Willfully Blind to the Machinery of Death: The State of Execution Challenges After Barr v. Lee

The Supreme Court “has never invalidated a [ ] chosen procedure for carrying out a sentence of death as the infliction of cruel and unusual punishment.” From the firing squad, to the electric chair, to an assortment of lethal injection protocols, it has consistently rejected Eighth Amendment challenges to methods of execution.

Moreover, in recent years the Court has placed an additional hurdle in front of death row prisoners challenging their executions. Even assuming that the planned method of execution is unconstitutionally painful, the Court held in Glossip v. Gross (2015) that a prisoner must identify an alternative method of execution that is “feasible, readily implemented, and . . . significantly reduce[s] a substantial risk of severe pain.”1 In other words, a planned method of execution does not violate the Eighth Amendment unless the death row prisoner can establish both that the method poses a substantial risk of severe pain and that this pain is intolerable in comparison to known and available alternatives.

Given this background, the Supreme Court’s opinion last summer in Barr v. Lee (2020), vacating U.S. District Court Judge Tanya S. Chutkan’s ruling that the federal government’s planned execution protocol constituted cruel and unusual punishment, may come as no surprise. As five members of the Court reasoned in a brief, per curiam opinion issued in the middle of the night, the planned method of execution—a single-drug protocol consisting of a massive dose of the sedative pentobarbital sodium—was similar to execution protocols adopted by multiple states and upheld against Eighth Amendment challenge by several courts of appeals. Additionally, the Court had previously rejected a challenge to the use of pentobarbital in the execution of a prisoner with a unique medical condition in Bucklew v. Precythe (2019). Perhaps even more astoundingly, the district court issued its opinion on the morning of the first scheduled execution. A more last-minute reprieve would be hard to find.

Indeed, both proponents and critics of the Supreme Court’s death penalty jurisprudence have generally agreed that the result in Lee was inevitable. The government, for example, argued in its briefing that Bucklew controlled, and Deputy Attorney General Jeffrey A. Rosen later took to the pages of the New York Times to call the plaintiffs’ claims a “guerrilla war” against the death penalty itself. Several critics, meanwhile, have suggested that cases like Glossip and Bucklew represent the original sin of the Court’s death penalty jurisprudence by making it “nearly impossible” to challenge lethal injection protocols. According to this interpretation, the district court was, for better or for worse, engaged in open defiance of the case law when it enjoined the planned federal executions.

Yet both sides overlook several key innovations that distinguish Lee from the Court’s previous jurisprudence. Glossip, unlike Lee, emphasized deference to the district court’s factual findings. Moreover, while it is certainly possible that the majority in Bucklew intended to foreclose all Eighth Amendment challenges to methods of execution, Justice Brett Kavanaugh concurred separately in that case to specifically describe circumstances in which such a challenge might be successful. Finally, the district court’s opinion was fundamentally distinct from other cases in which the Court has criticized last-minute stays of execution. For all these reasons, the Court’s opinion in Lee is better understood as taking a small but significant step toward a total refusal to examine execution procedures for potential Eighth Amendment violations.

* * *

Even without the Supreme Court’s late-night intervention, the executions at issue in Lee were guaranteed to garner massive public attention. Following a de facto moratorium on federal executions lasting more than fifteen years, on July 25, 2019, the United States Department of Justice announced plans to execute five federal death row inmates using a new Federal Execution Protocol (henceforth, the 2019 Protocol). This protocol replaced the existing three-drug lethal injection procedure with a massive dose of a single drug: pentobarbital.

Predictably, this announcement triggered a flood of litigation by federal death row prisoners. After the government refused to delay the planned executions to allow for discovery, four of the five inmates with scheduled execution dates moved to preliminarily enjoin their executions. In November 2019, the district court granted the motions for preliminary injunction on the basis of the plaintiffs’ claims that the 2019 Protocol violated the Federal Death Penalty Act (FDPA), but did not rule on their remaining claims, including those under the Eighth Amendment.

Both the D.C. Circuit and Supreme Court declined to stay this injunction. That said, Justices Samuel Alito, Neil Gorsuch, and Brett Kavanaugh issued a statement indicating their belief that the government was likely to prevail on the merits of the FDPA issue. In April 2020, a divided panel of the D.C. Circuit vacated the injunction. Curiously, no two judges were able to agree on the statutory requirements of the FDPA. Nonetheless, two judges concluded that the 2019 Protocol did not exceed statutory authority under the FDPA, and so vacated the injunction. The full D.C. Circuit declined to rehear the case en banc in May 2020, and the Supreme Court denied certiorari on June 29, 2020.

