county jail, Cox obtained the details he needed to testify by reading William’s preliminary hearing transcript.[115] It is not clear how the informant got hold of the transcript, although it seems possible that it was given to him by prosecutors in the case because when he later recanted, Cox said that he testified against Williams because the Deputy District Attorney had threatened to charge him as an accessory to murder in a separate case if he did not.[116] Kino Christian was victimized by an informant who obtained access to his codefendant’s legal papers after the codefendant invited the informant to review them when he offered to help him with his case.[117]
News reports are another common source of case information regurgitated at trial by jailhouse informants. In Thomas Malik’s case, the CRU that investigated Malik’s conviction concluded that the informant in the case cobbled testimony together from reading newspaper stories.[118] The informant in the wrongful conviction of Kenneth Nixon also obtained details about Nixon’s case from watching TV news, as did the informants in several other cases in the dataset.[119]
In some cases, the jailhouse informant is privy to details of the crime because he was, in fact, either the actual perpetrator or otherwise involved. This may have been true in the case against Mark Purnell. A woman named Tameka Giles was shot during an attempted robbery, and police eventually charged Purnell with the crime. The victim’s husband, Ernest Giles, had initially identified a man named Kellee Mitchell as one of the robbers. Other evidence linked Mitchell to a revolver that may have been the murder weapon. Mitchell, however, was not charged in the case. Instead, he provided jailhouse informant testimony alleging that Purnell had confessed to the killing while the two men had been together in jail on unrelated charges.[120]
Of course, the easiest way for the informant to concoct a false narrative is to get the key facts directly from corrupt cops and prosecutors, a scenario that occurred all too often in the cases studied.[121] For example, the informant who testified against Lacino Hamilton, Oliver Cowan, was allegedly told by police detectives that “if he were willing to ‘help’ homicide detectives with some of their murder cases, the detectives would help him with his case.”[122] They also allegedly told Cowan that “he didn’t have to worry if he didn’t have any information; detectives would take care of that.”[123] The detective then allegedly gave Cowan “two copies of a pre‑written statement” and instructed Cowan “to sign one and to memorize the other in preparation for a preliminary hearing.”[124] Similarly, according to the jailhouse informant who testified against Lee Harris and years later recanted his testimony, he was approached in jail by two detectives who “showed him a photograph of Harris and asked if he knew the man in the photo.”[125] When the informant said he didn’t, the detective told him he was “about to get know him,” and instructed him “to talk to Harris and then to call the police to report that Harris had confessed.”[126] The informant further stated “his testimony was scripted and that it was all a lie.”[127]
Jailhouse informants’ falsehoods are not limited to fabricating confessions made by fellow inmates. They also invent encounters with those inmates to create a plausible cover story. For example, the jailhouse informant in Keith Davis’s case claimed that Davis bragged about killing the victim.[128] Davis and the jailhouse informant were housed in different sections of the prison, but according to the jailhouse informant, Davis came to his cell to purchase alcohol from his cellmate. After Davis was convicted, he filed a motion for a new trial. At a hearing on the motion, Itisham Butt, the jailhouse informant’s cellmate, testified that the jailhouse informant’s story was pure fabrication. Butt testified that he was a devout Muslim who didn’t drink and never had, and that not only did he not sell alcohol to Davis, he had never met him.[129] Numerous other cases illustrate jailhouse informants fabricating meetings or conversations that never occurred in order to bolster their secondary confession.[130] Wrongful convictions in these cases could have been prevented but were not because evidence that would have disproven the lies was either withheld by the prosecution, never uncovered by defense counsel due to shoddy investigation, or simply ignored by the jury.
Inmates often learn from each other that snitching on a fellow inmate is the quickest ticket to freedom.[131] Charles Jason Lively, for example, was falsely convicted of killing a doctor.[132] The jailhouse informant that testified against him later reported that a fellow inmate told him that to get out of jail, “[a]ll you’ve got to do is tell them that the Lively boy said he killed that doc.”[133] Albert Ian Schweitzer was originally falsely fingered as a participant in the Hawaii rape‑murder discussed above by a man named Frank Pauline, Jr., who told police that he had been present during the crime and falsely identified Schweitzer and his brother as the assailants. He later admitted that he had lied about the incident, explaining that his half‑brother, John Gonsalves, “had talked him into cooperating with police to help Gonsalves get his drug charges dropped.”[134]
These cases illustrate how simple it often is for inmates to generate false but believable testimony, either with the help and direction of cops and prosecutors or through their own entrepreneurship. Seemingly credible details can be mined from a variety of available sources, and potential informants are encouraged by fellow inmates, police detectives, and prosecutors who convince them, if they need convincing, that snitching is the quickest means of escape from incarceration.
B. The Deals
Prior research on jailhouse informants has documented a wide range of inducements offered to informants to convince them to cooperate.[135] Both police and prosecutors have leverage available to induce potential informants to cooperate. Police decide who to arrest or whether to refrain from making an arrest and how to frame initial charges. They also work with prosecutors and can make favorable recommendations to prosecutors about their cases. Prosecutors have even more leverage at their disposal. As Alexandra Natapoff explains,
Prosecutors have a wide range of negotiating tools. They can delay an indictment or an existing case to give an informant a chance to testify before a grand jury or to go back into the community or organization and gather information, sometimes known as “working off” a charge. They can drop charges, alter charges, or promise to make favorable recommendations to the judge at sentencing.[136]
These inducements create powerful incentives for informants to falsely testify and are the engines that drive the informant machine.[137]
The inducements offered in the cases in this dataset are consistent with the prior research. First, the rewards for testifying as a jailhouse informant are often enormous. The jailhouse informant in the Barry Williams case was threatened with being charged in a separate case as an accessory to murder, which carried a fifteen to twenty‑five‑year sentence. By testifying against Williams, the jailhouse informant was instead given immunity in that case.[138] The jailhouse informant who gave a false statement implicating Robert DuBoise initially faced charges carrying two potential life sentences. In exchange for his statement, he pled guilty and was sentenced to five years.[139] One of the informants testifying against Michael Rhynes faced a potential life sentence for a robbery charge as a habitual offender. In exchange for his testimony, the robbery charge was reduced to a misdemeanor and a sentence of time‑served.[140] The jailhouse informant in Robert Gold‑Smith’s case received, among other things, a reduced felony aggravated battery charge to a misdemeanor.[141]
In other cases, the benefits were less extreme, but nonetheless sufficient to induce informants to falsely testify. In a case against a woman named Madeline Mendoza, for example, a jailhouse informant gave a series of statements falsely implicating Mendoza and two other suspects.[142] Prosecutors continued to rely on the informant even though her testimony evolved and morphed dramatically over time.[143] As reward for testifying in the “right way,” drug charges and a federal probation violation against the informant were dropped.[144]
The deals made often included cash payments as well, meaning that jailhouse informant testimony was often literally “purchased” by the State. A jailhouse informant in the Curtis Flowers debacle (Flowers was tried six times before the Supreme Court put an end to proceedings) who recanted his testimony, said he not only was coerced into testifying by threats, but was offered a share of the $30,000 reward money that had been posted in the case.[145] Other jailhouse informants in the dataset also received cash compensation for their false testimony.[146]
The problem of incentivized witness testimony is not limited to jailhouse informants,[147] although jailhouse informants tend to exemplify the most egregious forms of it.[148] In the Lamar Johnson case, for example, an alleged eyewitness named Elking who had purported to identify Johnson in a lineup later admitted that his identification was false.[149] He stated that prosecutors told him which suspects to pick out of the lineup, that he had been given cash payments by the State to help him move (records indicate the amount was over $4,000), and that he had been told that if he did not testify for the prosecution, that he might himself be charged with the offense.[150] Elking said he was told, “either you’re going to be on the winning side or the losing side. Which side do you want to be on?”[151]
Similarly, David Faulkner was wrongfully convicted based in part on testimony from an alleged eyewitness who said she saw Faulkner and the other defendants in the vicinity of the crime scene that day. In exchange for her testimony, the witness negotiated dismissal of criminal drug charges against her grandson, plus $10,000 in reward money.[152] An alleged eyewitness to a prison murder resolved six pending cases for no additional jailtime in exchange for false testimony that he saw Raymond Champagne stab a fellow prisoner.[153] In Robert DuBoise’s case, a witness who was not incarcerated testified that DuBoise had supposedly made incriminating statements to him.[154] At the time, the witness was also working with prosecutors in another murder trial in which he could have been charged as an accessory. After testifying for the State in both cases, all criminal charges against him were dropped.[155]
These cases illustrate the remarkable variety of deals and dealmaking that incentivizes false testimony. It also indicates that concerns regarding the impact of incentives on the truthfulness of witness testimony should begin with jailhouse informants, but not end there. Incentives provided to supposed eyewitnesses in exchange for their testimony are, if anything, even more prejudicial than those provided to informants, codefendants, and alleged accomplices. The self‑interested nature of informant and accomplice testimony should be apparent to jurors who can, at least in theory, discount the weight of such testimony in response. When fact witnesses are incentivized to provide incriminating testimony, the self‑interested nature of such testimony will likely go wholly unnoticed.
C. Failure to Disclose
The Sixth Amendment’s Confrontation Clause ensures defendants the right to confront witnesses against them.[156] Although there is no clear consensus whether that right also entitles a defendant to the pretrial discovery necessary to effectively cross‑examine witnesses, there is no doubt that such discovery is vital for effective impeachment of jailhouse informants. Even in the best of circumstances, jailhouse informant testimony is extremely difficult to cross‑examine effectively. In most cases, the jailhouse informant’s testimony constitutes little more than an assertion that the defendant admitted, in private, to involvement in prior criminal wrongdoing. Therefore, an informant’s testimony can swing the odds heavily toward conviction even if nothing more than the broadest alleged admissions by the defendant are recounted. The informant need not claim to know any of the details of the alleged crime, nor recount details of the alleged crime accurately. Because a supposed confession of a fellow inmate is almost never subject to corroboration, it can almost never be proven false absent evidence that the informant and the defendant never met or had opportunity to discuss the defendant’s alleged crime.
Although it remains a point of contention whether the Confrontation Clause mandates adequate pretrial discovery, there is no doubt that at minimum, the Due Process Clause requires prosecutors to disclose all material exculpatory evidence in the State’s possession. That right was established in Brady v. Maryland and remains a vital and essential component of due process.[157] When prosecutors fail to comply with their constitutional obligations under Brady, however, the difficult task of effectively impeaching jailhouse informants becomes nearly impossible. Prior research has demonstrated that failures to disclose exculpatory evidence are pervasive in cases involving jailhouse informants.[158] The exoneration cases involving jailhouse snitches are almost always accompanied by some type of prosecutorial misconduct related to the failure to disclose exculpatory information to the defense. Most commonly, prosecutors fail to disclose either that a deal had been made with an informant or the full extent or scope of the deal.[159] In the Lamar Johnson case, for instance, prosecutors never disclosed that the jailhouse informant in the case had written a letter to them asking for a reduction in his sentence or a pardon from the governor in exchange for his testimony or that a deal had been made in the case[160] Notwithstanding what they told the jury, prosecutors requested that the informant be granted an early release and be entered into a rehabilitation program.[161]
The prosecutor in Lacino Hamilton’s case neglected to disclose that the jailhouse informant, who faced fifteen years on a burglary charge, was allowed to plead guilty in exchange for a one‑year sentence (that he in fact was never required to serve) and inaction on a parole violation.[162] In the Lee Harris case, prosecutors failed to disclose that the jailhouse informant was released on a non‑cash bond at the request of the prosecutor.[163] The prosecutor in Walter Ogrod’s case failed to disclose that the two jailhouse informants, who had cooperated with the prosecution, received leniency in exchange for their cooperation.[164]
Often, disclosure omissions concern the jailhouse informant’s criminal history, a complete record of which is essential for impeachment purposes. In Keith Davis’s case, prosecutors failed to disclose that the jailhouse informant had admitted as part of his plea agreement to participating in a gang murder.[165] Lamar Johnson’s prosecutors failed to reveal that the jailhouse informant’s arrest record was more extensive than what he testified to at trial.[166]
Disclosure omissions also frequently concern the informant’s prior history of working with the police as an informant. Such information is critically important for successfully impeaching a jailhouse informant’s credibility. The Innocence Project, an organization dedicated to assisting wrongly convicted defendants to prove their innocence largely through use of DNA testing, as well as several states, have urged mandatory disclosure of a jailhouse informant’s prior history of informing as part of pretrial discovery.[167] In Clifton Caldwell’s case, a jailhouse informant emerged at the last minute. At trial, the informant testified that Caldwell admitted that he held the arms of the victim while she was raped. He denied asking for any reward in exchange for his testimony. Instead, he stated that his motive for coming forward with the information was his strong feelings about rape because his girlfriend had been raped and his sister had been raped by their father as a child. But prosecutors failed to disclose that the informant previously had worked for police in a similar kind of case involving rape allegations, which might have cast doubt on the informant’s testimony.[168]
Prosecutors in the Lee Harris case failed to disclose that the informant had been paid at least $1,000 for cooperating with them in a prior case.[169] Prosecutors in the Lamar Johnson case failed to disclose that the informant had testified as a jailhouse informant in a Missouri murder trial.[170] Prosecutors routinely fail to disclose other types of impeachment evidence, such as that an informant used racial slurs to refer to the defendant.[171] In the Olin “Pete” Coones case, the prosecutor failed to disclose a complete criminal history of the jailhouse informant, providing only a watered‑down version. He also failed to disclose the demands that the jailhouse informant made prior to agreeing to testify, or the promise the prosecutor had made to intervene with the Kansas Department of Corrections on his behalf, or that the jailhouse informant had offered to inform on an additional inmate from whom he had stolen information.[172] In another case, after a detective retired from the police force, he flew to England to attend the wedding of the daughter of a jailhouse informant with whom he had worked.[173] This somewhat extraordinary relationship was also never disclosed to the defendant.[174]
Prosecutors also routinely fail to disclose the number of contacts between state agents and informants, as well as the substance of conversations and communications between them. For example, in the Robert DuBoise case, the prosecutor told the jury in “closing arguments that he had no previous contact” with the jailhouse informant.[175] That claim was false. In fact, the prosecutor had overseen the indictment of the jailhouse informant with regard to one of the charges that was included in the plea deal.[176] In the Lamar Johnson case, prosecutors failed to disclose that approximately one month before trial the jailhouse informant had written a letter to them offering his assistance and listing the benefits he wanted in exchange. Prosecutors also did not disclose that after Johnson was convicted, the jailhouse informant was granted relief consistent with what he asked for in the letter.[177] In the Walter Ogrod case, prosecutors failed to disclose that the jailhouse informant previously cooperated with prosecutors in numerous other cases, including twelve murder prosecutions, and that one of those cases involved the same prosecutor who handled Ogrod’s second trial.[178]
Just as with jailhouse informants, prosecutors routinely fail to disclose critically important exculpatory information concerning other types of incentivized witnesses. For example, the grandmother who testified as a witness against David Faulkner also threatened to provide testimony that was favorable to Faulkner if drug charges against her grandson were not dismissed. The charges were dismissed, and while prosecutors disclosed this fact to the defense a few days prior to trial, they never disclosed the grandmother’s demand.[179]
In some cases, a prosecutor’s failure to disclose that a deal has been made, or that incentives have been provided in exchange for an informant’s testimony, is compounded by the prosecutor’s false assertions to jurors at trial that no such incentives have been offered. Needless to say, an informant’s claim that the defendant has confessed to the crime will be compelling—and indeed often dispositive—only if the jury finds the claim credible. One reason jurors so often believe informants is that prosecutors routinely vouch for the accuracy of informant testimony, putting their own authority and credibility behind the witness in a way that may relieve jurors of their responsibility to make their own credibility assessment.
