On September 30, 2015, Richard Wayne Glossip sat in a holding cell at the Oklahoma State Penitentiary in McAlester, dressed in fresh clothes, having eaten what prison officials told him would be his last meal. The IV lines had not yet been inserted. The witnesses had not yet been seated in the observation room. But Glossip had been through this before — twice before, in fact, in 2009 and 2011 — and he had no particular reason to believe that this time would be any different from the others. He was going to die. The State of Oklahoma had decided that he would die for ordering the murder of Barry Van Treese, the owner of an Oklahoma City motel, in January 1997. It had decided this twice over — once in 1998, and again in 2004 after his first conviction was vacated and a second jury was impaneled to reach the same verdict. And now, on the last day of September 2015, Oklahoma was prepared to execute him by lethal injection.

What stopped it was not evidence of innocence. It was not a court order prompted by newly discovered proof that Glossip had not committed the crime. It was a pharmacist’s error: the Oklahoma Department of Corrections had obtained the wrong drug — potassium acetate instead of potassium chloride — for use in its three-drug execution protocol. When prison officials noticed the discrepancy at the last minute, they halted the execution. Oklahoma Governor Mary Fallin granted Glossip a two-week stay while the Department of Corrections sorted out its drug supply problem. Two weeks later, the execution was rescheduled.

It never happened. A different set of circumstances — legal challenges, growing public attention to the case, and the extraordinary intervention of Oklahoma’s own law enforcement apparatus — combined to keep Glossip alive long enough for what should have been found decades earlier to come to light. And what came to light, in 2023, when Oklahoma Attorney General Gentner Drummond concluded a comprehensive reinvestigation of the case, was this: the prosecutors who had twice secured Glossip’s conviction had suppressed evidence that was directly favorable to the defense. The lead witness against Glossip — the man who had actually committed the murder, who had received a life sentence in exchange for his testimony, and who was the sole direct evidence connecting Glossip to the crime — had told police a story that contradicted the account he gave at trial. That contradiction had been documented in law enforcement records. Those records had not been turned over to the defense. They had been, in the technical language of constitutional criminal procedure, Brady material — material that the Constitution required the prosecution to disclose — and they had not been disclosed.

Richard Glossip spent more than twenty-five years on death row, survived nine scheduled execution dates, and was twice within hours of a lethal injection needle being inserted into his arm, because Oklahoma prosecutors had withheld evidence that their own law enforcement colleagues had documented, and that a properly informed defense could have used to challenge the only testimony that connected him to a murder he says he did not commit. He is, as of this writing, still on death row. No prosecutor has been disciplined. No criminal charges have been filed. The Oklahoma Court of Criminal Appeals — the state’s highest criminal court — initially rejected the Attorney General’s own recommendation that Glossip’s conviction be vacated. The spectacle of an attorney general fighting against his state’s own courts to prevent an execution he believes would be wrong is, in the annals of American criminal justice, virtually unprecedented. It is also, in the history of prosecutorial accountability, almost certainly going to produce nothing: no one who put Richard Glossip on death row will lose their license, their pension, or their freedom.

January 7, 1997: The Crime That Built a Conviction on a Single Witness

Barry Van Treese owned and managed the Best Budget Inn, a low-rent motel on the outskirts of Oklahoma City. It was the kind of establishment that existed at the margins of the formal economy — cheap rooms, transient guests, the kind of place where people stayed when they had nowhere else to go. Richard Glossip managed the property for Van Treese, handling day-to-day operations, collecting rent, and keeping the peace among a population of guests that included recovering addicts, parolees, and people in acute poverty. Justin Sneed was a 19-year-old handyman who lived at the motel, doing maintenance work in exchange for a room.

On the morning of January 7, 1997, Van Treese was found beaten to death in Room 102 of the Best Budget Inn. He had been bludgeoned with a baseball bat. The room showed signs of a violent struggle. The murder weapon, later identified as a bat, was discovered on the property. Justin Sneed was quickly identified as a suspect — he had access to the room, had been seen near it, and could not account for his whereabouts during the relevant period. When police questioned him, Sneed initially denied any involvement. Then, under sustained interrogation, he changed his story.