While the plaintiffs’ petition for certiorari was pending, the federal government announced new execution dates for three of the four prisoners whose executions had been enjoined the previous November, as well as for a fourth prisoner. Three of these executions, the first since 2003, were scheduled for the week of July 13, 2020.

The district court’s second injunction, In re Federal Bureau of Prisons’ Execution Protocol Cases (2020), thus came as a major shock in a case that appeared to have been resolved, as it granted the plaintiffs’ renewed motion on the basis of their Eighth Amendment claims. When a unanimous panel of the D.C. Circuit declined to vacate the injunction and called for expedited briefing on the merits just before midnight, many—including staffers at the district court—went to bed on July 13 believing that the first of the scheduled executions would not take place. The rest, of course, is history: by the next morning, the Supreme Court had vacated the injunction, and the first prisoner, Daniel Lewis Lee, after remaining strapped to the execution gurney for several hours while last-minute litigation in other courts played out, had been executed.

* * *

What, then, makes the Supreme Court’s order in Lee so special? Let’s begin with the science. Although the Court cautioned in Baze v. Rees (2008) against lower courts becoming “boards of inquiry charged with determining ‘best practices’ for executions,” it would be impossible to evaluate method-of-execution challenges without some weighing of the scientific evidence. Much of Justice Alito’s majority opinion in Glossip, for example, is devoted to affirming the district court’s findings that the lethal injection protocol in that case was unlikely to pose a substantial risk of severe pain. That opinion emphasized deference to the district court’s factual findings under a clear error standard of review.

In Lee, by contrast, the Court’s per curiam opinion makes no mention whatsoever of the district court’s conclusion that the 2019 Protocol was very likely to result in “extreme pain and needless suffering” by causing flash pulmonary edema, a medical condition resulting in “sensations of drowning and suffocation.” Instead, the Court simply pointed to the existence of competing expert testimony by the government arguing that pulmonary edema would not occur until after the prisoners were either dead or incapable of feeling its effects.

This section of the majority opinion is notable for two reasons. First, it relies on the existence of competing scientific evidence as a reason to permit an execution to go forward, rather than as a reason to halt it so that the evidence can be evaluated. While this may seem callous given that the death penalty is, as Justice Sonia Sotomayor argued in dissent, “the most irreparable of harms,” it is true that a preliminary injunction is only justified where the movant has demonstrated a likelihood of success on the merits. If the scientific evidence in a case is truly equally balanced, then the movant has not met this burden, and an injunction should be denied.

Still, this reasoning takes on a more sinister tone when combined with the Supreme Court’s second innovation in Lee: its refusal to credit or even discuss the district court’s factual findings. In a vacuum, the existence of competing expert testimony might be a reason not to enjoin an execution; however, the district court did weigh this testimony, and it concluded that one side’s evidence was stronger. This, after all, is what the district courts are charged to do: evaluate the factual record and apply the law to the facts. Appellate courts, including the Supreme Court itself, are not meant to “reverse the finding of the trier of fact simply because [they are] convinced that [they] would have decided the case differently.” Per the Court’s opinion in Glossip, it should have reviewed the district court’s factual findings for clear error, while emphasizing that petitioners—in this case, the federal government—had the burden of persuasion on the issue.

Even if the Court had chosen not to defer to the district court’s factual findings, its refusal to even discuss them suggests a disturbing development in death penalty jurisprudence. After all, in Glossip, the Court did conduct a detailed evaluation of the scientific evidence; it did not simply apply a rubber stamp to the lower court’s findings. But the Court heard Glossip on the merits: the case was fully briefed—including twelve separate briefs by amici curiae—and argued. The prisoners in Lee had no such opportunity to be heard.