Prosecutors vouch for the credibility of jailhouse informants in a variety of ways. First, the mere fact that the prosecutor proffers the witness is an implicit form of vouching.[180] By putting the jailhouse informant on the witness stand and treating the jailhouse informant as a friendly witness, the prosecutor communicates to the jury that the prosecutor believes the jailhouse informant is telling the truth. Jurors likely assume that the prosecutor has vetted the jailhouse informant and wouldn’t put on a witness that the prosecutor was not convinced was truthful. Prosecutors often vouch more explicitly for jailhouse informants as well. One way they do this is by reiterating the jailhouse informant’s (false) denial of his motives for testifying on behalf of the State. For instance, in the prosecutor’s closing argument in the Lamar Johnson trial, the prosecutor reassured jurors that the jailhouse informant had no incentive to lie:
The man [referring to the informant] may be a burglar, he may be someone who carries a gun, I think he had another charge there too but he’s a man that draws the line…. ‘This was a terrible waste of a life. It was cold‑blooded murder and you draw the line. Even criminals, people in jail have got some morals [and] say you know, enough is enough on this murder stuff. There’s just too much murder. I can’t keep my mouth shut and turn my face because of what has happened. Mock [the jailhouse informant] stood up and was counting, counting as a honest, God‑fearing man to tell you the truth.[181]
Another way prosecutors deceptively vouch for witnesses is by falsely reassuring jurors that the jailhouse informant is not receiving anything in return for their testimony. These claims amplify the informant’s own false denials and lend them further credence. In the Carnevale case, for example, the prosecutor told the jury that the jailhouse informant “was a credible witness because he had nothing to gain from his testimony.”[182] This statement was plainly false, as confirmed by discovery uncovered later indicating that when the jailhouse informant agreed to cooperate in the case, the informant faced charges carrying more than one hundred years of penalties, and that as a result of the cooperation, the informant instead received a mere fourteen‑month sentence, well below the guidelines minimums.[183] Prosecutors made the same false claims to jurors in numerous cases in the dataset, and jurors repeatedly believed them.[184]
IV. Red Flags
The only truly effective “fix” for the unreliable jailhouse informant testimony problem is to ban its use altogether.[185] Until criminal legal system actors muster the will to take that step, wrongful convictions stemming from false jailhouse informant testimony will continue to occur. In the meantime, legal actors can at least be cognizant of the dangers inherent in the practice.
The dataset here underscores at least four powerful signs, or “red flags,” that a wrongful conviction has occurred. These red flags include: (1) the fact that one or more incentivized witnesses has recanted their testimony at or after trial; (2) the appearance of a “serial snitch,” that is, an informant with a lengthy history of cooperation with police or prosecutors; (3) the opportunistic emergence of jailhouse informants, either at the eleventh hour, or as new witnesses following a prior unsuccessful prosecution; and (4) a jailhouse informant’s denial of receiving any inducement in exchange for testimony.
A. Recantations
A witness’s recantation occurs when a witness disavows statements or testimony provided at an earlier point in time. Recantations are a common feature in wrongful convictions predicated on false witness testimony, although courts, for a wide variety of reasons, tend to discount or dismiss their significance.[186] The number of recantations from jailhouse informants and other incentivized witnesses in the exoneration cases is significant,[187] and the often decades‑long gap between a jailhouse informant’s recantation and the defendant’s ultimate exoneration is, unfortunately, shocking. All parties involved in the appellate and post‑conviction process, but particularly post‑conviction and habeas courts, conviction integrity units, innocence projects, and prosecutors and defense attorneys need to take recantations far more seriously than they now do.[188] As it stands, snitches regularly recant, but their recantations are often either not believed or thought to be immaterial.[189] The latter conclusion is sometimes based on the ironic assertion that because everyone already knew the snitch was lying, the recantation adds no new information, and as result is ‘‘inconsequential,”[190] lacking sufficient probative value to warrant relief. [191]
Certainly, the dataset includes numerous instances in which a jailhouse informant recanted their trial testimony and that recantation was ignored. Ralph Birch, for example, was convicted of murder in 1989, based in part on the testimony of a jailhouse informant who testified that Birch had confessed the killing to him—a jailhouse informant who had been promised “favorable sentencing and prosecution recommendations” on charges against him in two different states.[192] The jailhouse informant recanted in 2008, and two other witnesses in the case also recanted. Based on the recantations and other new evidence, Birch filed a habeas petition but the petition was denied. According to the judge, the witness recantations and other exculpatory evidence presented “were either unreliable or inconsequential.”[193] Birch was exonerated after thirty‑one years of incarceration and twelve years after the jailhouse informant and other witnesses had recanted their trial testimony.[194]
Lee Harris likely would not have been exonerated were it not for a former cellmate, Robert Chattler, who was determined to help him after Chattler had been released from prison. Chattler located an inmate named David Toles who had testified against Harris at Harris’s trial. At the time, Toles was in a prison in Wisconsin. Chattler wrote him a letter accusing him of lying against Harris, and Toles reportedly explained in an undated and unsigned letter that, “I was twenty‑four when this happen [sic] and each year following it I’ve lived in pure hell. Don’t worry Mr. Harris I’m coming to get you my brother. I know you [won’t] ever be able to forgive me for what happen [sic] but I want you to know I’m so very sorry for what you had to face all those years of your life.”[195] In 2004, Toles consented to an interview with the Cook County State’s Attorney’s office. During the interview, Toles recanted his testimony, stating that he had been told to lie by police detectives. Toles admitted during the interview that “his testimony was scripted and that it was all a lie.”[196] The recantation was enough to get the district attorney’s office to agree to new tests on some of the evidence, but some evidence had been lost and tests on other evidence were inconclusive. Harris remained in prison. Years later, the case was brought to the attention of the Cook County CIU, which conducted an investigation but declined to vacate Harris’s charges. A high‑profile article by an investigative reporter helped keep the case alive, and the discovery of yet more exculpatory evidence finally led to Harris’s exoneration in 2023—nineteen years after the jailhouse informant first recanted.[197]
In the epically misguided campaign to falsely convict Curtis Flowers—which stretched over nearly two decades and six trials and prompted an investigation by American Public Media (APM)—three different witnesses, two of whom were jailhouse informants who testified against Flowers about supposed jailhouse confessions, recanted testimony they had previously given against Flowers.[198] One of the recanting informants was Odell Hallmon, who testified at Flowers’s third trial. Hallmon later spoke with an APM reporter via cellphone from prison where he was serving a life sentence for a triple murder committed in 2016. Hallmon told the reporter that he had agreed to testify against Flowers after the district attorney agreed to drop drug charges against him and not to pursue charges in other cases. “‘As far as him telling me he killed some people, hell naw, he ain’t ever told me that. That was a lie,’ Hallmon said in the APM episode. ‘I don’t know nothing about this. . . . It was all make‑believe. Everything was all make‑believe on my part.’”[199]
In some of the exoneration cases, the incentivized witness’s recantation occurred at the trial itself in front of the jury. This happened, for instance, at Raymond Champagne’s trial. Champagne was charged with killing an inmate while he was in custody. A man named Butler initially gave police a statement claiming that he had seen Champagne stab the victim, even though it was later demonstrated that Butler could not have seen the incident from his jail cell given the layout of the wing. While on the stand, Butler recanted his earlier statement and told the jury that the investigating officer “threatened to send him back to Walpole unless he testified for the prosecution.”[200] The jury nonetheless convicted Champagne.
The fact that jailhouse informants will sometimes recant on the witness stand and the jury will still vote to convict underlines just how prejudicial confession evidence tends to be. Once heard, confessions simply cannot be unheard.[201] What seems especially troubling in the wake of exonerations is that no one, including prosecutors or judges, is ever surprised when evidence subsequently comes to light that a jailhouse informant lied, and there are rarely any investigations into what went wrong or how the testimony was proffered or admitted.
B. Serial snitches
Serial snitches—that is, informants who testify in multiple cases on behalf of the prosecution—are a common source of false testimony,[202] and the knowing use of serial snitches is a strong indicator that police or prosecutors are engaged in misconduct.[203]
Serial Snitches foster misconduct in a variety of ways. First, snitches who have developed relationships with police detectives or prosecutors are personally familiar with the incentives available to them if they deliver incriminating evidence. This relationship makes them de facto state agents, and thus should constitutionally bar them from attempting to elicit incriminating statements from persons who have already been charged with crimes.[204] For example, after Olin “Pete” Coones was wrongfully convicted in part based on a jailhouse informant named Rupert, defense lawyers obtained three letters written by Rupert in post‑conviction discovery. In these letters Rupert asked for an attorney to advise him on cooperating and setting forth his demands in exchange for his testimony. To sweeten the deal, Rupert offered to inform on yet another inmate from whom he had stolen information.[205]
Second, as the dataset shows, the cops and prosecutors most likely to be attracted to employing serial snitches are those least concerned with witness integrity, procedural propriety, or evidentiary reliability. The number of cases in which detectives were found to have induced false confessions and were also associated with proffering serial snitches provides evidence of this correlation. For example, the corrupt cops who induced false confessions from Thomas Malik and his codefendants leaned on a serial snitch named Shabazz. [206] Following Malik’s conviction, they continued to use Shabazz as an informant in numerous additional cases, and his testimony in these cases was so flawed that a judge in Manhattan issued a permanent injunction in 2003 barring Shabazz from contacting any law enforcement agency regarding matters outside his own cases.[207]
In Bernard Howard’s case, the police detective who handled the jailhouse informant, Robert Twilley, had, according to the detective, used that informant in twenty other cases. The same detective elicited false confessions from Howard and his codefendant by “threaten[ing] to physically harm them unless they confessed.”[208] In Lacino Hamilton’s case, prosecutors relied on a serial snitch named Cowan, who police credited with helping convict at least six other people of murders. Wayne County prosecutors were aware of Cowan’s history, but instead of avoiding using him, they sought to obfuscate that history. According to Hamilton’s defense team, “evidence showed that the prosecutor who had questioned Cowan at the preliminary examination had made numerous objections to defense questions that prevented the defense from learning about Cowan’s involvement as an informant in other cases.”[209] Moreover, Cowan actively sought to minimize his history as a serial snitch, falsely claiming that he had been “a witness in only two or three other cases.”[210]
Third, the claims of some serial snitches that they possess a special “gift” of winning the trust of fellow inmates and getting those inmates to confess their crimes to them are often dubious. The likelihood that any inmate will confide their guilt to another inmate while awaiting trial—and knowing that snitching is one of the only paths out of extended incarceration for many fellow inmates—is always relatively small.[211] Accordingly, the odds that testimony about an alleged confession is truthful or reliable grow increasingly weaker the more frequently the snitch claims to have witnessed one. The serial snitch who testified against Bernard Howard provides an example. The snitch, Twilley, claimed on cross‑examination “that 40 to 50 different inmates had confessed their crimes to him” without any special inducement.[212] Twilley insisted that his motives were pure and denied testifying pursuant to any deals, claiming instead that his testimony was motivated simply by his “moral principles” which “don’t make me feel good when somebody brags on killing somebody.”[213] These claims were blatant lies. In fact, Twilley had earlier pled guilty to second‑degree murder and been sentenced to a term of twelve to twenty‑five years but received a substantial sentence reduction in exchange for his testimony in this and several other cases.
C. Eleventh‑Hour Witnesses
Jailhouse informants who emerge out of the woodwork on the eve of trial or close thereto,[214] often after failed prior trials, are very common in the dataset and constitute another major red flag.[215] For example, three days before Marvin Cotton’s trial was set to begin, the prosecutor disclosed for the first time that a jailhouse informant would testify. The prosecutor somewhat implausibly claimed that Cotton had admitted involvement in the crime while the two men were awaiting court appearances, even though they were not housed in the same cell and could not even see each other.[216] The informant eventually recanted his testimony against Cotton and admitted he had been promised a sentence reduction in exchange for his testimony.[217] Similarly, the jailhouse informant in the Keith Davis case came forward just weeks before the trial was set to begin.[218]
Jailhouse informants are especially suspect when they emerge as new witnesses to help bolster the State’s case following an earlier failed attempt to win a conviction. The jailhouse informant in Davis’s case, for instance, was used to successfully (though wrongfully) prosecute him after his first trial for murder ended in a mistrial.[219] In Olin “Pete” Coones’s case, the resort to jailhouse informants to shore up a weak case was especially galling. What evidence there was (much of it never disclosed to Coones) indicated that the victim had framed Coones and then committed murder‑suicide. Nonetheless, a jury convicted Coones, but the conviction was thrown out because the prosecution failed to disclose exculpatory evidence. Two weeks before the retrial, the prosecutor announced that a new witness—a jailhouse informant—would testify that Coones had confessed. Coones was convicted and sentenced to fifty years to life.[220]
Curtis Flowers, who ultimately was tried six times for a murder that he did not commit, was the victim of especially egregious efforts by the prosecution. Two jailhouse informants testified against him in his first trial. One of those informants then recanted. At his third trial, another jailhouse informant testified that Flowers had confessed. That witness also later recanted, as did several other witnesses in the case.[221] And of course, in Walter Ogrod’s case, it was the emergence of two snitches—one of whom was a serial informer—after the first trial had ended in mistrial that led to his conviction and death sentence.[222]
D. Claims of No Expectation of Reward
Finally, a jailhouse informant’s claim that they are testifying without any expectation of reward, while common in the wrongful conviction cases studied, is inherently suspect and should also be treated as a red flag. Such claims, all of which proved false, were numerous in the dataset. In the Juwan Deering case, for instance, three jailhouse informants testified, two of whom asserted that Deering had confessed to starting a fire that killed five children and claimed that they did not receive any benefits for testifying against Deering.[223] A police detective also testified that “none of the informants received any benefits,” and were not promised anything by the detectives, and the prosecutor reiterated the same claim during closing arguments.[224] In fact, a special prosecutor later appointed to investigate the case issued a report finding, inter alia, that all three informants received substantial benefits in exchange for their testimony.[225] In Barry Williams’s case, both the jailhouse informant and the prosecutor told the jury that there was no deal for his testimony. In fact, the jailhouse informant was “given probation on the armed robbery charge, relocated, and given an undetermined amount of cash.”[226] In Marvin Cotton’s case, the jailhouse informant denied that he had been either threatened or promised any benefit in exchange for his testimony. Rather, he explained to the jury that he came forward “[b]ecause . . . one of the guys who was in the bullpen was the wrong guy and no sense [of] him going down for something somebody else did.”[227] The informant later recanted and admitted that both the police detective and the prosecutor had promised that he would be released early on his one‑year sentence and a probation revocation charge would be dropped, and that he did in fact receive those benefits.[228]
Although there might be an occasional jailhouse informant who comes forward without any expectation of reward, those instances are almost certainly rare. Jailhouse informants face substantial reputational costs from working as snitches, and their claims to do so out of the goodness of their hearts will almost always prove false. An assertion of no expectation of reward, therefore, in most cases evidences the informant’s willingness to say whatever they deem helpful to the State’s case regardless of veracity. It also indicates that prosecutors (who may be tolerating or inducing potentially false testimony) may be complicit in potential deception. Such dubious testimony provides a compelling basis for additional discovery from the State to allow defendants to scrutinize the truth of the implausible assertion and assess other aspects of the State’s investigation which may share similar tendencies to shade the truth.
Recanting witnesses, serial informants, and informants that appear at the eleventh hour along with jailhouse informants who claim no expectation of reward are all frequently recurring features of cases involving jailhouse informants. All are critical red flags that should tip off judges and defense counsel that, at minimum, careful scrutiny of the State’s use of informants and possibly other aspects of the case, is warranted.