Here is what Sneed eventually told police, in the account that formed the basis of Richard Glossip’s prosecution: Glossip had approached him days before the murder and proposed that Sneed kill Van Treese. The motive, according to Sneed, was that Glossip had been stealing money from the motel — money Van Treese was coming to collect — and needed Van Treese eliminated to avoid being caught. In exchange for committing the murder, Sneed said Glossip had promised him money and continued access to his motel room. Sneed admitted to the killing. He said he had done it at Glossip’s direction and for Glossip’s benefit.

On the basis of this testimony — and essentially no other direct evidence — Richard Glossip was charged with first-degree murder in the form of “murder for hire.” He was not charged as the actual killer; the evidence was clear that Sneed had wielded the bat. He was charged as the person who had hired Sneed to commit the murder. In Oklahoma, a person who procures a murder is as guilty as the person who commits it, and subject to the same range of punishments, including death. The prosecution’s entire theory of Glossip’s guilt depended on the jury believing Justin Sneed — the confessed killer, the man who had physically beaten Barry Van Treese to death — when Sneed said that Glossip had put him up to it.

There was no physical evidence linking Glossip to the murder. There was no documentary evidence — no payment, no record, no corroborating witness — of any arrangement between Glossip and Sneed. The prosecution had no murder weapon with Glossip’s fingerprints, no witness who had heard Glossip discuss killing Van Treese. There was Justin Sneed’s testimony. There was Sneed’s cooperation agreement, which gave him a life sentence — rather than death — in exchange for his willingness to testify. And there was the prosecution’s skill in presenting that arrangement to the jury as reliable rather than profoundly compromised.

The First Trial, the Vacated Conviction, and the Second Death Sentence

Glossip’s first trial in 1998 resulted in a conviction and a death sentence. His defense at trial — that Sneed was lying, that Sneed had committed the murder on his own initiative, and that Glossip had no involvement in planning or ordering it — did not persuade the jury. Sneed’s testimony, delivered with apparent conviction by a man who had everything to gain from being believed and who had already admitted to the killing himself, was sufficient to tip the scales.

The first conviction did not survive appellate review. In 2001, the Oklahoma Court of Criminal Appeals vacated the conviction on grounds of ineffective assistance of counsel — finding that Glossip’s trial attorney had failed to adequately investigate and present a defense, and that this failure had prejudiced the outcome. The court ordered a new trial. This reversal was legally significant, but limited: it acknowledged a constitutional deficiency in the process, not a deficiency in the evidence. The evidence that had convicted Glossip the first time — principally Sneed’s testimony — remained available for a second prosecution.

The second trial, in 2004, produced the same outcome: conviction, death sentence. A new jury, hearing essentially the same case built around the same primary witness, reached the same conclusion. Oklahoma’s courts affirmed the conviction through the state appellate process. Federal habeas corpus proceedings began — the long, slow, procedurally constrained process by which federal courts review constitutional claims in state criminal cases. Through those proceedings, the execution dates were repeatedly set and repeatedly stayed as courts worked through the constitutional questions Glossip’s attorneys raised.

But none of the legal proceedings produced what would only become visible years later: the documented evidence that prosecutors had possessed, and had not disclosed, that directly undermined the reliability of the testimony on which both convictions rested.

What Was Hidden: The Brady Material Oklahoma Suppressed

The constitutional rule established in Brady v. Maryland, 373 U.S. 83 (1963), requires that prosecutors disclose to the defense any material evidence that is favorable to the defendant — evidence that could be used to challenge the prosecution’s case or to impeach its witnesses. The rule is one of the foundational guarantees of the American criminal justice system. Its violation is not a minor procedural irregularity. It is a constitutional violation that, when it affects the outcome of a trial, requires reversal of the conviction and a new proceeding in which the withheld evidence can be properly presented.

What Oklahoma prosecutors withheld in the Glossip case falls squarely within the core of what Brady protects.