Viewed in context, Lee raises unfortunate implications. It suggests that if a lower court refuses to stay or enjoin a pending execution, the Supreme Court will credit that court’s factual findings; but if a lower court does block an execution, the Supreme Court will simply substitute its own view of the evidence, without even permitting briefing and oral argument regarding that evidence.2

* * *

Because the Supreme Court rejected the district court’s conclusion that the 2019 Protocol posed a substantial risk of severe pain, it did not discuss the second prong of its test in Glossip: the need to identify feasible and readily available alternatives. That said, there is some reason to believe that Justice Kavanaugh may have intended Lee to foreclose an argument that he himself praised in Bucklew. In that case, Justice Kavanaugh wrote separately to “underscore the Court’s additional holding that the alternative method of execution need not be authorized under current state law.” He also specifically noted that death by firing squad might serve as one such alternative, citing dissenting opinions by Justice Sotomayor that praised the firing squad as a more humane method of execution.

In context, Justice Kavanaugh’s concurrence is most sensibly read as establishing a narrow set of circumstances under which he might break from the rest of the Bucklew majority to vote in favor of a death row prisoner’s method-of-execution challenge. The prisoners in Lee, presumably relying on this concurrence, argued—among other alternatives—in favor of execution by firing squad, and the district court concluded that this method was available, readily implemented, and much less likely than the 2019 Protocol to result in severe pain. Like Justice Kavanaugh in Bucklew, the district court emphasized previous dissenting opinions in which Justices had singled out the firing squad as a comparatively humane and reliable method of execution.

It is, of course, impossible to know what each Justice in the Lee majority was thinking, or indeed, who penned the per curiam opinion at all—another consequence of the Court’s decision to vacate the injunction without hearing the case on the merits. But even though the opinion in Lee makes no mention of the alternative methods of execution proffered by the prisoners in that case and approved by the district court, Lee potentially represents an even more restrictive approach to method-of-execution challenges than Justice Kavanaugh hinted at in Bucklew.

The government repeatedly argued in its briefing that the firing squad was not a feasible alternative, because it would require a complete reworking of the planned execution procedure, and because it is a method only employed by one state, Utah, and an infrequently used method at that.3 As the district court observed, this reasoning is tautological. If a death row prisoner is permitted to plead only an alternative method of execution that is currently authorized by law and can be implemented without significant changes to the existing execution protocol, then governments could foreclose all method-of-execution challenges simply by refusing to authorize any other method.

Justice Kavanaugh’s concurrence in Bucklew specifically stated that this argument was incorrect, and that “all nine Justices today agree on that point.” Indeed, the Eleventh Circuit, which had previously held that a proposed alternative method of execution must be authorized under state law, later noted that “Bucklew demonstrates our conclusion . . . was incorrect.” The government in Lee, however, continued to argue in favor of this approach throughout the entire litigation. The Supreme Court’s refusal to address the topic of alternatives in Lee, then, may be read by the lower courts as an implicit blessing to treat this aspect of Bucklew as dictum, and to further bar the courthouse doors to death row prisoners bringing method-of-execution challenges.

* * *

Finally, the Supreme Court’s criticism of the injury stemming from a delayed execution twists the concept of irreparable harm against the very prisoners it should protect. There is nothing new about the Court’s emphasis on the harm caused to the government by a preliminary injunction. The Court’s opinion in Winter v. Natural Resources Defense Council, Inc. (2008) explicitly argued that the public interest outweighed any irreparable injury to marine mammals caused by the United States Navy’s use of sonar in training exercises. Lee did not introduce this reasoning to the topic of capital punishment. The Court has long been critical of last-minute execution delays, holding in Hill v. McDonough (2006) that the “last-minute nature of an application,” or an application that “could have been brought” earlier, is a valid reason to deny a stay of execution. Reiterating this holding in Bucklew, the Court concluded that “last-minute stays should be the extreme exception, not the norm.”

Lee’s innovations are twofold. First, as a case in which the government, not the prisoners, petitioned the Supreme Court for last-minute relief, it represents an expansion of the Court’s willingness to use its “shadow docket”—the term Professor William Baude has coined for the Court’s orders and summary decisions not heard as merits cases—to grant extraordinary relief to the federal government. As Professor Stephen Vladeck has explained, the Solicitor General under the Trump administration applied for such emergency relief far more aggressively than during the Bush or Obama administrations. In the past year, in particular, the Supreme Court has granted a number of last-minute stays to the federal government, as several commentators have observed. The Court’s willingness to do so in the arena of capital punishment, where the harm is, to quote Justice Sotomayor in Lee, “most irreparable,” threatens to prevent death row prisoners from bringing suit at all once they have a scheduled execution date.