V. Policy Responses
The only sure way to prevent wrongful convictions stemming from false jailhouse informant testimony is to ban the use of such testimony altogether. The argument for doing so is straightforward: Jailhouse informant testimony is inherently unreliable. While some informants may testify truthfully, many do not, and there is virtually no way to distinguish truthful from false testimony. In the same way that coerced confessions have long been viewed as inherently untrustworthy and thus inadmissible (even though some coerced confessions may be true), jailhouse informant testimony should be flatly barred from use.[229] Although there is a compelling constitutional case for a ban, the United States. Supreme Court has thus far refused to take this path.[230] Accordingly, a more incremental approach to regulating use of jailhouse informants will likely have a greater chance of success. Fortunately, reformers in numerous states have already begun taking major strides in this direction.
The red flags identified above, along with the other findings from this study, suggest several such avenues that reformers may wish to pursue in future efforts to reduce the risk of wrongful convictions. Because the data suggest a high prevalence of police and prosecutorial misconduct in cases involving jailhouse informants—at least among those cases eventuating in exoneration—any reforms aimed at improving the accuracy of jailhouse informant testimony must be crafted to dissuade and detect such misconduct. The reforms already adopted by leading reform states will help address these issues, but additional reform efforts are needed if substantial progress is to be made.
This Section, discusses several reform initiatives that might respond to the red flags identified above. It begins by reviewing initiatives that reform‑minded jurisdictions have already implemented. These include new recordkeeping and disclosure requirements regarding jailhouse informant usage, expanded discovery obligations, corroboration requirements, and administrative supervision of informant use. While each of these initiatives is useful and will help mitigate the problems associated with jailhouse informant use, the problems identified in this study suggest that even more far‑reaching reforms are necessary.
The Article thus proposes three additional types of reform. First, because the root problems identified here concern transactions that take place in black box environments, more aggressive efforts to increase transparency are necessary to deter the kinds of misconduct associated with the cases under review. To this end, the Article proposes the creation of legal databases that track misconduct by police and prosecutors. Importantly, these databases should record any allegations related to evidentiary reliability of any kind, including but not limited to problematic or contested confessions obtained by investigating detectives.
In addition to enhanced tracking of police and prosecutorial misconduct, greater transparency with respect to all aspects of prosecution‑informant interactions is critically important. Transparency is essential to combat the kinds of subterfuge and deception that enables wrongful convictions.
Second, even if abolition of jailhouse informants is not a politically viable option, more restrictive limitations on the use of informant evidence might be feasible. The Article proposes several such limits that would help limit use of jailhouse informants in situations most likely to generate unreliable convictions.
Third, the Article proposes reforming post‑conviction procedures in a way that would make it easier for individuals who were wrongfully convicted because of false testimony from incentivized witnesses, including but not limited to jailhouse informants, to obtain post‑conviction review.
A. Current Reform Initiatives
Efforts to more closely regulate the use of jailhouse informants, led by states including Texas, Illinois, and Connecticut, are currently underway. These efforts are long overdue and should help to dramatically reduce the contribution of jailhouse informant testimony to wrongful convictions. Hopefully, other states will follow their lead. The question raised in this Section is whether these current reform efforts go far enough given what the data reveals about the dynamics of jailhouse informant‑related wrongful convictions. Although a full review of the panoply of currently implemented reform measures is beyond the scope of this Article, it is worth considering what successful implementation of some of these measures can be expected to accomplish, and where further policy reforms might still be needed.
1. Informant Tracking Requirements
Several states have adopted tracking requirements that obligate prosecutors to document important facts related to use of jailhouse informants. Texas, for example, requires prosecutors to track any “use of testimony” made by jailhouse informants, regardless of whether that testimony was presented at trial, and any benefits “offered or provided to” such witnesses.[231] The effort to track testimony given by informants is essential to helping law enforcement officials, defense counsel, and judges identify repeat or serial jailhouse informants. As noted above, serial informants have proven to be a significant source of false testimony and are a major contributor to wrongful convictions. The regular use of serial informants is also correlated with police and prosecutor misconduct. Law enforcement officers and prosecutors can use this information to vet claims brought to them by potential snitches looking for benefits. Defense counsel can benefit by using the information to more effectively impeach jailhouse informant witnesses. Defense counsel can also investigate an informant’s prior testimony. Indications that an informant provided false or unreliable information in other cases would likely have a significant impact on juror credibility determinations.
Texas’s tracking requirements, however, fall short in at least one major way: The Texas statute does not require that the information tracked be made easily accessible by, for instance, adding it to a database that might be accessed by defense counsel and prosecutors in other jurisdictions.[232] Some jurisdictions in Texas, as well as a variety of other states, have mandated maintenance of databases to house such information and make it available. Nebraska, for example, requires prosecutors to “maintain a searchable record” of such information.[233] Connecticut requires that the information be provided to a central state agency that will maintain “a state‑wide record of such materials.”[234]
2. Enhanced Discovery Obligations
Texas and other jurisdictions also now mandate disclosure of basic facts necessary for cross‑examination and impeachment of jailhouse informant witnesses.[235] This includes the informant’s “complete criminal history, including any charges that were dismissed or reduced as part of a plea bargain,” benefits offered or provided in exchange for testimony, and information concerning other cases in which the informant testified or offered to testify against another inmate and any benefit offered or provided in exchange.[236] Other states have instituted similar disclosure mandates, in some cases expanding the scope of mandatory disclosure to include more specific information related to contacts with law enforcement or concerning recantations that might have been made by the informant.[237]
As the cases in this study make clear, disclosure failures are a regular feature of wrongful convictions involving jailhouse informants. Complete, timely disclosure of the information mandated by these new rules will provide important safeguards to defendants and enhance the jury’s ability to make accurate credibility assessments of jailhouse informant testimony. Certainly, full disclosure of the inducements offered or agreed to in exchange for an informant’s testimony will short‑circuit the ability of informants to falsely portray their testimony as motivated by benevolent intent when, in truth, they are expecting valuable benefits.[238] It will also limit the ability of prosecutors to vouch for a jailhouse informant’s credibility by painting an untruthful portrait of the jailhouse informant’s motives.
It should be acknowledged that mandatory disclosure rules, while valuable, provide only modest gains over current practice. After all, Brady and Giglio v. United States[239] already established that prosecutors must produce most of the material that is mandated for disclosure. Certainly, information about benefits provided to a testifying witness or charges that were dismissed or reduced as part of a plea bargain in exchange for testimony fall within the heartland of a prosecutor’s Brady obligations. Similarly, any evidence that the informant recanted their statement or story is highly material to the witness’s credibility and should be disclosed by the State.
Making disclosure of this information mandatory by rule, nonetheless, will still provide some benefits. First, categorical disclosure obligations should ensure that prosecutors cannot withhold exculpatory evidence based on judgments that the information is not “material” and thus not within Brady’s scope.[240] While prosecutors already have an ethical duty to disclose such material, statutory mandates provide teeth to those ethical duties.[241] Second, as several of the dataset cases illustrate, sometimes responsibility for an absence of effective cross‑examination or impeachment can be traced to inadequate representation by incompetent or overburdened defense counsel. Mandatory disclosure of the information needed to undertake a minimally competent cross‑examination of jailhouse informants removes at least one barrier to an effective defense and should ensure that counsel has the basic information at hand without necessitating independent investigation or burdensome discovery requests.
3. Modifications to Evidentiary Rules to Permit More Effective Witness Impeachment
Texas has also modified its rules of evidence to permit jailhouse informants to be impeached with “[e]vidence of a prior offense committed by a [jailhouse informant] . . . if the [informant] received a benefit . . . with respect to the offense, regardless of whether the [informant] was convicted of the offense.”[242] State evidence rules regarding witness impeachment based on prior convictions often do not encompass such prior offenses. This was the case in Texas prior to the enactment of Texas House Bill 34.[243]
The exchange of benefits for testimony defines the fundamental dynamic of informant practices. Reduction or dismissal of charges is one of the most powerful benefits that prosecutors can offer. But jurors simply cannot understand or appreciate the nature of jailhouse informant testimony unless they are informed of such benefits. Jurisdictions that seek to limit the ability of defense counsel to impeach informants with prior offenses dismissed in exchange for testimony certainly follow Texas’s lead in expanding the scope of impeachment.
However, by expressly limiting the impeachment rule to jailhouse informants, even the Texas rule does not go far enough. Rather, an expanded impeachment rule should apply to all incentivized witnesses, including eyewitnesses, fact witnesses generally, witnesses who testify to alleged confessions or admissions in non‑custodial settings, and accomplices and codefendants. The cases studied here demonstrate that while jailhouse informant testimony is highly unreliable and a major contributor to wrongful convictions, incentivized witnesses of all sorts seriously threaten the integrity of criminal trials. Disclosure and impeachment rules should be expanded to encompass all incentivized witnesses.
4. Corroboration Requirements
Uncorroborated testimony from accomplice witnesses has long been frowned upon, and many jurisdictions bar such testimony altogether. More recently, some jurisdictions have adopted corroboration rules applicable to jailhouse informant testimony as well. California was one of the earlier adopters of a corroboration requirement for jailhouse informants.[244] Other states, including Texas, have followed suit.
While any conviction based on a single uncorroborated piece of evidence is highly problematic,[245] the extreme unreliability of jailhouse informant testimony in particular makes any conviction based solely on such testimony factually worthless and should be flatly prohibited. Corroboration rules ensure that this is the case. However, while such rules provide a modicum of relief by barring conviction in cases in which the only substantive evidence comes from a jailhouse informant, in practice the rule rarely will provide much benefit to defendants. Even the most aggressive prosecutors are unlikely to go to trial with no other evidence that “connects the defendant with the crime,” which is all that the requirement requires.[246] Moreover, many corroboration statutes permit corroboration with other types of equally problematic evidence. California, for instance, allows a jailhouse informant’s testimony to be corroborated with testimony by an accomplice (even though both are subject to corroboration requirements), and in some instances, with the testimony of a second jailhouse informant as well.[247] Thus, at least on their face, (and as California courts have ruled) these statutes permit the testimony of one incentivized witness to be corroborated by the testimony of another incentivized witness.
In assessing the impact that corroboration rules might have had on the cases in the dataset, it seems clear that there would have been very little difference in the vast majority of cases. Indeed, there was no clear case in the dataset where a prosecutor attempted to take a case to a jury where jailhouse informant testimony was the only evidence that linked the defendant to the crime, and it would seem in general that such cases would be quite rare.
5. Pretrial Reliability Hearings
One of the most promising and potentially difference‑making reforms to get traction is the adoption of pretrial reliability hearings to allow a thorough vetting of informant testimony before it gets to a jury. Connecticut has enacted the most comprehensive set of procedures to date,[248] one that should serve as a model for other states. It requires, upon motion by a defendant, a pretrial judicial determination following a hearing regarding:
(1) The extent to which the jailhouse witness’s testimony is confirmed by other evidence;
(2) The specificity of the testimony;
(3) The extent to which the testimony contains details known only by the perpetrator of the alleged offense;
(4) The extent to which the details of the testimony could be obtained from a source other than the defendant; and
(5) The circumstances under which the jailhouse witness initially provided information supporting such testimony to . . . [law enforcement or] a prosecutorial official, including whether the jailhouse witness was responding to a leading question.[249]
A prosecutor’s failure to make at least “a prima facie showing that the jailhouse witness’s testimony is reliable” would result in its exclusion.[250] Illinois has adopted a similar pretrial reliability determination requirement.[251] Pretrial reliability hearings, such as those mandated in Connecticut and Illinois, are among the most promising avenues for reform and should become standard precautionary measures.[252] They provide an opportunity for the court to assess all the circumstances surrounding the informant’s emergence as a witness. The tracking and maintenance of relevant information in informant databases, and the timely disclosure of such information to counsel, will also make such hearings vastly more productive than they would otherwise be. The movement in reform‑minded states to adopt many of the reforms noted here represents a watershed moment for regulation of jailhouse informants and promises substantial improvement in reducing the number of wrongful convictions tied to their use. Moreover, the introduction of pretrial reliability hearings in particular, especially when combined with more stringent tracking and disclosure rules, could significantly impede dubious uncorroborated testimony from reaching juries.
Unlike any of the other reforms (except for the rarely applicable corroboration requirement), pretrial reliability hearings provide a direct mechanism to keep unreliable evidence out of a trial altogether. Given the growing body of persuasive evidence showing that jurors are generally ill‑equipped to properly discount jailhouse informant testimony even when presented with facts indicating the witness’s self‑interested motives, pretrial reliability hearings provide a mechanism to insulate jurors from prejudicial exposure to informant testimony. Trial judges who take their screening responsibility seriously can meaningfully reduce the chances that a defendant will be wrongfully convicted at trial as a result.
But the potential of even this promising reform measure is circumscribed. The two biggest weaknesses of pretrial reliability hearings are: (1) the paucity of evidence on which to make a determination that the informant lacks credibility, and (2) the need to rely on politically motivated, and often prosecutorial‑leaning, trial judges to devote more than de minimis attention to the issue. Both of these concerns are heightened by the general rule in American law that questions of a witness’s reliability or credibility go to the weight of the testimony, not its admissibility. Even with full disclosure of the facts and circumstances, it may be difficult to assess a jailhouse informant’s claim that the defendant made an inculpatory admission in his presence. It will simply be a question of one person’s word against another.
6. Administrative Supervision of Informant Use
Administrative reforms illustrate another promising approach to reigning in the most egregious instances of informant‑related misconduct. For example, on October 9, 2020, New Jersey’s attorney general issued a directive to all state enforcement and prosecuting agencies requiring compliance with a new set of procedures pertaining to the use of jailhouse informant witnesses.[253] The directive contains three main mandates. First, it requires all line prosecutors to obtain approval from the head of the county prosecutor’s office before any jailhouse informant can be called as a witness. Second, it requires line attorneys seeking permission to use jailhouse informant witnesses to provide essential information related to the credibility of a jailhouse informant, such as the informant’s criminal history, any benefits offered or provided in exchange for testimony, prior cases in which the jailhouse informant served as an informant for the state, any known prior recantations, and other information relevant to assessing credibility, to the county prosecutor for consideration in the decision whether to grant the request. Third, the directive requires county prosecutors’ offices to maintain records of jailhouse informants that have been approved to testify, including the information submitted as part of the review process. The information in these records will presumptively be available for discovery.
B. New Proposals
In addition to reform initiatives currently gaining traction around the country, the data from this study suggests the need for several additional types of reform.
1. Police Misconduct Databases
While tracking and disclosure requirements will enhance the abilities of police and prosecutors to vet potential informants more carefully and defense counsel to cross‑examine them more effectively, those requirements are predicated on good faith compliance by prosecutors and law enforcement. What the dataset in this study shows, however, is that wrongful convictions linked to jailhouse informants are most likely to occur when police and prosecutors are not acting in good faith. Not only do defendants and their lawyers need to know the alleged facts surrounding a jailhouse informant’s claims to have evidence, they need to know whether the prosecutors and law enforcement officers who produce such witnesses are themselves reliable and acting with integrity. One way to make it easier to do so is by ensuring that there is a central repository for allegations of misconduct by state officials and that legal actors have easy access to that information.
Some states, such as Massachusetts,[254] have already taken steps to maintain records of police and prosecutorial misconduct and to make such information available to judges and defense counsel.[255] Chicago and Washington, D.C. have also both established databases tracking police misconduct and provide models for more ambitious efforts.[256] Recognition of the significant link between misuse of jailhouse informants and official misconduct provides further support and motivation to make efforts to collect, maintain, and disseminate this information. Numerous scholars have called for enhanced transparency and accountability to combat the prevalence of police misconduct.[257] Public—especially defense bar—access to officer personnel files, public complaints, investigations, or findings of misconduct related to a detective or officer involved in a criminal investigation would greatly enhance the ability of defense counsel to function as an effective check on potential investigative misconduct. It would also enhance the ability of counsel to impeach law enforcement witnesses who normally receive the benefit of the doubt in the eyes of judges and jurors.