The most significant suppressed evidence concerned Justin Sneed’s initial account to police — the account he gave before he settled on the story that Glossip had hired him to commit the murder. According to the findings of the 2023 reinvestigation, Sneed initially told investigators that he had committed the murder alone, without any direction from Glossip and without any arrangement for payment. This initial account — which directly contradicted the testimony Sneed would later give at trial — was documented in police records. Those records were never turned over to the defense.

The significance of this suppression cannot be overstated. The entire prosecution of Richard Glossip rested on the proposition that Justin Sneed, having admitted to the killing, was telling the truth when he said that Glossip had hired him to do it. Sneed’s credibility was the trial. If the jury had known that Sneed had initially told police the exact opposite story — that he had acted alone, that there was no arrangement with Glossip — the jury’s assessment of Sneed’s reliability would have been fundamentally different. A witness who tells one story to police and a different story at trial is not a reliable witness. A witness who tells one story when he has nothing to offer and a different story after he has negotiated a cooperation agreement that saves his life is a witness with every incentive in the world to say whatever the prosecution needs him to say.

The withheld material also reportedly included records related to Sneed’s mental health treatment — records that could have been used to challenge Sneed’s competency, his susceptibility to suggestion during interrogation, or the conditions under which his cooperation agreement was negotiated. And it included notes and documentation from interrogation sessions that the defense never saw — sessions in which the evolution of Sneed’s account, from “I did it alone” to “Glossip hired me,” was captured in real time.

These documents existed. They were in prosecutors’ files. The defense never received them. And Richard Glossip, on the basis of a case that these documents could have significantly undermined, spent twenty-five years on death row and survived nine scheduled executions.

“I consider trial by jury as the only anchor ever yet imagined by man, by which a government can be held to the principles of its constitution.”

The 2015 Near-Execution and the Drug That Was Wrong

The September 30, 2015 near-execution of Richard Glossip has become one of the most widely reported episodes in recent American death penalty history — but for the wrong reasons. Most coverage focused on the bureaucratic farce of the wrong drug procurement: the Oklahoma Department of Corrections had obtained potassium acetate when its protocol called for potassium chloride. Prison officials noticed the error, the execution was halted, and Glossip was returned to his cell with the knowledge that his life had been extended not by any court’s recognition of his potential innocence but by a supply chain failure.

The wrong drug story obscured a more important development that was simultaneously unfolding. The 2015 near-execution generated significant national attention to Glossip’s case — attention that brought new scrutiny to the evidence underlying his conviction. Conservative legal commentators, including figures not normally associated with death penalty opposition, began examining the case and expressing discomfort with executing a man who might have been convicted on the basis of a single witness’s compromised testimony. Sister Helen Prejean, the anti-death-penalty activist whose work inspired the film Dead Man Walking, became involved in advocating for Glossip, lending her public credibility to the question of his guilt.

The 2015 episode planted the seed. Glossip’s case had attracted enough national attention that when new evidence began to emerge about prosecutorial misconduct in the original investigation and prosecution, there was an audience ready to understand its significance.

The Attorney General Turns: Drummond’s Extraordinary 2023 Finding

Oklahoma Attorney General Gentner Drummond is, by any reasonable characterization, a law-and-order Republican — a prosecutor’s prosecutor, elected in a state with one of the highest incarceration rates in the nation and a long history of aggressive criminal justice administration. He is not a death penalty opponent. He is not an advocate for criminal defendants. He is, in fact, the kind of officeholder who might be expected to defend Oklahoma’s conviction of Richard Glossip as a matter of prosecutorial solidarity and political prudence.

In 2023, having directed a comprehensive reinvestigation of the Glossip case by his office, Drummond reached a conclusion that was, by any measure, extraordinary: the conviction should be vacated. The prosecution had suppressed material evidence. The star witness had changed his story after negotiating a cooperation deal that saved his life. The reliability of the conviction could not withstand scrutiny. Drummond filed a brief with the Oklahoma Court of Criminal Appeals joining Glossip’s call for a new trial.

