The Court and the Killing State

Eric Berger* June, 2026

Secrets of the Killing State: The Untold Story of Lethal Injection. By Corinna Barrett Lain. New York: New York University Press. 2025. Pp. vii, 359. $29.95.

Introduction

If you haven’t paid close attention, you may assume that lethal injection brings about a peaceful and humane death. Professor Corinna Barrett Lain’s fabulous new book, Secrets of the Killing State: The Untold Story of Lethal Injection,¹ dispels that myth. Lain argues thoroughly and persuasively that lethal injection does not work like it is supposed to. Far from producing painless unconsciousness, lethal injection can inflict agony.

Many people don’t know this, in part because death penalty states like to keep their execution procedures secret. States conceal key information about their drugs, execution team members, and protocols. Some state protocols also include a paralytic, which can make an agonizing death appear peaceful.

State officials don’t do this because they’re sadistic, but rather, as Lain demonstrates, because they don’t know what they’re doing. Given political pressures to carry out death sentences, states often proceed with a deeply flawed protocol rather than delay executions. As a result, there have been several gruesome, botched executions in recent years. In light of all this and much more, Secrets of the Killing State concludes that lethal injection in the United States is broken and cannot be fixed (pp. 4, 273–74).

Professor Lain is one of the country’s leading death penalty scholars, and her volume is the best and most comprehensive treatment of lethal injection.² She examines closely the dangerous drugs, incompetent executioners, political pressures, drug supplier saga, and many other twists and turns. “Rest assured,” Lain tells us, “I brought the receipts” (p. 5).

Secrets of the Killing State lays out those receipts in meticulous detail. The book is mostly descriptive, but the sum of those receipts adds up to her thesis, which is that most states don’t have a clue how to put together a lethal injection protocol, but they do it anyway, resulting in torturous executions with alarming frequency. For lawyers, judges, reporters, and academics who need to know more about lethal injection, Secrets of the Killing State is an invaluable resource. Twenty years ago, while litigating a lethal injection case as a junior attorney, my colleagues and I tried to convince a skeptical judge that an execution procedure that looked peaceful might in fact cause agony. Lain’s book would have made our job much easier.

Secrets of the Killing State is also a great read, easily accessible to lay readers. Lain’s prose is snappy and pert. Law professors aren’t known for writing page-turners, but Secrets of the Killing State is one. The more you learn about states’ incompetence, the more you want to see what happens next. Were Arizona’s and Nebraska’s efforts to procure drugs off the international black market successful (pp. 152–59)? What happened when Oklahoma discovered it had conducted an execution using the wrong drug (pp. 225–29)? What do postexecution autopsies reveal (pp. 18–26)? Grisly and macabre, this stuff is real-life American gothic. Secrets of the Killing State is not for the fainthearted, but it’s an awfully compelling read.

For all the book’s merits, there is, perhaps, a disconnect between its voluminous, gruesome detail, on the one hand, and Lain’s breezy prose, on the other. Lain sounds like she’s writing for the beach, but she provides a scholarly level of ghastly evidence. All Americans ought to know something about the death penalty—the United States is one of the few Western democracies that still retains capital punishment—but at 359 pages, Lain’s book may be too meaty for casual readers and too breezy for serious ones. That said, style also serves a purpose, and most readers likely will appreciate Lain’s cheeky commentary, which keeps this morbid subject lively and entertaining.

If Lain skimps on any part of the story, it’s the role of courts. She could have devoted a whole chapter to judicial decisions but instead only addresses them in passing (pp. 237–40). While litigation pops up occasionally, Lain never tackles the question many readers will ask: How is this even legal? Put another way, if states don’t know what they’re doing and consequently inflict excruciating pain on the condemned, why don’t judges order them to stop?