The present study also suggests that additional data about police evidence‑gathering methods would be tremendously helpful. In particular, given the strong correlation between false confessions and use of false jailhouse informant testimony, data on the success rates of interrogating police detectives in obtaining confessions, at least in serious cases such as homicide investigations, could help defense counsel and courts identify potentially problematic informant usage. Maintenance of data about how often, and under what alleged circumstances, detectives have secured confessions in the interrogation room would assist both in investigating the reliability of a particular confession (in light of the interrogating officer’s track record) and challenging similar types of evidence that is also prone to officer manipulation, such as jailhouse informant testimony. Such data could also help guide defense counsel’s discovery requests, factual investigation, and impeachment of investigating detectives. Obviously, data about individual detectives’ track records in past cases would not be dispositive evidence of misconduct or evidence of unreliability on its own, but it might help signal situations in which further investigative resources are required or warrant further discovery.
At minimum, the strong link between false confessions and false informant testimony provides a powerful argument for inclusion of allegations that a particular officer induced a false confession—even if not confirmed—in any centralized database. Indeed, the creation of a centralized repository of allegations linking individual police officers with any type of false evidence would provide a robust tool that would greatly enhance the legal system’s ability to identify and root out patterns of misconduct that lead to wrongful convictions. Such a tool would be of obvious help to defendants and their lawyers, but it would also be useful to assist judges in conducting reliability hearings and crafting effective jury instructions. It would also help responsible prosecutors make better decisions about which witnesses to present and what evidence to proffer at trial.
2. Expanding the Data Tracked in Informant Databases
As noted above, several initiatives already adopted by some states impose new duties on police and prosecutors to collect and retain information about informant usage. The data in this study, however, provides support for tracking an even richer set of facts relevant to the informant’s credibility. These include how and when the informant was identified and contacted, a precise account of the incriminating evidence or statements the informant claims to have obtained during the initial contact and how the initial account might have changed during subsequent contacts, the number of meetings with state officials, all requests for rewards or benefits by the informant, and, of course, any inducements requested, promised, or provided. Ideally, recordings and transcripts of all such meetings would be preserved. All such details should be maintained in the informant database, and the database should also track any changes in charges or sentences already imposed, as well as the disposition of charges or sentencing outcomes not yet resolved at the time of the informant’s testimony.
Not only is tracking of such information crucial for effective investigation and witness impeachment, but it would allow prosecutors, judges, and defense attorneys to identify serial snitches who pose especially great threats to the truth‑seeking mission of the criminal legal system. Prosecutors concerned with the integrity of their cases would have a far greater ability to assess the credibility of any purported informant. Defense attorneys, of course, would be better positioned to develop impeachment strategies that would effectively expose false informant testimony. Judges would also benefit from informant databases because they would improve their ability to make educated reliability assessments.
3. More Restrictive Limits on Informant Evidence
As noted above, Texas and several other states have introduced corroboration requirements on jailhouse informant testimony.[258] But to make corroboration rules more effective, it is also necessary to increase their rigor by imposing explicit limits on the kinds of evidence that can be used for purposes of corroboration.
This study provides several examples of why increased rigor is needed. Given the routine collaboration among inmates to concoct false stories about confessions made by fellow inmates, or to recruit other inmates to serve as false witnesses in an effort to obtain benefits through snitching, an effective corroboration requirement should not permit one jailhouse informant’s testimony to serve as corroboration for a second informant’s testimony. Corroboration needs to be based on truly independent evidence, preferably physical evidence that ties the defendant directly to the crime. Indeed, because the heightened reliability concerns are essentially the same, testimony from any incentivized witness—whether it be a jailhouse informant or a purported eyewitness—should be deemed equally insufficient to satisfy a corroboration requirement.
Beyond corroboration requirements, the data in this study also suggests a pressing need for strict prohibitions on jailhouse informants who emerge shortly prior to, or during, trial. Such witnesses are especially hard for trial counsel (who are gearing up for or already in the thick of trial) to properly investigate. Moreover, such late‑appearing witnesses seem to be a common feature of wrongful convictions involving jailhouse informants. Not only are such witnesses hard to investigate and impeach effectively, but they also are a symptom that prosecutors themselves view their cases as weak enough to require evidentiary reinforcement (long after the fact investigation should be complete). They are, in short, a clear sign that the State is moving forward with a case it is unsure it can prove, often for good reason. Similarly, and primarily for this latter reason, such witnesses should also be flatly barred from use in any retrial. Many self‑serving snitches take advantage of a prosecutor’s desperation in such circumstances to obtain outsized benefits by promising to provide evidence against the defendant. The temptation by prosecutors to cut corners and overlook reliability concerns in an effort to save face at a second trial creates far too great a risk that an unreliable witness—and a type of witness the data shows is often viewed as highly persuasive by jurors—will play a determinative role in the case outcome. A flat rule barring jailhouse informants at retrials would insulate prosecutors from succumbing to such temptations and would have prevented several known wrongful convictions from occurring had it been in place.
4. Post‑Conviction Procedural Reforms
Even when the error is ultimately rectified, any time that an innocent person is convicted of a crime they did not commit, a heartbreaking injustice has occurred. But that injustice is magnified many times over when rectification takes decades to come about. The exonerees in this dataset waited on average twenty‑one years before obtaining vindication. This is far longer than what most exonerees have experienced. Why this is the case is unknown. It may be that recognition of the contribution of jailhouse informants to wrongful convictions is only belatedly being fully recognized by the bar, and especially by the Conviction Integrity Units that have been integral to many of the exonerations studied here.
Regardless of the reasons for them, the extended delays between conviction and exoneration experienced by the exonerees vividly highlight that appellate and post‑conviction processes simply failed to provide timely or effective relief. While a full review and critique of post‑conviction review processes is well beyond the scope of this Article, the experiences of the individuals in the cases studied here suggest a few much‑needed changes.
First, the dataset cases provide strong support for expanding disclosure requirements. The reforms discussed above that are aimed at requiring prosecutors to track and disclose benefits provided to informants at or prior to delivery of their testimony are important. But it is equally important to ensure that disclosure mandates extend to benefits or inducements granted not only prior to, but also after an informant’s testimony is completed. After all, promises are often implied rather than express, and rewards are routinely withheld until after the defendant has been convicted and sentenced. Accordingly, prosecutors’ duty to track and disclose benefits provided to informants must include future benefits, especially any charges dismissed or not pursued as a result of the informant’s guilty plea, as well as any sentencing or other disposition recommendation made by the prosecutor with respect to the informant. Although post‑trial discovery of an implicit deal will generally not be as useful to a wrongfully convicted defendant as would pretrial disclosure of the deal, post‑conviction disclosure would at least enable defendants to utilize the information on appeal and in post‑conviction proceedings.
Indeed, disclosure rules could be significantly strengthened by requiring full disclosure of all present and future benefits, rewards, and inducements. Such a rule would prevent prosecutors from conducting their arrangements with informants with a wink and a nod in an effort to hide the essentially transactional nature of jailhouse informant testimony from jurors. Prosecutors should be required to identify the specific benefits that the informant can expect if the informant testifies on behalf of the prosecution and disclose those benefits to the defense and the jury. Any subsequent benefit bestowed on the informant that was not disclosed prior to the informant’s testimony would be treated as a Brady violation that defendants could assert in appellate or post‑conviction proceedings as constitutional error.
Further, the high prevalence of recantations by incentivized witnesses and the high rate at which they are ignored suggests that reforms in the appellate and post‑conviction review processes are needed. Such reforms should ensure that recantation evidence, which casts serious doubt on the reliability of convictions, is more effectively and promptly utilized in pursuit of innocence claims. When witnesses come forward and admit that they lied at trial—especially those who were incentivized by the State to testify—not only has the evidentiary basis of the conviction been weakened, but the integrity of the entire investigation and prosecution has necessarily been called into question. Cases in which a jailhouse informant or some other incentivized witness recanted, sometimes quite soon after trial had concluded, and that recantation remained a mere footnote in the litigation file for years and sometimes decades, were unfortunately common in the dataset. Had those recantations triggered prompter review, years of wrongful incarceration of innocent people could have been avoided.
Practically speaking, there are several ways that appellate and post‑conviction procedures can be modified to improve system responsiveness to recantation evidence. First, the recantation of an incentivized witness might trigger at least an initial review by a CRU, if one exists in the jurisdiction. CRUs have a wide mandate to investigate potential miscarriages of justice and should reinvestigate a conviction in the “interests of justice” when confronted with evidence that an incentivized witness has admitted to perjuring themselves at trial.[259] In assessing the weight of such evidence, CRUs might distinguish between recantations that admit or allege misconduct or wrongdoing by state actors, and those that simply admit making false statements. Allegations of misconduct, if true, undermine the legitimacy of the conviction and should always warrant relief, particularly in light of the evidence presented in this study which shows the frequent link between informant use and official misconduct of various sorts.
Another potential response to recantation evidence would be to prompt appointment of counsel for purposes of undertaking a thorough investigation of the case. Although criminal defendants are constitutionally entitled to counsel during appellate proceedings, that right does not carry over to post‑conviction review.[260] Many—indeed most—persons pursuing post‑conviction review are thus forced to proceed pro se.[261] A rule—be it statutory or constitutional—that entitled criminal defendants to a lawyer following the recantation of an incentivized witness would be a significant boon to those innocent persons who were wrongly convicted and forced to attempt to vindicate themselves without outside legal assistance. Such a rule would also be consistent with the project, favored by some, of redeploying habeas corpus resources more towards the protection of the innocent.[262]
A third possible response might be to modify post‑conviction review processes to permit appellate courts to throw out convictions in which incentivized witnesses testified and later recanted. California, for example, authorizes courts to grant new trials where the convictions were secured with “false evidence.”[263] Perjured trial testimony fits squarely within that definition.[264] Given that we know jurors have a strong tendency to accept secondary confessions as powerful evidence of guilt, an even-minimally plausible recantation by an incentivized witness should be sufficient to warrant the grant of post‑conviction relief.
Finally, even in the absence of any formal changes to appellate or post‑conviction process, the innocence bar should weigh recantation evidence more heavily than it now does; indeed, all lawyers—prosecutors, judges, and the innocence bar—should take notice when an incentivized witness recants. A recantation made by an incentivized witness is not a minor footnote to a case. It is a clap of thunder indicating that something has seriously gone wrong. A searching review of the conviction is always warranted in such circumstances.
Conclusion
Thanks to the exoneration cases, we have a far better understanding than ever before of how the use of jailhouse informants contributes to wrongful convictions. Study of this dataset gives us a fine‑grained look at the many ways in which informant testimony is induced and fabricated, and reveals the frequent efforts of law enforcement officials to hide the mechanics by which false convictions are manufactured. Indeed, the most egregious cases show that in some areas, and especially during the tenure of wayward or unprincipled administrative leadership, criminal justice processes can be compromised by cultures that tolerate or embrace high levels of corruption. Hopefully, such instances of widespread misconduct or corruption are discrete and limited, and wider complicity is unrepresentative of the norm.
Regardless, there is now abundant evidence that the use of jailhouse informants—and indeed, a wide range of incentivized witnesses—is responsible for shocking miscarriages of justice. There is no doubt that by the time this Article is published, more wrongful convictions predicated on false jailhouse informant testimony will have been uncovered. It is past time to adopt more aggressive efforts to shut this practice down or, at least, mitigate its proven danger. States such as Texas, New Jersey, Illinois, and Connecticut have taken critical first steps in this direction. The accumulation of exoneration data shows that more must be done.
Appendix: Table of Jailhouse Informant‑Related Wrongful Convictions Leading to Exonerations (Jan. 1, 2020 to Mar. 1, 2024)
| Last Name | First Name | Race | Year Convicted | State | FC[265]†? | OM[266]‡? |
|---|---|---|---|---|---|---|
| Alvarez | Ramon | Hispanic | 2012 | CA | OM | |
| Bell | George | Black | 1999 | NY | FC | OM |
| Birch | Ralph | White | 1989 | CT | OM | |
| Caldwell | Clifton | White | 2002 | MA | OM | |
| Carnevale | Daniel | White | 2007 | PA | OM | |
| Champagne | Raymond | White | 1979 | MA | OM | |
| Christian | Kino | Black | 2009 | MI | OM | |
| Coones | Olin | White | 2009 | KS | OM | |
| Cotton | Marvin | Black | 2001 | MI | OM | |
| Davis | Ricky | White | 2005 | CA | OM | |
| Davis, Jr. | Keith | Black | 2017 | MD | OM | |
| Debelbot | Ashley | Black | 2009 | GA | OM | |
| Debelbot | Albert | Other | 2009 | GA | OM | |
| Deering | Juwan | Black | 2006 | MI | OM | |
| DuBoise | Robert | White | 1985 | FL | OM | |
| Edwards | Dartanion | Black | 2009 | MI | OM | |
| Edwards | Joshun | Black | 2009 | MI | OM | |
| Faulkner | David | White | 2001 | MD | OM | |
| Flowers | Curtis | Black | 1997 | MS | OM | |
| Garrett | Paul Shane | White | 2003 | TN | FC | OM |
| Gold‑Smith | Robert | White | 2016 | IL | ||
| Hamilton | Lacino | Black | 1995 | MI | OM | |
| Harris | Lee | Black | 1992 | IL | FC | OM |
| Harris III | James | Black | 2015 | NJ | ||
| Hill | Taron | Black | 2006 | NJ | ||
| Hinton | C’Quan | Black | 2009 | MI | OM | |
| Howard | Bernard | Black | 1995 | MI | FC | OM |
| Inman | Devonia | Black | 2001 | GA | OM | |
| Jimenez | Ricardo | Hispanic | 2007 | NY | OM | |
| Johnson | Gary | Black | 1999 | NY | FC | OM |
| Johnson | Lamar | Black | 1995 | MO | FC | OM |
| Johnson | Jesse | Black | 2004 | OR | FC | OM |
| Johnson | Kareem | Black | 2007 | PA | ||
| LaPena | Frank | White | 1977 | NV | ||
| Legion | Anthony | Black | 2001 | MI | OM | |
| Lively | Charles Jason | White | 2006 | WV | ||
| Lynch | Renay | Black | 1998 | NY | FC | OM |
| Malik | Thomas | Black | 1996 | NY | FC | OM |
| Mangum | Verle | White | 2003 | CO | FC | OM |
| Mendoza | Madeline | Hispanic | 1993 | IL | FC[267]* | OM |
| Mulero | Marilyn | Hispanic | 1993 | IL | FC | OM |
| Mulherin | Crystal | White | 2022 | MI | FC | |
| Nixon | Kenneth | Black | 2005 | MI | ||
| Ogrod | Walter | White | 1996 | PA | FC | OM |
| Purnell | Mark | Black | 2008 | DE | OM | |
| Rhynes | Michael | Black | 1986 | NY | OM | |
| Schweitzer | Albert Ian | Other | 2000 | HI | ||
| Smith | Paul | White | 2010 | CA | OM | |
| Smith, Jr. | Larry | Black | 1994 | MI | FC | OM |
| Walker | John | Black | 1977 | NY | OM | |
| Ward | Ramon | Black | 1995 | MI | FC | OM |
| Watkins | Joseph | White | 2001 | GA | OM | |
| Williams | Barry | Black | 1986 | CA | OM | |
| Wilson | Jackie | Black | 1983 | IL | FC | OM |
* Professor, Georgia State University. J.D. Yale Law School. Thanks to Nirej Sekhon, Caren Morrison, and participants in the GSU faculty retreat for their thoughtful comments on earlier versions of this paper. The case descriptions and summaries that provide much of the source material in this Article were originally compiled by Maurice Possley and Ken Otterbourg at the National Registry of Exonerations. I owe them and the many contributors to the NRE a debt of gratitude for their incredibly important work.