The spectacle of a state attorney general arguing against his own state’s conviction in a capital case — fighting alongside the defense rather than alongside the previous prosecution — is essentially without precedent in the modern history of American criminal justice. It reflected Drummond’s conclusion that the evidence of prosecutorial misconduct was so clear, and the risk of executing an innocent man so real, that professional loyalty to past prosecutorial decisions could not justify maintaining a conviction that the law and the evidence could not support.

The Oklahoma Court of Criminal Appeals did not immediately agree. In a decision that drew sharp criticism from legal observers, the court declined to vacate Glossip’s conviction, finding — in the face of the attorney general’s own arguments — that the Brady violations had not been established with sufficient clarity to warrant relief. The court’s reasoning was widely characterized as motivated reasoning: an institution protecting the integrity of its own prior decisions against evidence that those decisions had been built on a constitutional violation.

Drummond appealed to the United States Supreme Court.

The Supreme Court Speaks: Glossip v. Oklahoma (2024)

In 2024, the United States Supreme Court decided Glossip v. Oklahoma — a decision that, in the annals of capital punishment litigation, stands as one of the most remarkable in decades. The Court, in an opinion that cut across the conventional liberal-conservative fault lines of capital cases, reversed the Oklahoma Court of Criminal Appeals and remanded the case for new proceedings consistent with the finding that Brady violations had occurred.

The Court’s decision was, in its most fundamental dimension, a vindication of the constitutional rule that the government cannot hide evidence that might save a defendant’s life. The majority found that the suppression of Sneed’s initial statement to police, the records of his mental health treatment, and the documentation of the evolution of his cooperation with prosecutors constituted exactly the kind of material favorable evidence that Brady requires to be disclosed. The Court rejected the Oklahoma Court of Criminal Appeals’ narrow reading of the Brady obligation and found that the withheld evidence was material — that there was a reasonable probability that, had it been disclosed, the outcome of the trial would have been different.

The decision did not acquit Richard Glossip. It did not declare him innocent. It sent his case back to Oklahoma for proceedings in which the withheld evidence must be properly considered. But it established, at the highest judicial level in the American system, that Oklahoma had convicted Richard Glossip on the basis of a constitutionally deficient proceeding — that the prosecution had hidden evidence that the Constitution required it to disclose — and that this violation required correction.

As of this writing, those proceedings continue. Glossip remains on death row. Whether Oklahoma will retry him, whether the evidence available without Sneed’s compromised testimony is sufficient to sustain any prosecution at all, whether a man who has spent more than a quarter century on death row for a crime he may not have committed will ever be free — remains unresolved. And the prosecutors who built their careers on his conviction face no professional consequences whatsoever for the constitutional violations that the Supreme Court of the United States has now confirmed they committed.

“He has erected a multitude of New Offices, and sent hither swarms of Officers to harass our people, and eat out their substance… depriving us in many cases, of the benefits of Trial by Jury.”

Justin Sneed: The Architecture of a Cooperator’s Testimony

The Glossip case is, among other things, a case study in how American prosecutors construct convictions using cooperating witnesses — and in how the structural incentives that define cooperation agreements systematically corrupt the testimony they produce.

Justin Sneed was nineteen years old when he killed Barry Van Treese with a baseball bat in Room 102 of the Best Budget Inn. He confessed to the killing. In Oklahoma in 1997, a person who confessed to first-degree murder faced death. Sneed faced death. And then prosecutors offered him a deal: testify against Glossip, provide the testimony that would make a murder-for-hire case against the motel manager, and you will receive life in prison instead of the needle. Sneed took the deal. Sneed testified. Glossip received two death sentences. Sneed received life.

The cooperation agreement did not just give Sneed an incentive to testify against Glossip. It gave Sneed an incentive to testify in a specific way — in a way that supported the murder-for-hire theory, that placed Glossip at the center of the crime, and that justified the prosecution’s charging decision. A cooperating witness who tells a story that the prosecution finds unhelpful does not keep his deal. The cooperation agreement is, in practice, a performance contract: say what we need you to say, say it convincingly, and you live. Say something different, and the deal unravels.