Lain has her reasons not to spend too much time on legal cases. She is telling an intricate story as it is, and lethal injection’s risks matter, regardless of what courts do. They certainly matter to the condemned prisoner who will suffer an agonizing death. They also illuminate our culture’s disregard for the suffering of society’s most despised. And they matter to the broader politics of capital punishment. Lethal injection problems are one reason that, by multiple measures, the U.S. death penalty has declined dramatically in recent decades.³ Since 2000, the number of executions, death sentences, and death penalty states have all dropped substantially, as has popular support for capital punishment.⁴

But the story of lethal injection also illustrates something important about courts, a phenomenon that transcends the death penalty. Not only has the Roberts Court rejected Eighth Amendment lethal injection challenges,⁵ but, in so doing, it has also signaled that it doesn’t care about the details of the procedures. Following the Supreme Court’s lead, lower courts usually reject these challenges too.⁶ You may wonder how the Court can uphold all those protocols given the many lethal injection problems Lain identifies. The answer is that the Court simply ignores them.

This Review picks up where Secrets of the Killing State leaves off, arguing that the Supreme Court has designed its Eighth Amendment method-of-execution doctrine to avoid grappling with facts. Lain has written an important book about the facts of lethal injection, but the Supreme Court doesn’t want those facts to make a legal difference. To the contrary, the Court’s priority is to permit states to continue executions without judicial interference.

The Court’s super-deference here comes with grave costs. The Eighth Amendment prohibits punishments that are “cruel and unusual.”⁷ Part of this inquiry—the “cruel” part—requires knowing whether an execution is likely to inflict pain. That’s a factual inquiry, but the Court has, in essence, rewritten the Eighth Amendment to take facts off the table. The Court has some reason to worry about strategic litigation obstructing executions, but it has rewritten the law solely with that concern in mind. Indeed, the Court’s fact-lite jurisprudence essentially eviscerates the Eighth Amendment right in the method-of-execution context.

Part I of this Review summarizes crucial lethal injection facts. It begins by exploring why lethal injection is difficult to write about. Lain does not explicitly address the inherent challenges her subject poses, but a direct discussion of those challenges helps readers better appreciate the topic’s complexity. Part I then discusses dangerous drugs and incompetent personnel, which collectively create a substantial risk of serious pain.

Part II turns to courts’ role in this story. Specifically, it argues that the Supreme Court has turned a blind eye to the facts Lain recounts, crafting Eighth Amendment doctrine to minimize the importance of those facts. This precedent makes it nearly impossible for capital inmates’ method-of-execution challenges to prevail, no matter how likely a protocol is to inflict agony.

Part III observes that the Court’s avoidance of key facts extends beyond lethal injection to many other constitutional cases. Though this is not a recent phenomenon, it is especially pronounced in the Roberts Court. By minimizing the relevance of facts in various constitutional cases, the Court frequently resolves important issues without the benefit of crucial context. Such fact-lite review is always problematic, but it is especially worrisome in an era when the Court increasingly confronts cases with profound implications for the future of American democracy and the rule of law.

I. Lethal Injection Facts

A. The Inherent Challenges of Lethal Injection Facts

Eighth Amendment scholars face several challenges in describing lethal injection. The first is utilizing language that conveys lethal injection concepts succinctly but accurately. This is surprisingly difficult. For example, scholars sometimes refer to the procedure’s “dangers,”⁸ which sounds like euphemistic understatement. Of course lethal injection is “dangerous;”⁹ the whole point is that it kills people.

“Dangerous” in the Eighth Amendment context, however, connotes an execution method that creates a “substantial risk of serious harm”¹⁰—though (and this is another complication) the Supreme Court has altered the Eighth Amendment standard repeatedly, so one cannot articulate it succinctly with much confidence.¹¹ Putting aside the doctrinal intricacies, a “dangerous” lethal injection procedure is not one that merely kills—they all do that—but one that exceeds certain (admittedly imprecise) thresholds for the degree of risk and amount of pain. The flip side is also true. A “safe” lethal injection protocol is one that does not exceed those thresholds.