- See Timothy Cole Exoneration Rev. Comm’n, Timothy Cole Exoneration Rev. Comm’n Rep. (Dec. 9, 2016) [hereinafter Timothy Cole Exoneration Report], https://www.txcourts.gov/media/1436589/tcerc-final-report-december-9-2016.pdf [https://perma.cc/EYW4-JDDH]. ↑
- Id. ↑
- See id. ↑
- See Tex. Code Crim. Proc. Ann. art. 39.14(h‑1) (West 2017). ↑
- See id. at art. 39.14(b). ↑
- Luke G. Allen, Lies Behind Bars: An Analysis of the Problematic Reliance on Jailhouse Informant Testimony in the Criminal Justice System and a Texas‑Sized Attempt to Address the Issue, 98 Wash. U. L. Rev. 257, 270 (2020). See also Jolie McCullough & Justin Dehn, How Some See Texas as the “Gold Standard” Against Wrongful Convictions, Tex. Trib. (Sept. 20, 2017), https://www.texastribune.org/2017/09/20/texas-lawmakers-hope-prevent-wrongful-convictions [https://perma.cc/NT6Y-8DJV]. See Tex. Code Crim. Proc. § 38.075(c)(2025). ↑
- Texas is one of several states that have enacted some type of jailhouse informant reforms. See Peggy M. Tobolowsky, Preventing and Addressing Wrongful Convictions: The Texas Approach, 54 No. 5 Crim. L. Bull. 1, 26 n.70 (citing Timothy Cole Exoneration Report at 14). ↑
- Nick Muscavage, NJ Latest State to Probe Use of Jailhouse Informants, Law360 (Feb. 25, 2022), https://www.law360.com/articles/1462987/nj-latest-state-to-probe-use-of-jailhouse-informants [https://perma.cc/6C6Z-XLE8]. ↑
- See id. The directive “instruct[ed] prosecutors to meet a series of standards when relying on jailhouse informants. Under the directive, prosecutors must seek approval from the county prosecutor before using a jailhouse informant by providing the informants’ criminal history, the benefits they would receive for their testimony, and known recantations. The directive also instructs prosecutors to maintain a database of approved informants.” Id. ↑
- Id. ↑
- Id. ↑
- See 725 Ill. Comp. Stat. Ann. 5/115‑21 (West 2019) (requiring pretrial reliability hearings in cases involving charges of murder, sexual assault, and aggravated arson). Connecticut’s reforms are among the most far‑reaching. See Dave Collins, Lying Prisoners: New Laws Crack Down on Jailhouse Informants, AP News (Sept. 14, 2019), https://apnews.com/article/connecticut-ned-lamont-9f8858ef3fbf4965874d314ce41ec69c [https://perma.cc/52P8-9MV5] (including pretrial reliability hearings and discovery requirements). Illinois is also mandating pretrial reliability hearings. Id. Some state courts have held that pretrial reliability hearings are necessary before a jailhouse informant’s testimony will be deemed admissible. See, e.g., D’Agostino v. State, 823 P.2d 283 (Nev. 1992). Additional states, including Michigan, are considering adopting such reforms as well. See Elisha Anderson, Michigan Jailhouse Informant Testimony Has Led to Wrongful Convictions. New Bill May Help, Detroit Free Press (July 24, 2022), https://plus.lexis.com/api/permalink/5df2c3e4-c3c2-47a9-9a97-529f97775ae5/?context=1530671 [https://perma.cc/NK4X-DTB2] (discussing “new bill [that] seeks to put safeguards in place by creating a statewide system in Michigan to track when jailhouse informants testify and benefits they receive in exchange, helping to ensure that prosecutors and defense attorneys have information relevant to the informant’s credibility and defendants get fair trials.”). ↑
- See infra Part I. ↑
- See infra Part I. ↑
- See infra Part I. ↑
- See infra Part I. ↑
- See infra Part I. ↑
- See infra Part I. ↑
- The NRE is a joint endeavor of the University of California Irvine Newkirk Center for Science & Society, the University of Michigan Law School, and Michigan State University College of Law. The NRE provides in‑depth analysis of the factors that contributed to the wrongful convictions that eventuated in exonerations and a database that can be filtered for a wide number of characteristics and contributing factors. See National Registry of Exonerations, Univ. of Mich. L. Sch., https://www.law.umich.edu/special/exoneration/Pages/detaillist.aspx [https://perma.cc/J36M-JFYU]. ↑
- Id. The cases are identified in an Appendix to this Article. The NRE’s database continues to grow at a steady rate. ↑
- Id. Number current as of March 1, 2024. ↑
- See id. These figures are based on a filter of NRE data for jailhouse informant cases, and then sorting by date. The rate of discovery of wrongful convictions in general has been growing. Through March 1, 2024, 724 people were added to the Registry since Jan. 1, 2020—a total representing 20.8 percent of all recorded exonerations, thus indicating that the proportion of exonerations for jailhouse informant‑related wrongful convictions among wrongful convictions stemming from other causes has not changed even while the discovery of such wrongful convictions has grown. ↑
- Chief among the reasons for progress has been the path‑breaking work of innocence organizations like the Innocence Project and the Center for Wrongful Convictions, pro bono work by innocence lawyers across the country, as well as an explosion of popular media content focusing on wrongful convictions, including the Netflix series Making a Murderer and the American Public Radio podcast Serial. There has also been an explosion of innocence scholarship focused on all aspects of wrongful convictions, including but certainly not limited to those caused by jailhouse informants. See, e.g., Alexandra Natapoff, Snitching: Criminal Informants and the Erosion of American Justice (2d ed. 2022). ↑
- These and similar assessments of the data in the NRE were generated by using the filter tools available on the NRE website. See generally National Registry of Exonerations, supra note 19. ↑
- Id. No‑crime wrongful convictions are, unfortunately, a fairly common occurrence. See Jessica S. Henry, Smoke but No Fire: When Innocent People Are Wrongly Convicted of Crimes That Never Happened, 55 Am. Crim. L. Rev. 665 (2018). ↑
- National Registry of Exonerations, supra note 19. The use of jailhouse informants to advance racist agendas to subject Black defendants to the death penalty is another way in which the link between jailhouse informants and prosecutorial misconduct has been manifested. Disparate racial impacts are also apparent in the dataset studied here, as more than half of the exonerees were Black. For a discussion, see Sidney Balman, Ensuring Black Lives Matter When the Penalty Is Death, 15 Idaho Critical Legal Stud. J. 1, 27 (2022) who noted that “more than a dozen defendants were sentenced to death in Los Angeles County” on the basis of jailhouse informant testimony in L.A. County in the 1970s and 1980s. ↑
- This is based on data from the NRE database as of March 1, 2024 (2,088 cases tagged as involving official misconduct out of 3,481 total exonerations). See National Registry of Exonerations, supra note 19. ↑
- One of the worst such examples involved Chicago police detectives Reynaldo Guevara and Ernest Halvorsen, who were jointly responsible for thirty‑nine documented wrongful convictions. See National Registry of Exonerations, National Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Documents/Guevara_Links.pdf [https://perma.cc/PW67-VMLH]. As discussed below, findings that a group of New York detectives were engaged in a wide range of misconduct that led to several exonerations triggered a review of more than seventy cases. See infra text accompanying notes 38−45. ↑
- See Russell D. Covey, Suspect Evidence and Coalmine Canaries, 55 Am. Crim. L. Rev. 537, 539 (2018) [hereinafter Covey, Suspect Evidence and Coalmine Canaries]. For other cases in the data set, see Ken Otterbourg, Clifton Caldwell, Nat’l Registry of Exonerations (May 30, 2023) [hereinafter Otterbourg, Clifton Caldwell], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6589 [https://perma.cc/LPD8-8MLZ] (describing prosecutorial misconduct of many types, including making prejudicial statements to the jury and failing to disclose exculpatory evidence); Ken Otterbourg, Lamar Johnson, Nat’l Registry of Exonerations [hereinafter Otterbourg, Lamar Johnson], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6547 [https://perma.cc/36JX-ZB2E] (last updated Jan. 18, 2024). Detective Jackson, who handled the snitch, “pled guilty in 2010 to stealing money from a person he arrested and received a sentence of 18 months in federal prison,” and prosecutors “presented false evidence at trial by not correcting [jailhouse informant] Mock’s statements about never testifying in a criminal case and the extent of his criminal record.” Id. Maurice Possley, Madeline Mendoza, Nat’l Registry of Exonerations [hereinafter Possley, Madeline Mendoza], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6519 [https://perma.cc/8L99-PR9T] ](last updated July 9, 2024) (noting that detective involved in investigation of Mendoza, who was innocent, was involved in dozens of other cases that also resulted in exonerations). Disgraced Chicago police detective Jon Burge used the testimony of a jailhouse informant while defending himself against charges of torturing a criminal defendant. See Maurice Possley, Jackie Wilson, Nat’l Registry of Exonerations (Oct. 16, 2020) [hereinafter Possley, Jackie Wilson], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5818 [https://perma.cc/72XV-Q89A] (describing how a witness at Burge’s trial testified that the victim “Wilson told him that he had inflicted his injuries on himself and that they were not the result of torture”). The prosecutors then used Coleman is a jailhouse informant against Jackie Wilson at Wilson’s retrial. Id. ↑
- See Ken Otterbourg, Thomas Malik, Nat’l Registry of Exonerations [hereinafter Otterbourg, Thomas Malik], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6350 [https://perma.cc/7XCP-JBMK] (last updated Dec. 12, 2023). ↑
- Id. ↑
- Frances Robles & N. R. Kleinfield, Review of 50 Brooklyn Murder Cases Ordered, N.Y. Times (May 11, 2013), http://www.nytimes.com/2013/05/12/nyregion/doubts-about-detective-haunt-50-murder-cases.html [https://perma.cc/2EAU-M254]. ↑
- Id. ↑
- Id. ↑
- Id. ↑
- Id. ↑
- Id. ↑
- Id. ↑
- Id. ↑
- Otterbourg, Thomas Malik, supra note 30 (noting that an investigation conducted by The New York Times found that the district attorney’s office investigated Detective Scarcella ‘and found misconduct in more than seventy cases). ↑
- Robles & Kleinfield, supra note 38. ↑
- Otterbourg, Thomas Malik, supra note 30. ↑
- Id. ↑
- See id. The CRU also found that following Malik’s trial, Shabazz had continued to work as an informant in numerous other cases, providing such unreliable testimony in them that “a judge in Manhattan issued a permanent injunction . . . barring Shabazz from contacting any law‑enforcement agency regarding matters outside his own cases.” Id. ↑
- See, e.g., Lee O. Sanderlin & Alex Mann, Baltimore State’s Attorney Drops Keith Davis Jr.’s Criminal Cases, Balt. Sun, Jan. 14, 2023, at A1 (“Deputy State’s Attorney Thomas Donnelly said the case’s prosecutorial history shows ‘blatant disregard’ for the law and for the rules of professional conduct.”). ↑
- Maya Yang, Baltimore Man Tried Four Times for Same Killing Sees All Charges Dropped, Guardian, (Jan. 16, 2023, 11:35 AM), https://www.theguardian.com/us-news/2023/jan/16/keith-davis-baltimore-man-trial-four-times-charges-dropped [https://perma.cc/SYF2-Z5BP]. ↑
- Id. ↑
- Maurice Possley, Keith Davis Jr., Nat’l Registry of Exonerations (Feb. 15, 2023) [hereinafter Possley, Keith Davis Jr.], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6543 [https://perma.cc/Q2AA-DFN8]. ↑
- See id. Over time, the facts of the case increasingly suggested misconduct by numerous state actors. The judge granted the defense’s first motion for a new trial, concluding that the state failed to disclose the jailhouse informant’s full criminal history. Two months before the third trial, the Baltimore Civilian Review Board found that the officers who shot Davis had used excessive force and found serious errors in the forensic testimony presented by the state at trial. The Board also found that there were “serious discrepancies” between the officers’ trial testimony and accounts they had earlier provided to internal investigations shortly after the shooting. The third trial ended in a mistrial. Davis was retried a fourth time and convicted, but that conviction too was thrown out after the Maryland Court of Appeals found error in the jury selection process. Prosecutors then charged Davis with attempted murder stemming from a prison fight. Davis’s lawyer claimed vindictive prosecution and filed a motion to dismiss the indictment. While that matter was pending, prosecutors geared up to try Davis a fifth time. Both cases were finally dismissed when a new State’s Attorney was elected. ↑
- CBS Baltimore Staff, Baltimore Prosecutors Dismiss All Charges Against Keith Davis Jr., Who Was Tried 4 Times for Murder, CBS News , https://www.cbsnews.com/baltimore/news/baltimore-prosecutor-dismiss-all-charges-against-keith-davis-jr-wjz-pimlico-security-guard [https://perma.cc/B3AZ-Q3UH] (last updated Jan. 16, 2023, 4:47 AM). ↑
- Id. ↑
- See, e.g., Samuel R. Gross & Michael Shaffer, Exonerations in the United States, 1989−2012, Nat’l Registry of Exonerations, 1, 57, (June 2012), https://www.law.umich.edu/special/exoneration/documents/exonerations_us_1989_2012_full_report.pdf [https://perma.cc/6DEH-4UF4] (15 percent); Brandon L. Garrett, Judging Innocence, 108 Colum. L. Rev. 55 (2008) (finding in a study of the first 200 DNA‑based exonerations, 16 percent included false confessions); Samuel R. Gross et al., Exonerations in the United States 1989 Through 2003, 95 J. Crim. L. & Criminology 523, 544 (2005). ↑
- See infra Appendix. Seventeen of the fifty‑four cases in the dataset also involved a documented false confession. ↑
- See Covey, Suspect Evidence and Coalmine Canaries, supra note 29, at 542 (noting that official misconduct in these cases takes a wide variety of forms, including: “coercing confessions, coaching witnesses to lie, failing to disclose exculpatory or impeachment evidence to defendants, or affirmatively lying in court”). The article also documented a significant correlation between use of known unreliable evidence, including but not limited to jailhouse informants, and other forms of misconduct, including failures to disclose exculpatory evidence as mandated by Brady v. Maryland, 373 U.S. 83 (1963). Id. ↑
- See generally Robert Moran, Philly Agrees to $9.1 Million Settlement for Man Exonerated in 1998 Slaying of 4‑Year‑Old Girl, Phila. Inquirer, Nov. 5, 2023, at B1. ↑
- See id. ↑
- See Maurice Possley, Walter Ogrod, Nat’l Registry of Exonerations [hereinafter Possley, Walter Ogrod], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5752 [https://perma.cc/Y4QZ-VPA4] (last updated Nov. 6, 2023). ↑
- See id. Among the confessions obtained by Devlin and Worrell were those of Anthony Wright, convicted of raping a seventy‑seven‑year‑old woman and exonerated in 2016; Shaurn Thomas and Clayton “Mustafa” Thomas, Jr., convicted for the 1990 murder of a seventy‑eight‑year‑old man—Shaurn was exonerated in 2017 and his brother Clayton in 2019; and Wille Veasy, convicted of a 1992 murder and exonerated in 2019. Maurice Possley, Shaurn Thomas, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5154 [https://perma.cc/E6NT-QJLA] (last updated Aug. 13, 2021). ↑
- Possley, Walter Ogrod, supra note 57. ↑
- Aaron Keller, Former Philadelphia Detectives Indicted for Lying in Case of Wrongly Convicted Rape and Murder Defendant: DA, Law & Crime (Aug. 13, 2021, 4:36 PM), https://lawandcrime.com/high-profile/former-philadelphia-detectives-indicted-for-lying-in-case-of-wrongly-convicted-rape-and-murder-defendant-da [https://perma.cc/ZZ4J-PNA] (exonerating Anthony Wright). ↑
- Moran, supra note 55. See, e.g., Maurice Possley, Renay Lynch, Nat’l Registry of Exonerations (Jan. 16, 2024) [hereinafter Possley, Renay Lynch], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6735 [https://perma.cc/X8BJ-73HQ]. The false confession‑jailhouse informant link appeared, yet again, in the prosecution of Renay Lynch. According to Lynch, her alleged confession was false and “based on information the detectives fed her during questioning.” Id. ↑