This might not be a major problem if all last-minute petitions from death row prisoners were dilatory in nature. But Lee expands the Supreme Court’s distaste for the legal tactics of death row plaintiffs to a situation in which the delays were caused by the lower court and by the federal government itself, not by the plaintiffs. The federal government repeatedly referenced in its briefing the lengthy period of time between the plaintiffs’ sentences of death and their scheduled executions. Yet it was the government that waited eight years to announce a new execution protocol and then insisted on scheduling three executions in a single week. And far from bringing last-minute claims, the prisoners in Lee challenged the 2019 Protocol almost immediately after it was announced, months before their initially scheduled executions.

It is understandable that the Supreme Court would be displeased by a preliminary injunction granted on the morning of a scheduled execution, but it was the government that scheduled new execution dates—again, three in a single week—while the prisoners’ petition for certiorari at the Supreme Court was still pending, and the prisoners filed their renewed motion for a preliminary injunction just four days later. Once briefing concluded on July 1, any delay was the fault of the district court alone, not the prisoners. Lee thus suggests that the Court will treat the very existence of a delay in execution, regardless of explanation or fault, as a reason to reject claims brought by death row prisoners.

* * *

It is tempting for critics of the Supreme Court’s death penalty jurisprudence—and I suspect this Essay has revealed me as one of them—to direct their ire at the fundamental cases underpinning the Court’s approach, dismissing short opinions like Lee as the inevitable consequences of those fundamentals. This approach naturally lends itself to the conclusion that the current state of capital punishment in the United States cannot be fixed without overturning Glossip, or McCleskey v. Kemp (1987), or even Gregg v. Georgia (1976). It is likely no coincidence that it was Glossip in which Justices Ruth Bader Ginsburg and Stephen Breyer first announced their belief that the death penalty, “in and of itself,” violates the Eighth Amendment. These two Justices have since reiterated this stance in other capital punishment cases to come before the Court, including Lee.

This is one approach. The duty of a lawyer, however, is to represent a client, not to throw up one’s hands and conclude that no prisoner can ever successfully challenge their method of execution on Eighth Amendment grounds under current precedent. In this respect, more credit is perhaps owed to Justice Sotomayor, who has instead opted to argue against each subsequent restriction placed on death row prisoners’ Eighth Amendment rights, and in doing so provide a roadmap for future advocates to continue to argue against potentially inhumane execution procedures.

When Justice Harry Blackmun, dissenting in Callins v. Collins (1994), wrote that he “no longer shall tinker with the machinery of death,” this opposition to the very institution of capital punishment was in marked contrast to the approaches of his fellow Justices on either side of the issue. Other Justices, Blackmun reasoned, were simply tinkering around the edges, fighting over the scope and interpretation of procedural rules rather than engaging with the fundamentals of the death penalty as practiced.

Today, however, it is the Supreme Court’s pro-death penalty majority that refuses to tinker with the machinery of death. Its recent jurisprudence has consisted primarily of permitting executions to be carried out before they can be reviewed for potential infliction of cruel and unusual punishment. With Lee, the Court takes another small but significant step toward ensuring that the federal courts cannot look too closely at execution procedures. Perhaps, in this polarized age, it was inevitable that the Court would ultimately descend into two camps: a majority committed to nonreview of the death penalty, and a minority demanding its abolition. But if the majority has abdicated its oversight over execution procedures, then it is the dissenting Justices who must take up the duty of continuing to tinker.

• 1This language stems from an earlier opinion, Baze v. Rees (2008); however, it did not carry a majority of the Court until Glossip.
• 2The lower courts have also struggled to interpret the Supreme Court’s casual dismissal of the district court’s factual findings. In November 2020, the D.C. Circuit held that the Court had not intended to rule definitively on whether the 2019 Protocol was unconstitutionally painful, and so reinstated the remaining prisoners’ Eighth Amendment claims. Yet it also declined to enjoin any of the remaining scheduled executions, and six more prisoners were executed between November 2020 and January 2021.
• 3The firing squad is also authorized by statute in Mississippi and Oklahoma as a fallback method, should the state’s lethal injection protocol be deemed unconstitutional. Utah is the only state to have conducted an execution by firing squad since 1976; the most recent was that of Ronnie Lee Gardner on June 18, 2010.