This language can seem Orwellian.¹² A lethal injection procedure, quite obviously, cannot be considered “safe” in the ordinary sense of the word.¹³ Nevertheless, words like “dangerous” and “safe” provide convenient shorthand, and so we use them, even though they whitewash the fact that the procedures extinguish human life.

Another challenge is that risk and pain are impossible to quantify with precision, and yet the constitutional question hinges on the amounts of risk and pain.¹⁴ Part of the difficulty is that executions inevitably involve some pain and some risk of severe pain, so the Supreme Court has signaled that the Eighth Amendment must tolerate some of each.¹⁵ As the Court has repeatedly said, “because it is settled that capital punishment is constitutional, ‘[i]t necessarily follows that there must be a [constitutional] means of carrying it out.’ ”¹⁶

Consequently, some amount of pain is constitutionally tolerable. For instance, the modest pain resulting from the insertion of a catheter into a person’s vein ordinarily would not raise constitutional concerns (though arguably it does for people whose medical conditions make such insertions unusually problematic).¹⁷ Only when the pain is serious—commentators and courts often refer to “excruciating,” “agonizing,” or “intolerable” pain—does it become constitutionally problematic.¹⁸ Once again, though, neither judges nor scholars are entirely consistent in how they refer to the degree of pain.¹⁹ Even if they were, it is impossible to measure when pain passes the threshold from, say, “unpleasant” to “excruciating,” both because pain is inherently subjective and because we cannot interview the deceased after the execution.

Likewise, it is unclear precisely what level of risk becomes constitutionally problematic or the degree of risk each state’s protocol poses. Under the doctrine, a small risk of even severe pain does not trigger constitutional concerns. Rather, the Court has held that for an execution procedure to pose a constitutional problem, the risk of severe pain must be “substantial.”²⁰ As with the amount of pain, though, the degree of risk cannot be empirically quantified. Nevertheless, Eighth Amendment law purports to distinguish between acceptable and unacceptable degrees of risk.

Yet another challenge is that the landscape changes quickly. Twenty-seven states plus the federal government retain the death penalty,²¹ and while some of those states have not carried out any executions in years,²² many that conduct regular executions tend to alter their protocols frequently.²³ Just when you think you have a handle on a protocol, the state may start over with something entirely new, oftentimes because it couldn’t find the drugs for the old protocol. And while some states’ written protocols resemble each other because they include the same drugs,²⁴ each state has its own personnel, architecture, equipment, and so on. Those details really matter. A theoretically “safe” protocol can inflict agony if administered by incompetent personnel or with improper equipment (pp. 125–40).

Lethal injection is a moving target, too, because the scientific understanding of the drugs’ risks can change. Most lethal injection executions use either a one-drug protocol or a three-drug protocol.²⁵ The one-drug protocol delivers an overdose of an anesthetic.²⁶ The three-drug protocol kills the prisoner with an agonizing drug that stops the heart.²⁷ For years, the consensus was that one-drug protocols posed fewer dangers than three-drug protocols.²⁸ After all, whereas the third drug in the three-drug protocol causes excruciating suffering in the underanesthetized, the one-drug protocol utilizes only the anesthetic to put prisoners to sleep. Of course, states still need to administer the one-drug protocol competently, but the basic assumption was that removing the painful drugs significantly decreased the procedure’s risks.

As Lain recounts, newer studies call that assumption into question (pp. 88–97). Recent research suggests that when humans quickly receive a large dose of a drug, the result can be pulmonary edema, which creates the sensation of “drowning from within” (pp. 88–89). A 2020 study of over 200 execution autopsies revealed that pulmonary edema occurred in 84 percent of cases, including in 49 out of 58 one-drug executions using pentobarbital.²⁹ A different report concluded that the federal execution protocol is “virtually certain to cause prisoners excruciating suffering through their awareness of the sensations of suffocation and drowning caused by pulmonary edema.”³⁰ Pulmonary edema is actually a possibility with any protocol, but it is especially relevant to our understanding of the one-drug protocol because that method had previously been lauded as comparatively safe. These new concerns prompted Attorney General Garland to rescind the federal government’s one-drug execution protocol in the final days of the Biden administration.³¹