- See Maurice Possley, Bernard Howard, Nat’l Registry of Exonerations [hereinafter Possley, Bernard Howard], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5891 [https://perma.cc/W74X-TTQR] (last updated Feb. 18, 2022). ↑
- Anderson, supra note 12 (according to a police detective who testified at Twilley’s sentencing hearing (which was placed under seal), Twilley had provided crucial testimony in “numerous” homicide cases and had assisted police in ‘‘‘at least 20 cases”). ↑
- See Maurice Possley, Ramon Ward, Nat’l Registry of Exonerations [hereinafter Possley, Ramon Ward], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5694 [https://perma.cc/XGY4-B7V9] (last updated Mar. 28, 2022). ↑
- Possley, Bernard Howard, supra note 62. ↑
- Id. ↑
- Id. ↑
- Id. (discussing a prosecutor’s memo written during Howard’s trial that “detailed concerns about the [police] failure to disclose deals and promises, noting that doing so would require reversals of convictions”). ↑
- Id. ↑
- George Hunter, ‘You Have to Hang Your Hat on the Fact that You Didn’t Do It’; Detroit Man Formally Exonerated 26 Years Later in 3 Slayings, The Detroit News, Dec. 18, 2020, at A4. ↑
- Possley, Bernard Howard, supra note 62. See also Maurice Possley, Barry Williams, Nat’l Registry of Exonerations [hereinafter Possley, Barry Williams], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5946 [https://perma.cc/32AH-SWBG] (last updated May 23, 2024) (corrupt prosecutors). ↑
- See Possley, Jackie Wilson, supra note 29. ↑
- Id. ↑
- Id. ↑
- Id. ↑
- The practice dates at least to the 1980s in California. See Mirko Bagaric et al., Should We Flip on Flippers: A Rational Approach to Providing Penalty Reductions to Criminal Informers, 87 Tenn. L. Rev. 121, 147 (2019) (describing “cell block in the Los Angeles County Jail [that] was purposefully designed in the 1980s to give experienced informants, such as the one who tricked Lisker, easy access to ‘naive inmates’”). The findings of the L.A. County grand jury were presented at length in Report of the 1989−1990 Los Angeles Grand Jury: Investigation of Jail House Informants in the Criminal Justice System in Los Angeles County (June 26, 1990). ↑
- Maurice Possley, Paul Smith, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6403 [https://perma.cc/6FNM-TR8N] (last updated Oct. 13, 2022). ↑
- Salvador Hernandez, The Use of Jailhouse Snitches in Orange County Bungled at Least 57 Criminal Cases, Public Defender Says, L.A. Times (Oct. 4, 2023, 2:06 PM), https://www.latimes.com/california/story/2023-10-04/the-use-of-jailhouse-snitches-in-orange-county-bungled-at-least-57-criminal-cases-public-defender-says [https://perma.cc/N59T-B5KG]. ↑
- Id. ↑
- See Elizabeth Weill‑Greenberg & Jerry Iannelli, DOJ Finds Orange County Sheriff, DA Violated Civil Rights Using Illegal Jailhouse Informants, The Appeal (Oct. 13, 2022), https://theappeal.org/orange-county-jailhouse-informants-doj-investigation [https://perma.cc/B4EW-G54X]; U.S. Dep’t of Just., Civil Rights Div., Investigation of the Orange County District Attorney’s Office and the Orange County Sheriff’s Department (2022), https://www.justice.gov/media/1251036/dl [https://perma.cc/G6BS-P66B]. ↑
- See Maia Goodell, Government Responsibility for the Acts of Jailhouse Informants Under the Sixth Amendment, 101 Mich. L. Rev. 2525, 2530 n.37 (2003) (noting that, as reported in the case In re Benn, 952 P.2d 116, 138–39 (Wash. 1998), “[a] police officer in Washington testified that police sometimes illicitly placed ‘known snitches’ in the cells of pretrial defendants to obtain information”). Similar practices have come to light in Texas. See Transcript of Testimony of Ramiro Saldana at 27–41, Texas v. Padron, No. CR‑02‑004423‑F (Tex. Ct. Dist. Feb. 21, 2023) (recanting jailhouse informant describing jail “tank” that housed six inmates who were all working as snitches, and explaining that “the way it was all put together was so we could collaborate [sic] each other’s testimony”). ↑
- Stephen Bright has written some of the most damning criticism of this nation’s abject failure to live up to the promise of effective legal representation proffered in Gideon v. Wainwright, 372 U.S. 335 (1963). Bright has documented egregious failures by lawyers—drunk, sleeping through trials, or rushing defendants to quick pleas or, failing that, the shortest trials possible. He traces most of the system’s problems to failures to provide adequate resources to indigent defense. Because of low rates of compensation, competent attorneys often avoid taking appointed cases, and attorneys who do take such cases can only devote minimal time and energy to each individually because they can only make a decent income by handling large numbers of cases and disposing of them as quickly as possible. In jurisdictions that rely on public defenders offices to represent the indigent, consistent underfunding routinely leads to large caseloads that force even the most competent and dedicated defenders in the direction of a triage‑focused representational strategy. See Stephen B. Bright, Neither Equal Nor Just: The Rationing and Denial of Legal Services to the Poor When Life and Liberty Are at Stake, 1997 Ann. Surv. Am. L. 783, 816 (“The most fundamental reason for the poor quality or absence of legal services for the poor in the criminal justice system is the refusal of governments to allocate sufficient funds for indigent defense programs.”); Stephen B. Bright, Counsel for the Poor: The Death Sentence Not for the Worst Crime but for the Worst Lawyer, 103 Yale L. J. 1835, 1836 (1994) (documenting how “[p]oor people accused of capital crimes are often defended by lawyers who lack the skills, resources, and commitment to handle such serious matters”). ↑
- See, e.g., Maurice Possley, Crystal Mulherin, Nat’l Registry of Exonerations (Jan. 31, 2024), https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6747 [https://perma.cc/5ZQL-XUK5] (noting the trial lawyer failed to retain eyewitness identification expert to impeach alleged eyewitness). ↑
- Id. (noting that trial counsel failed to even attempt to impeach jailhouse informant). See also Maurice Possley, Daniel Carnevale, Nat’l Registry of Exonerations [hereinafter Possley, Daniel Carnevale], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5710 [https://perma.cc/LCT4-EACE] (last updated Feb. 28, 2022) (“Carnevale’s defense lawyer did not obtain an expert to review the ATF investigation of the blaze. No witnesses testified for the defense.”); Maurice Possley, Raymond Champagne, Nat’l Registry of Exonerations [hereinafter Possley, Raymond Champagne], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5772 [https://perma.cc/3N7K-9AAY] (last updated July 14, 2022) (noting the defense lawyer’s questions incentivized alleged eyewitness but failed to ask if he had any pending cases; later investigation revealed he had six pending cases at the time which he resolved without any added jailtime); Maurice Possley, Lacino Hamilton, Nat’l Registry of Exonerations [hereinafter Possley, Lacino Hamilton], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5828 [https://perma.cc/9TQP-SG5A] (last updated Jul. 27, 2021) (“Hamilton’s trial defense attorney, Robert Slameka, also represented three other men who were wrongfully convicted and exonerated–Marvin Cotton, Anthony Legion and Davontae Sanford.”); Possley, Bernard Howard, supra note 62 (“The petition also accused Howard’s trial defense lawyer of providing an inadequate legal defense by failing to call Tyiesha Washington as an alibi witness.”); Otterbourg, Lamar Johnson, supra note 29 (noting that the trial attorney failed to object to informant’s testimony that should have been excluded for lack of foundation). ↑
- Of the fifty‑four cases in the dataset, twenty‑seven‑half of the total sample were identified as involving inadequate legal representation. This is a far higher rate than what is present in the entire set of exoneration cases (977 out of 3,546—or 27.5 percent—as of July 2, 2024). The same holds true even if one looks only at homicide cases in the Registry. At the time of this writing, 1,423 of the 3,546 in the database involved homicide charges. Of those, 498—or 35 percent—of the cases involved documented inadequate legal defense. ↑
- “The danger that the investigators will go too far is magnified to the extent that the killing is brutal and horrifying and to the extent that it attracts public attention‑factors that also increase the likelihood that the murder will be treated as a capital case.” Samuel R. Gross, Lost Lives: Miscarriages of Justice in Capital Cases, 61 Law & Contemp. Probs. 125, 135 (1998). See also Myrna S. Raeder, See No Evil: Wrongful Convictions and the Prosecutorial Ethics of Offering Testimony by Jailhouse Informants and Dishonest Experts, 76 Fordham L. Rev. 1419, 1423−24 (2007) (“The public pressure on prosecutors has grown significantly in a world where news is 24/7, blogs are omnipresent, and commentators abound. . . . [I]t is unsurprising that prosecutors feel it necessary to solve major crimes quickly and publicly.”); William Lee Hon, Prosecuting Under Pressure, Tex. Dist. & Cnty. Att’ys Ass’n, https://www.tdcaa.com/journal/prosecuting-under-pressure [https://perma.cc/TLF8-NEJ5] (“Beyond the pressure from law enforcement and victims is the not‑infrequent pressure that the prosecutor feels from the public to obtain a particular result.”), both cited in Brief of The Innocence Project and Innocence Network As Amici Curiae in Support of Petitioner at 12, Hemphill v. New York, 143 S.Ct. 1082 (2023) (No. 22‑488), 2023 WL 174361 (arguing that especially in high profile cases substantial “public pressure to close a case will foster prosecutorial reliance on informant testimony”). ↑
- A recent study of fifty wrongful conviction cases found that one of the leading causal factors was an especially high profile or substantial media coverage of the case. See D. Kim Rossmo & Joycelyn M. Pollock, Confirmation Bias and Other Systemic Causes of Wrongful Convictions: A Sentinel Events Perspective, 11 Ne. U. L. Rev. 790, 805 (2019) (finding that one “common causal pattern consisted of a high‑profile crime (such as a horrible murder) that led to a rush to judgment (and a premature shift to a suspect‑based investigation), resulting in tunnel vision and confirmation bias—ultimately producing an evidence failure”). ↑
- Ogrod, as noted above, was tried for the killing of four‑year‑old Barbara Jean Horn. The murder sparked “massive media attention,” but remained unsolved for four years. It was not until Ogrod (who had not been a suspect in the case previously), was brought in as a possible witness and questioned for more than 14 hours, eventuating in the alleged confession, that Ogrod became the focus of the State’s case. Pressure on the state undoubtedly increased after the first trial ended in a mistrial with the jury voting 11‑1 to acquit. Possley, Walter Ogrod, supra note 57. ↑
- Maurice Possley, Lee Harris, Nat’l Registry of Exonerations [hereinafter Possley, Lee Harris], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6566 [https://perma.cc/6SVV-N4NJ] (last updated Feb. 4, 2024). ↑
- Id. The defense lawyers, Andrea Lyon and Shelby Keisman, said Harris was a scapegoat for the police’s failure to find the real killers. Keisman, in her opening statement, told the jury that a murder in the Gold Coast was different from most other murders in Chicago. “‘Gold Coast,’ Keisman said. ‘The name speaks for itself and the person to whom this happened came from a family of affluence, of prominence.’” Id. ↑
- Ken Otterbourg, Albert Ian Schweitzer, Nat’l Registry of Exonerations [hereinafter Otterbourg, Albert Ian Schweitzer], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6540 [https://perma.cc/SC5F-PD89] (last updated July 30, 2024). ↑
- Otterbourg, Thomas Malik, supra note 30; Robles & Kleinfield, supra note 32. In Malik’s case, major media coverage unsurprisingly arose after the horrific token booth firebombing in New York City. Id. Patterns of corruption also are common in certain types of cases in which cops and prosecutors have carved out a special dominion, in which there is little sympathy for defendants, and in which law enforcement officials might prioritize getting the defendant off the street over ensuring that the conviction is accurate. Gang cases in particular often fit these criteria. One such case in the dataset was that of Barry Williams, who was charged with the murder of an alleged member of the Crips gang. Prosecutors alleged that Williams “was a high‑ranking member of the Bloods.” Possley, Barry Williams, supra note 71. ↑
- Otterbourg, Clifton Caldwell, supra note 29. ↑
- Possley, Lacino Hamilton, supra note 84. ↑
- Id. ↑
- See Jonathan M. Golding et al., The Influence of Jailhouse Informant Testimony on Jury Deliberation, 28 Psych. Pub. Pol’y & L. 560, 560 (2022); see also Stacy Ann Wetmore et al., On the Power of Secondary Confession Evidence, 20 Psych., Crime & L. 339 (2014). ↑
- See Saul M. Kassin & Katherine Neumann, On the Power of Confession Evidence: An Experimental Test of the Fundamental Difference Hypothesis, 21 Law & Hum. Behav. 469, 471 (1997). ↑
- See Brown v. Mississippi, 297 U.S. 278, 287 (1936) (holding that convictions based on confessions induced by torture violated Due Process Clause of Fourteenth Amendment); Lyons v. Oklahoma, 322 U.S. 596, 605 (1944) (“A coerced confession is offensive to basic standards of justice, not because the victim has a legal grievance against the police, but because declarations procured by torture are not premises from which a civilized forum will infer guilt.”). More recently, the U.S. Supreme Court has shifted its interpretation of due process to focus the constitutional analysis on whether police misconduct was present in eliciting the confession. See Colorado v. Connelly, 479 U.S. 157, 164 (1986) (“Absent police conduct causally related to the confession, there is simply no basis for concluding that any state actor has deprived a criminal defendant of due process of law.”). State courts continue to exclude coerced confessions on the basis of their unreliability. See, e.g., State v. Graham, 768 P.2d 259, 265 (Kan. 1989) (overruled in part on other grounds) (“When a statement or confession is found to be involuntary it is excluded from evidence because it is unreliable and not trustworthy. A coerced confession is not based on knowledge and truth for it originates from coercion and a personal desire of the defendant to alleviate existing pressures.”). As Edward Rubin notes, coerced confessions “have been known for many centuries to provide unreliable information.” Edward Rubin, Criminal Injustice, 75 Vand. L. Rev. En Banc 1, 8 (2022). ↑
- See Saul M. Kassin & Holly Sukel, Coerced Confessions and the Jury: An Experimental Test of the “Harmless Error” Rule, 21 Law & Hum. Behav. 27, 31 (1997). ↑
- Id. at 43. ↑
- Jeffrey S. Neuschatz et al., The Effects of Accomplice Witnesses and Jailhouse Informants on Jury Decisionmaking, 32 Law & Hum. Behav. 137, 137 (2008). ↑
- Id. ↑
- See id. at 146. ↑
- See id. at 138. ↑
- Possley, Walter Ogrod, supra note 57. ↑
- Maurice Possley, Olin Coones, Nat’l Registry of Exonerations [hereinafter Possley, Olin Coones], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5866 [https://perma.cc/44NQ-FLHY] (last updated Nov. 27, 2022). See also Maurice Possley, Kareem Johnson, Nat’l Registry of Exonerations (May 2, 2020), https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5745 [https://perma.cc/4RGV-GJZP]. Jurors also fail to properly discount jailhouse informant testimony even when they are fully aware of the incentives to testify. Jailhouse informant “Younger admitted that he came forward in the hope of obtaining leniency and that in fact a life sentence for dealing drugs had been cut to 10 years. ‘I was trying to save my ass,’ he testified.” Id. ↑
- Otterbourg, Clifton Caldwell, supra note 29. ↑
- See Bruton v. U.S., 391 U.S. 123 (1968). ↑
- A judge who later reviewed Caldwell’s case noted that “[t]he inequity in the outcomes is not lost on the court.” Otterbourg, Clifton Caldwell, supra note 29. ↑
- See id. ↑
- See Baylee D. Jenkins et al., Testing the Forensic Confirmation Bias: How Jailhouse Informants Violate Evidentiary Independence, 38 J. Police & Crim. Psych. 93 (2021). ↑
