In Glossip v. Gross, decided June 29, 2015, the Supreme Court upheld Oklahoma’s use of the sedative midazolam as the first drug in its lethal injection protocol and set a demanding standard for inmates who want to challenge how a state plans to execute them. By a 5–4 vote, with Justice Samuel Alito writing for the majority, the Court held that a prisoner bringing an Eighth Amendment method-of-execution claim must prove two things: that the method creates a substantial risk of severe pain, and that a known, available alternative exists that would significantly reduce that risk. Fail either prong and the claim fails.
How the Case Reached the Court
The case grew out of Oklahoma’s April 29, 2014 execution of Clayton Lockett. Lockett received midazolam at 6:23 p.m. and was declared unconscious ten minutes later. About three minutes after that, witnesses reported that he began nodding, mumbling, and writhing on the gurney. He died at 7:06 p.m. of a heart attack, roughly 40 minutes after the drugs were first administered. A corrections department spokesman said a vein had apparently collapsed and the drugs were not entering Lockett’s system as intended.
Oklahoma revised its protocol after the botched execution but kept midazolam as the first drug. Richard Glossip and several other death row inmates sued corrections officials under 42 U.S.C. § 1983, arguing that the revised protocol still risked the kind of prolonged suffering witnesses had described. The district court sided with the state, the Tenth Circuit affirmed, and the Supreme Court took the case.
The Two-Part Test the Court Adopted
The majority tightened the framework the Court had introduced in Baze v. Rees (2008) into a clear two-part requirement. To win an Eighth Amendment challenge to an execution method, an inmate must establish both elements.
First, the method must present a substantial risk of severe pain. The risk has to be “sure or very likely to cause serious illness and needless suffering,” not a speculative or remote possibility of discomfort.1Justia. Glossip v. Gross
Second, the inmate must identify a specific alternative that is “feasible, readily implemented,” and would “significantly reduce” that risk.1Justia. Glossip v. Gross The state does not have to prove its own protocol is safe or research alternatives on its own. The inmate carries the entire burden: locate a better method, show it is available, and show the state could actually carry it out.
The Court reasoned that if no method could satisfy the Eighth Amendment, capital punishment itself would be effectively banned, which the majority said the Constitution does not permit given that the Fifth Amendment references the death penalty.2Oyez. Glossip v. Gross
The Glossip inmates failed both prongs. They did not persuade the Court that midazolam created a substantial risk of severe pain, and they did not propose a specific alternative drug the state could readily use.
One Botched Execution Is Not Enough
The majority also made clear that a single mishap like the Lockett execution does not by itself prove a protocol is unconstitutional. The question is whether the method, as designed and typically carried out, consistently creates a substantial risk of severe pain. A vein collapsing in one execution is a complication, not proof that the drug protocol is inherently cruel.
The Fight Over Midazolam
Midazolam became central because the drugs states had traditionally used were disappearing. Sodium thiopental, the anesthetic that anchored older three-drug protocols, went out of U.S. production in 2010, and European export controls and manufacturer restrictions cut off access to pentobarbital as well. Oklahoma turned to a 500-milligram dose of midazolam, a benzodiazepine sedative used for dental procedures and minor surgeries.
The inmates’ scientific argument focused on a “ceiling effect.” Midazolam, they said, has a dose above which additional drug does not deepen sedation. If that ceiling falls below the level needed to render someone truly insensate, no amount of midazolam could prevent the inmate from feeling the paralytic and heart-stopping drugs that follow.
The problem was proof. The inmates’ expert, Dr. Sasich, acknowledged in his report that he “was unable to determine the midazolam dose for a ceiling effect on unconsciousness because there is no literature in which such testing has been done.” Oklahoma’s expert, Dr. Evans, testified that 500 milligrams of midazolam made it “a virtual certainty” that any person would be unconscious enough not to feel the subsequent drugs.1Justia. Glossip v. Gross The district court credited Oklahoma’s expert, and under the deferential “clear error” standard, the Supreme Court declined to disturb that finding.
The Dissents
The 5–4 split produced two dissents that went in very different directions.