The existence of pulmonary edema, though, does not necessarily mean that the prisoner suffered; an unconscious person might not have experienced the drowning sensation, even though fluid accumulated in his lungs. The key question for these executions, then, is whether the prisoner was sensate to that drowning.³² Experts disagree.³³ The research probably isn’t conclusive either way, but the risk of pulmonary edema is nevertheless a complication that wasn’t on the radar a half-dozen years ago. A scholar like Lain, then, must keep track of both new state execution protocols and the latest scientific studies.

All of these complications lead to the scholar’s final challenge: Lethal injection’s many problems are both technical and interconnected. Because they are technical, they require careful explanation, and because they are interconnected, the author is tempted to present everything together before explicating all of those technicalities. We can see Lain navigating these sorts of challenges throughout her book. For example, Lain introduces the drug midazolam in chapter one and touches on it again in chapter two, but it is not until chapter three that we get a full explanation of why that drug choice is so problematic. Lain invariably circles back to tie up loose threads, but readers will sometimes have to be patient.

Professor Lain ultimately handles her subject’s many challenges admirably, providing a thorough and eminently readable account of lethal injection’s problems. The next two Sections summarize two of those problems to provide some important examples of facts the Court’s Eighth Amendment decisions ignore. Section B explores the dangers created by the states’ drugs, and then Section C examines the incompetent personnel who often administer lethal injection.

B. Dangerous Drugs

When Oklahoma became the first state to adopt lethal injection in 1977, it adopted a three-drug protocol that other states soon copied.³⁴ The three-drug protocol was ubiquitous in death penalty states until around 2010 when some states started experimenting with other protocols, usually involving just one drug.³⁵ Though most experts agree that the one-drug protocol poses fewer risks, the three-drug protocol nevertheless remains common today.³⁶ Sixteen states with the death penalty have used the three-drug protocol for their most recent lethal injection executions.³⁷ More importantly, several states that regularly carry out executions retain the three-drug protocol, including Alabama, Arkansas, Florida, Mississippi, Ohio, and Oklahoma.³⁸

The three-drug protocol is, in Lain’s words, “torturous” (p. 59). To understand why, it’s easiest to start with the third drug and move backwards. It is undisputed that the third drug, potassium chloride, causes excruciating pain as it courses through a person’s veins on its path to stopping the heart.³⁹ Some have likened it to being burned alive from the inside.⁴⁰ For a three-drug execution to be humane, then, the other drugs would need to anesthetize the prisoner from that agony.

The second drug certainly doesn’t anesthetize. It is a paralytic (often pancuronium bromide or vecuronium bromide), so it conceals the prisoner’s pain from onlookers.⁴¹ The paralytic is an important reason why we can’t know from simple observation whether a prisoner’s death is painful.⁴² To make matters worse, because paralytics also paralyze a person’s diaphragm, they can create the terrifying sensation of suffocation, adding still more suffering.⁴³

The safety of the three-drug protocol, then, hinges on the first drug, which is supposed to anesthetize the prisoner. Over the years, different states have tried various drugs for this purpose. Lain notes that sodium thiopental (also known as “thiopental”) was long considered “the gold standard” first drug (p. 70). An ultrashort-acting barbiturate, thiopental was used by anesthesiologists for short surgical procedures, so it seemed like a suitable choice for the three-drug protocol (p. 70). As we’ll see, though, barbiturates work only if competent personnel administer them correctly, and states often don’t employ competent personnel.⁴⁴

Eventually states ran into another problem, which was that they couldn’t get thiopental anymore, so they searched for replacements (pp. 142–45). They tried various solutions like pentobarbital, another short-acting barbiturate, but found that access to those alternative drugs soon disappeared as well (p. 147). Concerned about their standing with institutional investors, pharmaceutical companies didn’t want their drugs to be used for executions.⁴⁵ Likewise, European governments prohibited the export of drugs for lethal injection.⁴⁶ Time and again, states would identify a new source of drugs only to find that the source had dried up.