- Id. at 102. ↑
- See Maurice Possley, George Bell, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5976 [https://perma.cc/2KQB-JD2E] (last updated Oct. 11, 2023). ↑
- Maurice Possley, Kino Christian, Nat’l Registry of Exonerations [hereinafter Possley, Kino Christian], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6528 [https://perma.cc/985C-Z9R7] (last updated May 23, 2023). ↑
- Possley, Barry Williams, supra note 71. ↑
- Id. (“I lied on the stand against Barry Williams because I didn’t want to do the 25 years,’ Cox said. ‘When the District Attorney offered to pay me and relocate me, I agreed to say what he wanted me to say.”). ↑
- See Possley, Kino Christian, supra note 114. ↑
- Otterbourg, Thomas Malik, supra note 30. ↑
- See Maurice Possley, Kenneth Nixon, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5931 [https://perma.cc/9ZRA-X28F] (last updated June 29, 2023); see also Possley, Renay Lynch, supra note 61 (noting the informant who recanted trial testimony explained that the relied on “media accounts” to generate false testimony); Maurice Possley, Michael Rhynes, Nat’l Registry of Exonerations (Jan. 23, 2024) [hereinafter Possley, Michael Rhynes], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6740 [https://perma.cc/Z3AB-S676] (noting the informant who falsely testified against Rhynes admitted on cross‑examination that he first became familiar with facts of case from news accounts). ↑
- See Ken Otterbourg, Mark Purnell, Nat’l Registry of Exonerations (June 6, 2022) [hereinafter Otterbourg, Mark Purnell], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6318 [https://perma.cc/MM5N-LSJK]. ↑
- In Barry Williams’s case, the jailhouse informant admitted that the prosecutor “and his people coached me and told me what to say on the stand . . . . They told me to look at Barry in the courtroom, but I couldn’t. I was lying and Barry knew it. I have never heard Barry Williams say that he was responsible for killing anyone.” Possley, Barry Williams, supra note 71. See also Possley, Ramon Ward, supra note 64 (noting that at a hearing challenging the admissibility of the jailhouse informant testimony a witness testified “who said he had testified in three such cases said police fed him details so that he could then testify that defendants had admitted their crimes during jailhouse conversations”). ↑
- Possley, Lacino Hamilton, supra note 84. ↑
- Id. ↑
- Id. ↑
- Possley, Lee Harris, supra note 89. ↑
- Id. ↑
- Id. Defense lawyers “Lyon and Keisman said Harris’s statements had been pieced together from media accounts and from tidbits fed to him by detectives eager to solve the case.” Id. See also Maurice Possley, Marvin Cotton, Nat’l Registry of Exonerations [hereinafter Possley, Marvin Cotton], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5839 [https://perma.cc/7EGE-LV5B] (last updated Dec. 12, 2023); Maurice Possley, Anthony Legion, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5838 [https://perma.cc/596R-2FTR] (last updated Mar. 28, 2022) (statements prepared by detectives). ↑
- Possley, Keith Davis Jr., supra note 48. ↑
- Id. ↑
- See, e.g., Possley, Michael Rhynes, supra note 119 (noting that a jailhouse informant falsely claimed that Rhynes confessed to him in the jail’s law library after asking him to “look up something concerning a robbery”); Possley, Keith Davis Jr., supra note 48. See also Maurice Possley, Robert Gold‑Smith, Nat’l Registry of Exonerations (Feb. 15, 2023) [hereinafter Possley, Robert Gold‑Smith], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6577 [https://perma.cc/GR4T-H6E6]. An inmate secretly tape recorded a “conversation” he alleged he had with Gold‑Smith, but it appears he only recorded himself and a voice purported to be Gold‑Smith’s was mere whispers. [he was wearing a wire when the recording was made. Security footage later proved that the alleged conversation never took place]. Id. Otterbourg, Lamar Johnson, supra note 29 (“Mock testified about what he said he heard in the jail. He said he heard a man from a nearby cell who identified himself as Johnson shouting, ‘they didn’t have the gun.’ The following day, Mock testified, that same man talked about a robbery and murder on the city’s south side. (The police could find no record of this robbery or murder.”)); Possley, Marvin Cotton, supra note 127 (“Frazier said he couldn’t see Cotton during the conversation because they were in different holding cells”; “Frazier claimed that he was in a holding cell with a man named ‘Anthony’ who was a co‑defendant of Cotton. He said that ‘Anthony’ introduced Frazier verbally to Cotton, who was in an adjacent holding cell, but could not be seen because the cells were partitioned off. Frazier claimed he saw Cotton when he left his cell to go to court.”); Maurice Possley, Ricky Davis, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5684 [https://perma.cc/VW6W-X2FD] (last updated July 3, 2024) (“After Davis and Dahl were charged with the crime in 2002, Davis’s then‑girlfriend was in custody on an unrelated charge. Kristin Webb, who was also in the jail at the time, testified at the trial that the girlfriend told her that Davis had admitted that he committed the murder. However, the girlfriend also testified and denied that Davis ever admitted to the killing and denied she ever said so to Webb.”); Possley, Lacino Hamilton, supra note 84 (“At 9:15 a.m. on the following day, Oliver Cowan, another prisoner in the detention area, claimed to a Detroit police detective that Hamilton had admitted that he killed Bias. Cowan said Hamilton ‘confessed’ in the jail three days before—even though Hamilton had been in custody for less than 24 hours.”). ↑
- In the Lacino Hamilton case, for instance, the jailhouse informant who testified against Hamilton reportedly told another inmate that “if he were willing to ‘help’ homicide detectives with some of their murder cases, the detectives would help him with his case. The man said Cowan told him he didn’t have to worry if he didn’t have any information; detectives would take care of that.” Possley, Lacino Hamilton, supra note 84. ↑
- Maurice Possley, Charles Jason Lively, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5817 [https://perma.cc/FT6E-5VD9] (last updated Mar. 11, 2024). ↑
- Id. ↑
- Otterbourg, Albert Ian Schweitzer, supra note 91. The second of two jailhouse informants who testified against Michael Rhynes later admitted that the other informant recruited him to testify against Rhynes, and that both the other informant and an investigator for the DA’s office told him the key facts and what to say. Possley, Michael Rhynes, supra note 119. ↑
- See Natapoff, supra note 23, at 29; Russell D. Covey, Abolishing Jailhouse Snitch Testimony, 49 Wake Forest L. Rev. 1375, 1380–81 (2014) [hereinafter Covey, Abolishing Jailhouse Snitch Testimony]. ↑
- See Natapoff, supra note 23, at 24. ↑
- This fact has been recognized by numerous courts. See, e.g., State v. Patterson, 886 A.2d 777, 789 (Conn. 2005) (noting that a “[jailhouse] informant who has been promised a benefit by the state in return for his or her testimony has a powerful incentive, fueled by self‑interest, to implicate falsely the accused. Consequently, the testimony of such an informant, like that of an accomplice, is inevitably suspect”); United States v. Bernal‑Obeso, 989 F.2d 331, 334 (9th Cir. 1993) (“Our judicial history is speckled with cases where informants falsely pointed the finger of guilt at suspects and defendants, creating the risk of sending innocent persons to prison.”). ↑
- See Possley, Barry Williams, supra note 71. ↑
- Ken Otterbourg, Robert DuBoise, Nat’l Registry of Exonerations [hereinafter Otterbourg, Robert DuBoise], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5807 [https://perma.cc/MUX7-8L24] (last updated Feb. 13, 2024). ↑
- Possley, Michael Rhynes, supra note 119. ↑
- See Possley, Robert Gold‑Smith, supra note 130. ↑
- Possley, Madeline Mendoza, supra note 29. ↑
- Id. ↑
- See id. ↑
- See Maurice Possley, Curtis Flowers, Nat’l Registry of Exonerations [hereinafter Possley, Curtis Flowers], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5808 [https://perma.cc/VTP4-3DWH] (last updated Sept. 4, 2021). ↑
- See, e.g., Possley, Barry Williams, supra note 71 (noting that despite trial testimony that there was no deal for his testimony, jailhouse informant “was later given probation on [an] armed robbery charge, relocated, and given an undetermined amount of cash”; another jailhouse informant in the case was given a variety of benefits, including a check for several hundred dollars for himself and several hundred dollars of groceries for his family); Possley, Robert Gold‑Smith, supra note 130 (noting that in the case of Robert Gold-Smith a jailhouse informant receiveed, inter alia, a $100 commissary credit and a $1000 Crime Stopper reward). ↑
- See Christopher T. Robertson & D. Alex Winkelman, Incentives, Lies, and Disclosure, 20 U. Pa. J. Const. L. 33, 83–84 (2017) (noting the prevalence of government use of incentives to induce witnesses of all types to testify and describing that practice as “legalized bribery”). ↑
- See, e.g., Possley, Barry Williams, supra note 71 (noting that a witness told either he serves time on the robbery or Williams serves time on the murder); Otterbourg, Mark Purnell, supra note 120 (noting that the jailhouse informant was also identified by an eyewitness as possibly the shooter in the case); Possley, Raymond Champagne, supra note 84 (noting that an alleged eyewitness to prison murder resolved six pending cases for no additional jailtime in exchange for testimony). ↑
- Otterbourg, Lamar Johnson, supra note 29. ↑
- Id. ↑
- Id. ↑
- Maurice Possley, David Faulkner, Nat’l Registry of Exonerations (Mar. 15, 2022) [hereinafter Possley, David Faulkner], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6178 [https://perma.cc/PKY4-6JGP]. ↑
- Possley, Raymond Champagne, supra note 8484. ↑
- Otterbourg, Robert DuBoise, supra note 139. ↑
- Id. Otterbourg details evidence provided by second witness (not a jailhouse informant) who also testified against DuBoise: “Andrusckiewiecz testified about what he said DuBoise had told him. Id. At the time of the trial, Andrusckiewiecz was also preparing to be a star witness in another murder trial, a case where there was the potential for him to be charged as an accessory. Id. He was working closely with the State Attorney’s Office on that case, and his emergence as a witness in the DuBoise trial came through prosecutors rather than a police investigation. Id. This was not disclosed at trial, and he was never charged in connection with that separate murder.” Id. ↑
- U.S Const. amend. VI. ↑
- See Brady v. Maryland, 373 U.S. 83 (1963). ↑
- See Covey, Suspect Evidence and Coalmine Canaries, supra note 29, at 549 (discussing empirical data indicating that potential Brady violations in particular, and prosecutorial misconduct claims in general, are more strongly correlated with use of jailhouse informants than with any other type of evidence). ↑
- See id. at 572 (noting that empirical research indicates that Brady violations are correlated more highly with testimony from jailhouse informants than any other form of evidence). ↑
- Otterbourg, Lamar Johnson, supra note 29. ↑
- Id. ↑
- Possley, Lacino Hamilton, supra note 84. ↑
- Possley, Lee Harris, supra note 89. ↑
- Possley, Walter Ogrod, supra note 57. ↑
- Possley, Keith Davis Jr., supra note 48. ↑
- Otterbourg, Lamar Johnson, supra note 29. ↑
- See Caleb Linton, Like Putting Lipstick on a Pig: Why the History of Crime Control Should Compel the Prohibition of Incentivized Witness Testimony Under Fundamental Fairness Principles, 113 J. Crim. L. & Criminology 391, 412−13 (2023) (citing Innocence Project materials urging adoption of the requirement). ↑
- Otterbourg, Clifton Caldwell, supra note 29. ↑
- Possley, Lee Harris, supra note 89. ↑
- Otterbourg, Lamar Johnson, supra note 29. ↑
- See, e.g., id. ↑
- Possley, Olin Coones, supra note 106. See also Maurice Possley, Juwan Deering, Nat’l Registry of Exonerations [hereinafter Possley, Juwan Deering], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=6042 [https://perma.cc/4G93-A5CW] (last updated Jan. 12, 2023) (noting that prosecutors failed to disclose evidence indicating an ongoing relationship between jail informants and the Oakland County Sheriff’s Office and the Oakland County Prosecutor’s Office). ↑
- Possley, Jackie Wilson, supra note 29 ↑
- Id. (“Trutenko also revealed that after he left the state’s attorney’s office in 1991 and went into private practice, he and Coleman became friends. In 1992, Coleman paid for Trutenko to fly to England to be the godfather of Coleman’s daughter who was being christened. Slosar asked, ‘You never thought for one second to tell another person in your office that you became the godfather of (the daughter of) the jailhouse informant, is that correct?’”). Prosecutors’ failure to comply with their disclosure obligations under Brady are, of course, not limited to the jailhouse informants in the cases. They also fail to disclose other types of exculpatory evidence. See, e.g., Possley, Renay Lynch, supra note 61 (noting that prosecutors failed to disclose that fingerprints found at crime scene were tested and excluded both defendants in the case). ↑
- Otterbourg, Robert DuBoise, supra note 139. ↑
- Id. Jailhouse informant Saladino testified that he “had asked for nothing and been offered nothing in exchange for his testimony.” Saladino explained his motivation for testifying was that “he didn’t condone violence against women and because he didn’t want DuBoise to take all the blame for the crime,” which he apparently accomplished by implicating others in addition to DuBoise. Id. ↑
- See Otterbourg, Lamar Johnson, supra note 29. Mock, the jailhouse informant, wrote in the letter that “I am willing to testify as long as I do not have to return to the Department of Corrections once I [testify]. I can’t. I won’t live in protective custody or any institution once I testify.” Id. Mock stated that he didn’t want parole, which he viewed as “too uncertain,” but instead a reduction in his sentence or a pardon from the governor. Id. “I am positive that this can be worked out for the good of all. I will uphold my end of this situation as I am positive you will fulfill your obligations to me.” Id. ↑
- Possley, Walter Ogrod, supra note 57. ↑
- Possley, David Faulkner, supra note 152. See also Ken Otterbourg, Ralph Birch, Nat’l Registry of Exonerations [hereinafter Otterbourg, Ralph Birch], https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5769 [https://perma.cc/DC2L-KHW8] (last updated Apr. 4, 2024) (noting that deals made with testifying witness Yablonski to dismiss pending charges against her prior to her trial testimony were not disclosed); Possley, Raymond Champagne, supra note 84 (noting that prosecutors failed to disclose that a purported eyewitness in the Champagne case was also a suspect in a second case and avoided criminal charges in that case). ↑
- See Covey, Abolishing Jailhouse Snitch Testimony, supra note 135, at 1413. ↑
- Otterbourg, Lamar Johnson, supra note 29. ↑
- Possley, Daniel Carnevale, supra note 84. ↑
- Id. ↑
- See, e.g., Otterbourg, Robert DuBoise, supra note 139 (“During closing arguments, Ober pointed to the strength of the forensic evidence and also vouched for Butler’s credibility, telling jurors that Butler had ‘received nothing’ for his testimony.”); Possley, Walter Ogrod, supra note 57 (“[Prosecutor] Rubino also argued that Wolchansky had nothing to gain for testifying—a claim that was not true based on records later revealed.”). ↑
- In an op‑ed assessing Texas’s legislative reforms, the NY Times concluded that the best solution “would be to bar the use of compensated informants outright.” N.Y. Times Ed. Bd., Editorial, Texas Cracks Down on the Market for Jailhouse Snitches, N.Y. Times (July 15, 2017), https://www.nytimes.com/2017/07/15/opinion/sunday/texas-cracks-down-on-the-market-for-jailhouse-snitches.html [https://perma.cc/LYB6-GHQX]. See also Covey, Abolishing Jailhouse Snitch Testimony, supra note 135. ↑