Sotomayor: A “Cruel Catch-22”
Justice Sonia Sotomayor wrote the principal dissent, calling the alternative-method requirement “a cruel ‘Catch-22.'” The state, she said, can shield its chosen protocol from constitutional scrutiny by making sure no alternative drugs are available. When pharmaceutical companies refuse to sell execution drugs to corrections departments, the inmate cannot point to an available alternative, because none exists for anyone. Under the majority’s rule, that failure defeats the claim regardless of how painful the state’s method might be.1Justia. Glossip v. Gross
Breyer: Reopen the Question of the Death Penalty
Justice Stephen Breyer, joined by Justice Ruth Bader Ginsburg, wrote a separate dissent that skipped past the midazolam question. He argued it was “time to reopen the question” of whether the death penalty itself violates the Eighth Amendment, citing serious unreliability in who is sentenced to death, arbitrariness in who is actually executed, and unconscionably long delays. Breyer noted that the average delay between sentencing and execution had grown from two years in 1960 to roughly 18 years by 2015, and that at then-current rates, clearing the existing death row population would take more than 75 years.1Justia. Glossip v. Gross
What Glossip Means for Later Challenges
Four years later, the Court applied the Glossip framework in Bucklew v. Precythe (2019). Russell Bucklew suffered from cavernous hemangioma, a disease that causes blood vessel tumors in the head, neck, and throat, and argued that Missouri’s lethal injection protocol was uniquely dangerous for him. He proposed nitrogen hypoxia as an alternative. The Court rejected his claim, holding that the Glossip two-part test governs every method-of-execution challenge, whether the attack is on the protocol generally or based on an individual medical condition.3Supreme Court of the United States. Bucklew v. Precythe The Court found Bucklew’s nitrogen hypoxia proposal was not “sufficiently detailed to permit a finding that the State could carry it out relatively easily and reasonably quickly.”
Read together, Glossip and Bucklew make federal challenges to lethal injection protocols extraordinarily difficult. The inmate has to design a workable alternative and prove the state can implement it, while corrections departments and pharmaceutical companies control access to the drugs. Several states have added secrecy laws shielding the identity of their drug suppliers, adding another obstacle.
Midazolam itself remains in use. Alabama, Arkansas, Mississippi, and Oklahoma have all authorized or recently used midazolam as the first drug in a three-drug protocol. Arizona agreed after litigation over the 2014 execution of Joseph Wood that it would never again use the drug. Florida and Tennessee have moved to other protocols. Ohio, which used midazolam to execute Dennis McGuire in 2014, saw a federal magistrate bar its further use of the drug in 2017.
What Happened to Richard Glossip
The lead petitioner’s own case took a dramatic turn after the 2015 ruling. Richard Glossip was convicted in 2004 of first-degree murder for allegedly orchestrating the killing of his employer, Barry Van Treese. The prosecution’s case rested almost entirely on the testimony of Justin Sneed, who admitted to committing the murder and received a life sentence for testifying against Glossip.
Oklahoma Attorney General Gentner Drummond later concluded that the prosecutor “knowingly elicited false testimony from Sneed and failed to correct it.” Sneed had testified at trial that he had never been prescribed lithium or seen a psychiatrist, but he had been diagnosed with bipolar disorder and prescribed lithium, and the prosecutor’s own handwritten notes reflected a pretrial conversation in which Sneed mentioned “lithium” and his doctor.4Supreme Court of the United States. Glossip v. Oklahoma An independent investigation also uncovered seven boxes of previously undisclosed documents, including letters suggesting Sneed had considered recanting.
On February 25, 2025, the Supreme Court overturned Glossip’s conviction in Glossip v. Oklahoma, holding that “a criminal defendant is entitled to a new trial when the prosecution knowingly failed to correct false testimony, and this error could have contributed to the verdict.”5Justia. Glossip v. Oklahoma Drummond announced his office would retry Glossip but would not seek the death penalty. On May 14, 2026, an Oklahoma County judge set bail at $500,000, and Glossip was released after 29 years in prison. The retrial remains pending.
The 2015 method-of-execution ruling continues to control lower court decisions on lethal injection, even as the man whose name it carries walked out of prison after the state’s own attorney general conceded his conviction was built on false testimony the prosecution knew about.