State officials tried to improvise, but they didn’t always do this well. The craziest—but most illustrative—decision was to use midazolam as the three-drug protocol’s first drug. Unlike thiopental, midazolam is not a barbiturate and is not approved for use as a sole anesthetic in a painful procedure.⁴⁷ Doctors often give it to reduce anxiety prior to the delivery of anesthesia.⁴⁸ Midazolam might produce mild anesthesia so that the subject stops responding to verbal prompts, but it cannot reliably protect against excruciating stimuli like surgery or injected potassium chloride.⁴⁹

Midazolam is a bonkers choice for a three-drug protocol’s first drug. In Oklahoma, the decision to use midazolam was made not by a doctor or pharmacist, but by a Department of Corrections lawyer, who later admitted he worked under intense time pressure and didn’t consult medical personnel.⁵⁰ The lawyer testified that he “looked on-line . . . . Went past the key Wiki leaks, Wiki leaks or whatever it is . . . .”⁵¹

Oklahoma’s decision to begin its three-drug protocol with midazolam contributed to the botched execution of Clayton Lockett, a horrendous episode Lain recounts in vivid detail (pp. 7–29). When officials administered the potassium chloride, Lockett began writhing on the gurney, arching his back and clenching his jaw. (p. 23). He tried to break free from his restraints with a “gigantic” jerk (p. 24). When it was evident that Lockett was not anesthetized, the State closed the execution chamber’s blinds (p. 24). Lockett did not die until forty-three minutes after officials had started administering the drugs (p. 26). Remarkably, despite this highly publicized botch, Oklahoma used midazolam again when it executed Charles Warner. His final words were “My body is on fire.”⁵²

Notwithstanding Oklahoma’s egregious problems, seven other states used midazolam in executions over the following months and years.⁵³ Oklahoma itself halted executions for over six years after the Lockett and Warner fiascos, but when it finally resumed in fall 2021, it once again opted for a three-drug protocol beginning with midazolam.⁵⁴ Predictably, problems continued. John Grant convulsed and vomited for fifteen minutes during another Oklahoma execution using a three-drug protocol beginning with midazolam.⁵⁵ Likewise, during an Alabama execution with midazolam, Ronald Bert Smith Jr. gasped and coughed for thirteen minutes, clenching his fists and raising his head.⁵⁶ This suffering is the direct result of the drugs states selected.

C. Incompetent Personnel

Incompetent personnel compound these problems. As we already have seen, state officials sometimes select the wrong drugs.⁵⁷ Beyond that, though, state personnel also need to administer the drugs safely, and they often lack the training and competence to do so. Simply put, unqualified personnel heighten the risk of a botched execution.

Before an execution begins, states need to prepare the drugs for injection. Some drugs are supplied in powder form. To make such drugs injectable, staff must mix the powder into fluid.⁵⁸ For over a decade, that task in Missouri fell to Dr. Alan Doerhoff (p. 125). To Missouri’s credit, its execution team included a doctor. Less to its credit, Doerhoff came with red flags; two different hospitals had revoked his hospital privileges (p. 126). Doerhoff, it turned out, was not particularly competent. At a deposition, he admitted that when he mixed drugs for lethal injection he “sometimes transpose[d] numbers” and often made mistakes.⁵⁹ Indeed, Doerhoff sometimes injected less than half the anesthetic Missouri’s protocol called for.⁶⁰

To inject the drugs, states also need to set a catheter into the prisoner’s veins. Because many prisoners have compromised veins due to medical problems or a history of intravenous drug use, this can be a difficult task (p. 107). States, though, sometimes assign this task to personnel with suboptimal skills. For example, some states employ phlebotomists to set the IV line (p. 132). A phlebotomist’s job is to draw blood, so a competent phlebotomist is surely more capable than, say, a prison guard. Setting an IV line, however, is a “much more complex process” than drawing blood.⁶¹ As Lain explains, most states don’t permit phlebotomists to set an IV line in medical settings (p. 133).