This is why Sneed’s initial statement to police — the statement in which he said he had acted alone — was so damaging to the prosecution’s case. If the jury had known that Sneed’s initial story had not included Glossip at all; that Glossip had only become part of Sneed’s account after the cooperation deal was on the table; that the version of events in which Glossip hired Sneed to commit the murder was a version that emerged during the negotiation of a deal that saved Sneed’s life — the jury’s assessment of Sneed as a witness would have been dramatically different.

The suppression of that initial statement was not an accident. It was not an oversight. A law enforcement file documenting a suspect’s initial account of a crime — an account that contradicts the testimony the suspect later provides after entering a cooperation agreement — is not the kind of document that gets accidentally misplaced. It is the kind of document that, when found in a prosecutor’s file rather than in the defense disclosure package, indicates a deliberate decision not to share it. Someone in the Oklahoma County District Attorney’s office knew that Sneed had initially said something different. Someone decided not to tell the defense. That decision put Richard Glossip on death row for twenty-five years.

The Accountability Vacuum: What Happens to Prosecutors Who Break the Law

In the quarter century since Richard Glossip was first convicted, the prosecutors who built their case on withheld evidence have faced no professional consequences. No disciplinary proceedings have been initiated against them by the Oklahoma Bar Association. No criminal charges have been filed under the federal statute — 18 U.S.C. § 242 — that criminalizes the deprivation of constitutional rights under color of law. No civil judgments have been entered against them in their personal capacities. The Supreme Court of the United States has confirmed that a constitutional violation occurred in the case they prosecuted. The attorneys who committed that violation are, as far as the accountability systems of American law are concerned, in good standing.

This outcome is not unusual. It is, in the statistics of prosecutorial accountability, essentially the norm.

Prosecutorial immunity is the doctrine that makes accountability for conduct like that demonstrated in the Glossip case effectively impossible through civil litigation. Under Imbler v. Pachtman, 424 U.S. 409 (1976), a prosecutor acting within the scope of the prosecutorial function — including the decision about what evidence to disclose to the defense — enjoys absolute immunity from civil liability under 42 U.S.C. § 1983. This immunity is absolute in the sense that it applies even when the prosecutorial conduct was deliberately unconstitutional. A prosecutor who intentionally withholds Brady material, who knowingly presents false testimony, who hides evidence that could save an innocent person’s life, cannot be sued for money damages under federal civil rights law. The doctrine is one of the most sweeping grants of official immunity in American law — far broader than the qualified immunity that protects police officers, which at least requires that the violated right be clearly established.

The state bar disciplinary systems that theoretically police attorney conduct — including prosecutorial misconduct — have an equally dismal record. Studies examining bar discipline of prosecutors who have committed Brady violations have found that the rate of disciplinary action is vanishingly small: somewhere between one and two percent of cases in which courts find Brady violations result in any bar discipline against the responsible prosecutor. The reasons are structural: bar disciplinary boards are run by lawyers who practice before the prosecutors they are being asked to discipline; the standard of proof required for discipline is high; the culture of the legal profession defaults toward protecting practitioners rather than policing them; and the prosecutors who commit Brady violations often do so in ways that are characterized, generously, as negligence rather than intentional misconduct — a characterization that reduces the likelihood of discipline even when the violation is documented.

The criminal prosecution of prosecutors for Brady violations is, in the practical experience of American criminal justice, essentially nonexistent. The federal statute at 18 U.S.C. § 242 theoretically applies: it criminalizes the willful deprivation of constitutional rights under color of law, and a willful Brady violation — the deliberate decision not to disclose evidence the Constitution requires be disclosed — fits the statutory description. But the Department of Justice has almost never used § 242 against prosecutors. State prosecutors cannot typically be prosecuted by other state prosecutors within the same system, creating an obvious conflict. And the political costs of prosecuting a sitting or former district attorney for conduct committed in the prosecution of a criminal case are sufficiently high that no ambitious state attorney general has found it worth the effort.