- For further exploration of the disturbingly common disregard of recantations in the criminal process, see Russell D. Covey, Recantations and the Perjury Sword, 79 Alb. L. Rev. 861 (2015–2016). ↑
- Jailhouse informants recanted in at least 16.6 percent (9 out of 54) of the cases studied here. Other types of witnesses—often alleged eyewitnesses—recanted in an additional 15 percent of the cases (8 out of 54). Several of the cases involved recantations of both types. National Registry of Exonerations, supra note 19. ↑
- The Michael Rhynes case is both a refreshing counterexample and a sobering one. In post‑conviction proceedings, the two jailhouse informants who had testified against Rhyne testified that their trial testimony had been false. The judge presiding at the hearing vacated Rhynes’ conviction, finding the recantations credible and observing that there was no other physical or forensic evidence that connected Rhynes to the crime. Rhynes’ conviction was thus vacated and he was released after spending thirty‑seven years, four months and five days in prison. Possley, Michael Rhynes, supra note 119. ↑
- The most common reason given by informants for coming forward and admitting that their prior testimony was false is their feeling of regret and remorse for falsely implicating an innocent person. For example, the jailhouse informant who testified against Michael Rhynes began trying, according to him, to tell authorities that his testimony was false two years after Rhynes was convicted: “I have no knowledge if Michael Rhynes was involved in any crime or not, but I feel horrible that I have contributed to him being incarcerated because I lied to get a better deal for myself.” Id. See also Possley, Lee Harris, supra note 89. ↑
- See Otterbourg, Ralph Birch, supra note 179. For other cases involving recantations, see Possley, Marvin Cotton, supra note 127 (noting that a jailhouse informant gaves a statement recanting in 2010 and another in 2014); Possley, Walter Ogrod, supra note 57. A habeas petition filed by Ogrod “included a series of affidavits, including one from John Hall, who by then was dead, stating that Wolchansky, who also was dead, had never talked to Ogrod ‘in any detail.’ Hall said that he provided all of the information to Wolchansky. In addition, Hall’s widow gave an affidavit stating that Hall had told her that Ogrod had not confessed in jail.” Id. ↑
- In the infamous Troy Davis case, Davis was tried for the murder of a Savannah police officer. The state presented nine factual witnesses in the trial. Seven of those witnesses, including a jailhouse informant named Kevin McQueen, recanted their testimony. A federal judge, in a hearing convened to evaluate Davis’s long‑ignored innocence claim, the judge concluded that Davis had failed to establish his actual innocence. He dismissed McQueen’s recantation as follows: “[g]iven that Mr. McQueen’s trial testimony was so clearly fabricated, and was actually contrary to the State’s theory of the case, it is unclear why the State persists in trying to support its veracity. Regardless, the recantation is credible, with the exception of the allegation of prosecutorial inducements, but only minimally reduces the State’s showing at trial given the obviously false nature of the trial testimony.” In re Davis, No. CV409‑130, 2010 WL 3385081, at *49 (S.D. Ga. Aug. 24, 2010) (internal citations omitted). Davis was subsequently executed notwithstanding substantial doubt about his guilt. See Patrick S. Metze, Troy Davis, Lawrence Brewer, and Timothy Mcveigh Should Still Be Alive: Certainty, Innocence, and the High Cost of Death and Immorality, 6 Charleston L. Rev. 333 (2012). ↑
- Otterbourg, Ralph Birch, supra note 179. ↑
- Id. ↑
- Id. ↑
- Possley, Lee Harris, supra note 89. ↑
- Id. ↑
- Id. ↑
- Possley, Curtis Flowers, supra note 145. ↑
- Id. Veal, another jailhouse informant, and Harris, a purported eyewitness who was afraid he would be arrested if he didn’t identify Flowers, also recanted. ↑
- Possley, Raymond Champagne, supra note 84. ↑
- Id. See also Possley, Barry Williams, supra note 71 (noting that a jailhouse informant recanted in 2001 but Williams not exonerated until 2021); Otterbourg, Mark Purnell, supra note 120 (noting that Mitchell recanted to a fellow inmate in 2012, and later, formally, in 2017). ↑
- See Otterbourg, Thomas Malik, supra note 30; Possley, Olin Coones, supra note 106; Possley, Ramon Ward, supra note 64; Possley, Walter Ogrod, supra note 57; Possley, Lacino Hamilton, supra note 84; Possley, Bernard Howard, supra note 62. Such snitches played a prominent role in several of the jailhouse informant dataset cases, including those of Thomas Malik, Olin Coones, Ramon Ward, Walter Ogrod, Lacino Hamilton, and Bernard Howard. ↑
- See Possley, Lacino Hamilton, supra note 84; Possley, Bernard Howard, supra note 62. ↑
- See Kansas v. Ventris, 556 U.S. 586 (2009). ↑
- Possley, Olin Coones, supra note 106. ↑
- Otterbourg, Thomas Malik, supra note 30. ↑
- Id. ↑
- Possley, Bernard Howard, supra note 62. ↑
- Possley, Lacino Hamilton, supra note 84. ↑
- Id. ↑
- While an awareness about how snitching works is something that prisoners usually learn while incarcerated, jurors may be entirely unaware of such practices. Some commentators have argued that greater use of “jailhouse informant” experts to provide background information on snitching practices would help jurors to better evaluate jailhouse informant testimony. See, e.g., J. Vincent Aprile II, Trolling for Informants Inside Detention Facilities, Crim. Just., Summer 2020, at 57, 58 (“Due to the probability that jurors will lack both experience and education with regard to prison life and jailhouse informants, these types of cases should be fertile ground for the defense use of a teaching expert to educate the jury about such matters as the culture of confinement, the inmates’ common expectation of the potential benefits of being an informant, and how informants are recruited by the government.”). ↑
- Possley, Bernard Howard, supra note 62. ↑
- Id. ↑
- See Possley, Marvin Cotton, supra note 127; Otterbourg, Clifton Caldwell, supra note 29. ↑
- See, e.g., Possley, Jackie Wilson, supra note 29 (noting a jailhouse informant testified against him at second trial after the initial murder conviction was thrown out due to revelations of use of torture by Chicago Detective Jon Burge); Ken Otterbourg, Albert Debelbot, Nat’l Registry of Exonerations, https://www.law.umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=5965 [https://perma.cc/E82J-TYN3] (last updated May 25, 2021) (jailhouse informant used as a rebuttal witness). ↑
- Possley, Marvin Cotton, supra note 127 (noting that a jailhouse informant denied receiving anything in exchange for his testimony, explaining that he came forward with the alleged confession “[b]ecause . . . one of the [other] guys who was in the bullpen was the wrong guy and no sense [of] him going down for something somebody else did.”). ↑
- Id. In a later statement, the informant, a man named Ellis Frazier, Jr., admitted that he had been promised a reward for testifying. “The detective that did my statement said I would get out earlier than my one‑year sentence after I testified. The prosecutor said it too. And they wouldn’t give me time on my probation [revocation] charge.” According to Frazier, he was released seven months early and his probation revocation was dismissed. Id. ↑
- Possley, Keith Davis Jr., supra note 48. ↑
- Id. ↑
- Possley, Olin Coones, supra note 106. ↑
- Possley, Curtis Flowers, supra note 145. ↑
- Possley, Walter Ogrod, supra note 57. ↑
- Possley, Juwan Deering, supra note 172. ↑
- Id. ↑
- Id. ↑
- Possley, Barry Williams, supra note 71. ↑
- Possley, Marvin Cotton, supra note 127. ↑
- Id. Jailhouse informant Saladino testified that he “had asked for nothing and been offered nothing in exchange for his testimony.” Saladino explained his motivation for testifying was that “he didn’t condone violence against women and because he didn’t want DuBoise to take all the blame for the crime,” which he apparently accomplished by implicating others in addition to DuBoise. Id. ↑
- Critics of the proposal to ban jailhouse informant testimony often invoke the loss of investigative or informational value that they suppose, or contend, such a ban would entail. But this argument is a red herring. Banning jailhouse informant testimony is not the same as banning the use of informants to obtain informational leads, identify suspects, or otherwise assist in an investigation. It simply bars the state from putting the informant on the witness stand to testify. Indeed, a ban might even permit jailhouse informants to testify but simply prohibit the state from providing rewards or incentives to the witness in exchange. That type of ban would be consistent with the normal rules of evidence that make it unethical, and even unlawful, to compensate fact witnesses for their testimony. ↑
- In an amicus brief filed in Kansas v. Ventris, a case concerning the admissibility of testimony by a jailhouse informant obtained in violation of the Sixth Amendment right to counsel, the National Association of Criminal Defense Lawyers argued that “the Court should not allow lower courts to admit unreliable jailhouse snitch testimony for impeachment purposes” because it is a type of evidence that is “especially prone to unreliability.” Brief of the National Association of Criminal Defense Lawyers as Amicus Curiae in Support of Respondent, Kansas v. Ventris, 556 U.S. 586 (2009) (No. 07‑1356), 2008 WL 5409458 (U.S.). The Supreme Court declined to follow that suggestion. Kansas v. Ventris, 556 U.S. 586 (2009) (holding that informant’s statements taken in violation of the Sixth Amendment were admissible for purposes of impeachment). ↑
- Tex. Code Crim. Proc. art. 2A.111(b)(2025). ↑
- See Allen, supra note 6, at 272–73. ↑
- Neb. Rev. Stat. Ann. § 29‑4703 (West 2019). ↑
- See 2019 Conn. Acts 19‑131 (Reg. Sess.). ↑
- Tex. Code Crim. Proc. Ann. art. 39.14(h‑1) (West 2017). ↑
- Id. ↑
- See 725 Ill. Comp. Stat. Ann. 5/115‑21 (West 2019); Conn. Gen. Stat. Ann. § 54‑86o (West 2019). ↑
- See United States v. Bagley, 473 U.S. 667, 682 (1985) (explaining that impeachment evidence is exculpatory evidence for Brady purposes and must be disclosed “if there is a reasonable probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different”). ↑
- Giglio v. United States, 405 U.S. 150 (1972) (finding the prosecutor’s failure to disclose that witness received promise of immunity in exchange for testimony violated due process and the court’s holding in Brady v. Maryland, 373 U.S. 83 (1963)). ↑
- Bagley, 473 U.S. at 676 (citing Giglo, 405 U.S. at 154) (interpreting Brady to require disclosure only of “material exculpatory evidence”). ↑
- See Kyles v. Whitley, 514 U.S. 419, 437 (1995) (explaining that “the rule in Bagley (and hence Brady) requires less of the prosecution than the ABA Standards for Criminal Justice, which call generally for prosecutorial disclosures of any evidence tending to exculpate or mitigate”). ↑
- Tex. Code Crim. Proc. Ann. art. 38.075 (West 2017). ↑
- See Timothy Cole Exoneration Report, supra note 1, at 15 (noting that “[c]urrently, jurors are not required to be informed about the informant’s previous charges that may have been dismissed or modified as part of a plea bargain in unrelated cases”). ↑
- See Cal. Penal Code § 1111.5 (2012) (barring convictions “based on the uncorroborated testimony of an in‑custody informant”). ↑
- See Boaz Sangero & Mordechai Halpert, Why a Conviction Should Not Be Based on a Single Piece of Evidence: A Proposal for Reform, 48 Jurimetrics J. 43, 44–53 (2007) (arguing that possibility of false‑positive often, and perhaps always, renders probative value of single, uncorroborated piece of scientific evidence minimal “in the overwhelming majority of cases”). ↑
- People v. Davis, 159 Cal. Rptr. 3d 154, 157 (Cal. Ct. App. 2013). ↑
- People v. Huggins, 185 Cal. Rptr. 3d 672, 235 (Cal. Ct. App. 2015) (holding that accomplice witness may corroborate jailhouse informant, and vice versa). California only restricts corroboration by another in‑custody informant if the prosecutor is unable to prove that the two informants did not discuss the subject matter of their testimony. See Giglio v. United States, 405 U.S. 150, 150 (1972). ↑
- Conn. Gen. Stat. Ann. § 54‑86p (West 2019). ↑
- Id. ↑
- Id. § 54-86p(b). ↑
- 725 Ill. Comp. Stat. Ann. 5/115‑21(d) (West 2019) (mandating reliability hearings absent defendant’s waiver). ↑
- Numerous scholars have advocated for pretrial reliability hearings. See, e.g., Jessica A. Roth, Informant Witnesses and the Risk of Wrongful Convictions, 53 Am. Crim. L. Rev. 737, 780 (2016); Natapoff, supra note 23. ↑
- See Directive 2020‑11 from Gurbir S. Grewal, N.J. Att’y Gen., to Cnty. Prosecutors (Oct. 9, 2020), https://www.nj.gov/oag/dcj/agguide/directives/ag-Directive-2020-11_Jailhouse-Informants.pdf [https://perma.cc/784V-P4M7]; see also Jennifer Sutterer, Snitches Cause Stitches: The Need for Legislative Reform on Jailhouse Informant Testimony Laws, 48 J. Legis. 337, 352 (2022) (discussing directive). ↑
- See Francy R. Monestime, The (De)value of Unsubstantiated Allegations Against the Police, 92 Fordham L. Rev. 1109, 1153 n.363 (2023) (describing a bill that would create a database that would, inter alia, catalog and make available “records of confirmed misconduct infractions committed by officers”). ↑
- The database, which was published in the summer of 2023, “includes 3,400 disciplinary records involving 2,100 officers accused of misconduct, ranging from not properly responding to an incident to being convicted of a crime,” according state media reports. Deborah Becker & Ally Jarmanning, Mass. Police Misconduct Database Is Released, but Excludes Many Complaints, WBUR (Aug. 22, 2023), https://www.wbur.org/news/2023/08/22/mass-police-misconduct-database-post-complaints [https://perma.cc/M3C9-FATS]. ↑
- See Antonio M. Romanucci, Three Pillars for Police Reform, Trial, Sept. 2023, at 18, 21 (describing efforts in Chicago and D.C. and calling for national deployment of such databases). ↑
- Id. at 21. ↑
- Texas’s corroboration requirement, for example, provides that “[a] defendant may not be convicted of an offense on the testimony of a person to whom the defendant made a statement against the defendant’s interest during a time when the person was imprisoned or confined in the same correctional facility as the defendant unless the testimony is corroborated by other evidence tending to connect the defendant with the offense committed.” Tex. Code Crim. Proc. Ann. art. § 38.075(a) (West 2017). ↑
- See Elizabeth Webster, Postconviction Innocence Review in the Age of Progressive Prosecution, 83 Albany L. Rev. 989, 1000 (2020) (citing Barry Scheck, Conviction Integrity Unit Best Practices, Innocence Project (Oct. 15, 2015), https://www.innocenceproject.org/wp-content/uploads/2016/09/Conviction-Integrity-Unit.pdf [https://perma.cc/KXU9-SQP3]). ↑
- See Pennsylvania v. Finley, 481 U.S. 551, 555 (1987). ↑
- See Garza v. Idaho, 586 U.S. 232, 246 (2019) (noting that “most postconviction petitioners” proceed “pro se”). ↑
- See, e.g., Stephanie Roberts Hartung, Habeas Corpus for the Innocent, 19 U. Pa. J.L. & Soc. Change 1, 3 (2016) (observing that fundamental “purpose of federal habeas corpus review” is “to remedy wrongful convictions of the innocent”). ↑
- See Cal. Penal Code §1473 (2024). ↑
- Edward J. Imwinkelried, Debunked, Discredited, but Still Defended: Revising State Post‑Conviction Relief Statutes to Cover Convictions Resting on Subsequently Invalidated Expert Testimony, 48 Seton Hall L. Rev. 1095, 1101 (2018) (discussing amendments to statutes authorizing grants of new trials in cases in which new scientific evidence contradicts the opinions provided by experts at trial). ↑
- † False Confession ↑
- ‡ Official Misconduct ↑
- * Mendoza’s case is not identified in the NRE as involving a false confession. However, the detectives in the case did induce a false confession from her codefendant, Marilyn Mulero, which also implicated Mendoza. Mulero’s confession was subsequently determined to be false and she was exonerated in 2022. See Possley, Madeline Mendoza, supra note 29. ↑