Perhaps it should not be surprising, then, that states have had numerous problems setting IV lines. During a 2009 attempted execution, Ohio stuck Rommel Broom with a needle eighteen times.⁶² Blood ran down his arm, and one stick hit his ankle bone, causing him to scream out in pain.⁶³ Ohio ultimately gave up; Broom later died of COVID-19.⁶⁴ A similar story played out in Alabama in 2018 when executioners tried to find a vein for two-and-a-half hours, sticking Doyle Lee Hamm eleven times, likely puncturing his bladder and femoral artery before finally giving up.⁶⁵

When states can’t access a peripheral vein, they sometimes try instead to access a central vein using one of two methods. States sometimes attempt a “cut-down,” which involves cutting through skin, tissue, fat, and muscle to the underlying central vein.⁶⁶ Other times states attempt to set a central line, which requires using a larger needle, a guide wire, and additional steps to locate the central vein (p. 111). Both methods require specialized equipment and surgical experience that execution teams usually lack, but that doesn’t stop states from trying anyway (pp. 106–15). The result, as Lain puts it, is often “a lot of suffering,” almost literally death by a thousand cuts (p. 114).

Horrifically, states sometimes continue with the execution even when they have been unable to set the catheter properly (p. 113). When they do, the drugs can infiltrate the tissue surrounding the vein (p. 112). Infiltration is excruciatingly painful, sometimes resulting in chemical burns on the prisoner’s skin and/or golf ball-sized bulges at the point of infiltration.⁶⁷

Infiltration also results in some of the intended anesthetic not fully entering the prisoner’s veins.⁶⁸ Consequently, infiltration during three-drug executions can cause the inmate to be underanesthetized, which in turn makes him more vulnerable to potassium chloride’s searing pain.⁶⁹ For example, after Ohio’s execution team failed to site the catheter properly in Joseph Clark’s veins and then began administering the three-drug protocol, Clark started convulsing, crying out, “It don’t work. It don’t work.”⁷⁰

Other steps in the procedure also require expertise that state actors often lack. The timing of “pushing” each drug can be tricky. Inject the drugs too quickly, and the anesthetic might not take full effect before the injection of the second and third drugs, meaning that the prisoner would not be properly anesthetized.⁷¹ Inject the drugs too slowly, and the anesthesia might wear off before the potassium chloride enters his veins.⁷²

There are also dosage issues. To ensure a barbiturate’s anesthetic efficacy, anesthesiologists routinely adjust the dosage to account for factors like weight and anxiety level.⁷³ Only anesthesiologists have the expertise to navigate these timing and dosage complications, but states rarely include anesthesiologists on their execution teams (p. 132).

Anesthesiologists would also be helpful in monitoring prisoners’ anesthetic depth. Modest anesthesia is sufficient to render a person insensate to light stimuli, like gentle shaking, but not to stronger stimuli, like severe pain.⁷⁴ One enormous danger of the three-drug procedure is that execution officials may incorrectly believe a prisoner is fully anesthetized when, in fact, he can still feel the potassium chloride’s intense burning. Most execution teams don’t include an anesthesiologist, though, so they assume that the prisoner is insensate when he appears asleep.

In short, safe executions require competent personnel, which many states lack. Years after he designed the nation’s first three-drug protocol, Jay Chapman lamented, “It never occurred to me when we set this up that we’d have complete idiots administering the drugs.”⁷⁵ Blame the personnel, not the protocol, Chapman tells us.

Chapman’s critique is a bit off the mark. Part of the problem with the three-drug protocol is that it requires a high degree of expertise. Chapman designed an execution protocol that demands skills most execution team members simply lack.