The Death Penalty’s Unique Accountability Crisis

The Glossip case has a dimension that makes it categorically different from most prosecutorial misconduct cases: if the constitutional violations had not been discovered, or had been discovered too late, Richard Glossip would be dead. The state of Oklahoma would have executed him. And the evidence of prosecutorial misconduct would likely have remained buried in files that no one would have thought to reopen because the person most interested in reopening them would have been gone.

This is the specific horror of prosecutorial misconduct in capital cases. Wrongful convictions that result in incarceration are devastating — they rob years and decades from the lives of innocent people, they destroy families, they inflict harms that no compensation can fully address. But they are, at least theoretically, remediable. The person can be released. The conviction can be vacated. Compensation can be paid. In capital cases, there is a remedial horizon after which the harm becomes absolute and irreversible: the moment of execution. After that moment, the discovery of prosecutorial misconduct produces nothing except an apology — if that — and the grim accounting of a life taken by a justice system that was supposed to protect it.

Cameron Todd Willingham was executed by the State of Texas in 2004 for the arson-murder of his three daughters. The fire investigation evidence that had convicted him was later found, by independent forensic experts commissioned by the Texas Forensic Science Commission, to be scientifically invalid — built on folklore and discredited burn pattern analysis rather than validated forensic methodology. Willingham was executed anyway. The governor at the time, Rick Perry, replaced members of the Forensic Science Commission before they could complete their review of the Willingham case and issue a formal finding. No one was accountable. The accountability would have come after Willingham was already dead, which meant it would not come at all.

Richard Glossip survived nine execution dates. Cameron Todd Willingham survived zero. The difference between them, in terms of the justice system’s interest in their actual guilt or innocence, was not meaningful — it was procedural, circumstantial, and in the case of the 2015 near-execution, pharmaceutical. A state that is determined to execute a person it has convicted will find a way to do so unless something extraordinary intervenes. In Glossip’s case, something extraordinary intervened — a pharmacist’s error, a national media moment, a Republican attorney general with an unexpected commitment to constitutional integrity. Most people in Glossip’s position do not get those interventions. They get the needle.

The Broader Oklahoma Pattern: A State That Executes at Scale

Oklahoma is not simply a state that has one problematic capital case. It is a state with one of the most aggressive execution records in American history — and a corresponding record of capital convictions that have not always withstood subsequent scrutiny.

Oklahoma leads all other states in executions per capita over the modern era of the death penalty — a fact that reflects both the state’s crime rate and its political culture, in which support for capital punishment has historically been near-universal among elected officials. The state has executed more than 120 people since the Supreme Court reinstated the death penalty in 1976. It has done so using a lethal injection protocol that has been the subject of multiple Eighth Amendment challenges, culminating in the Supreme Court’s decision in Glossip v. Gross, 576 U.S. 863 (2015) — a case about Oklahoma’s use of midazolam in its execution protocol that was, by an irony that should not be lost, brought by Richard Glossip’s co-plaintiffs while Glossip himself was simultaneously fighting his own conviction on grounds of innocence.

The state’s aggressive execution schedule has produced, alongside its large total of executions, a number of cases in which serious questions about the reliability of capital convictions have emerged. Oklahoma has exonerated several death row inmates over the years. It has also had cases — including the Glossip case — in which questions about prosecutorial conduct, witness reliability, and evidence integrity have been raised and have taken years to resolve through the appellate process, while the condemned person waits on death row for the resolution to come before the execution date does.

The structural pressure that an aggressive capital punishment system places on prosecutors is itself a contributor to the kinds of violations that the Glossip case illustrates. Prosecutors in capital cases face enormous political and professional pressure to secure convictions. The cooperating witness — the person who knows the most because they participated in the crime — is often the only available path to conviction. The Brady obligations that require disclosure of evidence that undermines that cooperating witness’s testimony stand directly in the path of securing the conviction that the political and prosecutorial culture demands. In that environment, the temptation to suppress inconvenient police records is not just comprehensible — it is, in the actuarial experience of American capital justice, a significant contributor to the wrongful conviction rate.