The problem, then, isn’t that executioners are idiots, but that states have assigned them a difficult task for which they are usually unqualified.⁷⁶ As Lain puts it, most parts of the execution protocol are “left to nonmedical prison personnel for whom there are no minimum qualifications at all” (p. 131). Even when states persuade medical personnel to assist, they usually don’t have the specialized expertise necessary to monitor anesthetic depth properly or to insert a catheter into a central vein.⁷⁷

The main reason execution teams lack expertise is that most medical personnel don’t want to help kill people. The Hippocratic Oath forbids doctors from doing harm, and most doctors interpret it to forbid them from assisting with executions, even if their participation might minimize suffering.⁷⁸ Numerous medical organizations have issued strong statements opposing doctor and nurse involvement in executions.⁷⁹ While some medical professionals may choose to take part nonetheless, most don’t.⁸⁰ As a result, many states are left to rely on nonmedical personnel to carry out tasks that require medical training.

It should be no surprise, then, that things often go wrong.⁸¹ Austin Sarat estimated that 75 of the country’s 1,054 lethal injection executions as of 2010 had been botched.⁸² At over 7 percent, the lethal injection botch rate was higher than that of any other execution method.⁸³ A newer study of executions from 2022 found a 37 percent botch rate.⁸⁴ And remember that these botch rates would not include seemingly peaceful three-drug executions in which the prisoner was paralyzed but might have experienced excruciating pain. The actual rate of excruciating executions, then, may be higher, perhaps significantly so. Quite simply, lethal injection can and does inflict agony.

II. Lethal Injection Law

Professor Lain’s discussion of all these troubling facts raises an inevitable question: Do courts really tolerate this?

In a word, yes. Despite voluminous evidence of lethal injection’s dangers, courts usually reject Eighth Amendment challenges. Indeed, the Supreme Court has issued a series of increasingly deferential opinions, signaling to executioners that they can do what they want. The more we learn about lethal injection’s dangers, the less the Supreme Court seems to care. If anything, the Court has designed Eighth Amendment doctrine to ignore the very facts Lain presents.

For years, Supreme Court case law on lethal injection was thin. The Court’s few Eighth Amendment method-of-execution decisions before the lethal injection era were cryptic.⁸⁵ After Oklahoma adopted lethal injection in 1977,⁸⁶ the Court avoided ruling on the merits of Eighth Amendment challenges to lethal injection for three decades⁸⁷—though it occasionally decided related cases.⁸⁸ Finally, as lower courts began identifying serious problems with state lethal injection procedures in the mid-2000s,⁸⁹ the Supreme Court decided to get involved.

The Roberts Court gave an early clue as to how it would consider lethal injection facts before it ever decided such a case. In 2007, the Court was considering two certiorari petitions involving similar three-drug protocols beginning with thiopental. In one case, Taylor v. Crawford, there was a developed record of Missouri’s incompetence.⁹⁰ The trial court, in fact, had written:

After learning more about how executions are carried out in Missouri, through the interrogatories submitted to the John Doe defendants, reviewing the chemical dispensary logs, reviewing the videotape of the execution chamber and listening to the testimony of John Doe I, and to the testimony of the other expert witnesses . . . , it is apparent that there are numerous problems.⁹¹

The record in the other case, Baze v. Rees, was thinner.⁹² Though bench trials occurred in both cases, the Baze trial court’s findings of fact mostly summarized the written protocol.⁹³ Unlike the Taylor trial court, it did not closely examine the personnel’s qualifications or practices.⁹⁴

The Court granted certiorari in Baze, the thin case.⁹⁵ Because the three-drug protocol’s safety depends on the proper administration of the anesthetic,⁹⁶ a court can’t meaningfully review it in the abstract. A three-drug protocol beginning with thiopental can potentially be painless, if the state ensures that the prisoner is properly anesthetized. That same protocol, though, can be agonizing if the prisoner is insufficiently anesthetized. (pp. 66–72).