What Justice Looks Like When It Finally Comes — and What It Does Not

The Supreme Court’s decision in Glossip v. Oklahoma is the most significant thing that has happened for Richard Glossip’s legal case in twenty-five years. It confirms, at the highest institutional level, that the prosecution violated the Constitution in securing his convictions. It requires Oklahoma to conduct new proceedings that give proper weight to the suppressed evidence. It is, in legal terms, a significant victory.

But consider what it actually delivers to Richard Glossip: not freedom, not exoneration, not the return of twenty-five years on death row, and not accountability for the people who put him there. It delivers a new legal proceeding — one in which he will have to relitigate, in his late fifties, a crime from 1997, with witnesses who are decades older, memories that have faded, and the institutional disadvantages of any person who has spent a quarter century in maximum security incarceration. It delivers the possibility of a just outcome. It does not deliver justice.

And the prosecutors who committed the Brady violations that the Supreme Court has now formally identified? They remain lawyers in good standing. They retain their bar licenses. They retain whatever professional positions, reputations, and pensions they have accumulated in the years since the Glossip prosecution. The constitutional violation they committed has been identified, litigated to the Supreme Court, and resolved in a published opinion that will be cited in law school classrooms for decades. Their professional standing, in the accountability architecture of American law, is unaffected.

This is the system’s failure mode, rendered in its starkest possible form. A man nearly died — nine times — because prosecutors hid evidence. The system eventually identified this. It took twenty-five years. The remedy is a new proceeding. The accountability for the underlying misconduct is zero.

Reform Blueprint: Seven Changes That Would Prevent the Next Richard Glossip

The Glossip case is not merely a story about one wrong conviction. It is a story about the structural conditions that make such convictions possible and that allow them to persist for decades while the condemned person lives under the shadow of execution. Addressing those structural conditions requires specific reforms:

1. Abolish absolute prosecutorial immunity for Brady violations and mandate personal civil liability. The Supreme Court’s doctrine of absolute prosecutorial immunity, established in Imbler v. Pachtman, must be revisited. Prosecutors who deliberately withhold material evidence in violation of Brady v. Maryland should not be shielded from personal civil liability for the constitutional violations they commit. Congress has the authority to abrogate immunity under 42 U.S.C. § 1983 through legislation, and should exercise that authority to create a cause of action against prosecutors who commit willful Brady violations. The prospect of personal civil liability is one of the few deterrents that would meaningfully change prosecutorial behavior in cases where institutional culture incentivizes nondisclosure.
2. Create an open-file discovery obligation in all criminal cases, codified by federal statute for cases with federal constitutional dimensions. The Brady doctrine, as currently implemented, requires prosecutors to disclose evidence they have identified as favorable to the defense — a standard that depends on prosecutorial good faith. An open-file discovery rule — one that requires prosecutors to provide the defense with access to the complete law enforcement file, not just the subset the prosecution has selected for disclosure — eliminates the good faith dependency. Several states, including North Carolina, have enacted open-file discovery statutes. Federal legislation should establish this standard for all felony prosecutions.
3. Require independent review of all capital convictions when Brady material is identified. When a court finds that a Brady violation occurred in a capital case — at any stage of the proceedings — an independent review panel, convened outside the normal prosecutorial chain of command, should be required to examine the entire record of the original investigation and prosecution for additional undisclosed material. The Glossip case suggests that Brady violations rarely occur in isolation: a prosecution culture that withholds one piece of evidence is likely to have withheld others. A comprehensive independent review, triggered by any confirmed Brady finding in a capital case, would provide a systemic check that the normal appellate process does not offer.
4. Enact mandatory bar discipline for confirmed Brady violations, with presumptive suspension. State bar associations should adopt rules that make suspension a presumptive consequence of a confirmed Brady violation — a judicial finding, not merely an allegation — and that create a mandatory reporting requirement when courts find that prosecutorial misconduct affected the outcome of a proceeding. The current culture, in which Brady violations almost never result in bar discipline, reflects the bar’s structural unwillingness to police its own members in the prosecutorial context. A mandatory discipline rule, with clear presumptive consequences, would change the institutional calculus that currently makes nondisclosure a low-risk strategy.
5. Establish a federal Conviction Integrity Unit within the Department of Justice with authority to investigate Brady violations in state capital cases. The federal interest in the integrity of capital convictions is direct and substantial: the federal government conducts capital prosecutions, and federal courts review state capital cases through habeas corpus. DOJ should have a standing unit — adequately resourced and institutionally independent from the prosecutorial culture of the criminal division — tasked with reviewing capital cases in which Brady violations have been alleged or confirmed, and with referring cases of apparent willful prosecutorial misconduct for federal criminal investigation under 18 U.S.C. § 242.
6. Reform the cooperating witness system to require independent judicial validation of cooperation agreements before testimony can be used in capital cases. The structural problem at the heart of the Glossip case — a cooperating witness who changes his story after entering a deal that saves his life — reflects endemic problems with the use of cooperating testimony in high-stakes prosecutions. In capital cases, cooperation agreements should require judicial approval and should include an independent judicial officer’s review of any prior inconsistent statements the cooperating witness has made. The deal cannot be finalized, and the testimony cannot be presented, until a judge has reviewed the complete record of the cooperating witness’s prior statements and determined that disclosure obligations have been fully met.
7. Enact a federal moratorium on executions in any state where a capital conviction has been found, by any federal court, to have involved a Brady violation, until all pending capital cases in that state have been independently reviewed for compliance. A Brady violation in a capital case is not an isolated event — it is a symptom of a prosecutorial culture. A state whose prosecutors have been found to have suppressed evidence in one capital case is a state in which similar conduct may have occurred in other capital cases. A federal moratorium on executions, triggered by any confirmed Brady violation in a capital case and lasting until an independent review of all pending capital cases has been completed, would create a structural incentive for states to take Brady compliance seriously and would prevent the irreversible harm of executing someone whose conviction may have been similarly compromised.

The Man in the Cell

Richard Wayne Glossip has been on death row in Oklahoma since he was thirty-five years old. He is now in his early sixties. He has spent the better part of his adult life in a maximum security cell, preparing — and then not preparing, and then preparing again — for a death that Oklahoma has repeatedly scheduled and repeatedly been unable to carry out. He has been within hours of execution. He has eaten last meals. He has said goodbye. And he has watched, from his cell, as the evidence that his conviction was built on a constitutional violation slowly accumulated to the point where the Attorney General of the state trying to execute him concluded that it was wrong to do so.

What a human being carries through twenty-five years of that experience is not something the legal record captures. The briefs and opinions and habeas petitions and execution protocols document the legal history of his case. They do not document what it costs a person to live inside it — to wake up every morning in the knowledge that the state has decided you should die for something you say you did not do, that the system designed to discover the truth has instead built a wall of procedural obstacles between you and any court that might finally see the whole picture.

Richard Glossip may still be executed. Oklahoma may retry him and obtain a third conviction. Or the evidence available without Justin Sneed’s compromised testimony may prove insufficient to sustain any prosecution, and he may finally be free. The constitutional question has been answered by the Supreme Court: his convictions rested on a Brady violation. The human question — whether a man who maintained his innocence through twenty-five years of death row, nine scheduled executions, and one wrong-drug reprieve was telling the truth all along — is still, in a functional sense, open.

What is not open is the question of whether the system worked. It did not. It took twenty-five years, a pharmacist’s error, a Republican attorney general’s crisis of conscience, and a Supreme Court decision to produce the legal recognition that the constitutional violation had occurred. The people who committed that violation are untouched by it. And the man who bore the consequences of it is still in his cell.

That is the state of American prosecutorial accountability in 2026. That is what the Take America Back project exists to document, to challenge, and to change. The Founders built a system premised on the idea that the government would be held to its own law — that the rules constraining the exercise of state power over individual lives would be enforced, and that those who violated them would be answerable. In the case of Richard Glossip and the prosecutors of Oklahoma County, that premise has not been honored. It is past time to honor it.