The “safety” of the protocol, then, boils down to the details of administration. Without any evidence about how a particular state administers its protocol, courts have little to go on. The Supreme Court nevertheless took the case without a detailed record about protocol implementation.⁹⁷

The Court’s decision to grant certiorari in Baze is even more notable against the backdrop of contemporaneous district court rulings identifying serious problems with state lethal injection protocols. In addition to Taylor, a federal district court concluded in 2006 that “the [execution] team members [in California] almost uniformly have no knowledge of the nature or properties of the drugs that are used or the risks or potential problems associated with the procedure.”⁹⁸ Also that same year, another federal court found that Tennessee’s protocol “poses a substantial risk that [the prisoner] will not be unconscious when the second and third drugs are administered” because the State failed to check for consciousness before delivering the second and third drugs, failed to select adequately trained executioners, and failed to monitor the drug administration.⁹⁹

It would have been harder for the Court to avoid grappling with lethal injection facts had it taken a case like Taylor with more detailed district court findings about the procedure’s dangers. The Court avoided that problem by taking the case without such findings. Baze, then, might have been an unfriendly grant, a strategic move by pro-death penalty justices to take a case with a weak record. By taking the weak case, the Court seemed like it was trying to avoid inconvenient facts.

Whatever its motivation, the Baze decision made it harder for plaintiffs to prove an Eighth Amendment violation. At quick glance, Chief Justice Roberts’s Baze plurality opinion seemed to restate the Eighth Amendment test requiring a “substantial risk of serious harm.”¹⁰⁰ The plurality, though, added that the plaintiff must proffer an alternative method of execution, which “must effectively address a ‘substantial risk of serious harm.’ ”¹⁰¹ In other words, it was no longer sufficient for the plaintiff to show that the State was planning to execute him with a dangerous procedure; he also had to suggest an alternative procedure that was “feasible, readily implemented, and in fact significantly reduce[s]” that risk.¹⁰²

The Court’s next foray into lethal injection came in 2015. Glossip v. Gross involved Oklahoma’s three-drug protocol beginning with midazolam.¹⁰³ As protocols go, it’s hard to do worse than this one,¹⁰⁴ but the Court was untroubled.

To begin, Glossip ratcheted up the Eighth Amendment risk standard. Whereas the Baze plurality had asked whether the procedure created a “substantial risk of serious harm,”¹⁰⁵ Glossip required that the plaintiffs prove that the challenged protocol was “sure or very likely to cause serious illness and needless suffering.”¹⁰⁶ After Glossip, it wasn’t enough to show a “substantial risk” of serious pain; the pain had to be “sure or very likely.”¹⁰⁷

Though this standard is high, it would still seem to allow some room for courts to consider an execution protocol’s particulars (assuming the plaintiff had proffered an alternative). Glossip, however, made other moves to discourage serious factual inquiry. For one, by including the phrase “needless suffering,” the Glossip formulation implicitly gave states some wiggle room. Justice Alito took pains to explain that Oklahoma and other states had turned to midazolam because “[a]nti-death-penalty advocates” had successfully cut off those states’ access to thiopental and pentobarbital.¹⁰⁸ In relating this (misleading) story,¹⁰⁹ Alito signaled that whatever suffering Oklahoma’s protocol inflicted was not “needless” because circumstances forced Oklahoma to select an imperfect drug. In Justice’s Alito’s telling, anti-death penalty activists had given the State no choice, so courts shouldn’t judge Oklahoma’s protocol too harshly.

Glossip made still other moves that limited the legal relevance of facts. For example, drawing on Baze, it emphasized that “federal courts should not ‘embroil [themselves] in ongoing scientific controversies beyond their expertise.’”¹¹⁰ Lethal injection cases, the Court explained, “test the boundaries of the authority and competency of federal courts.”¹¹¹ Its point was that courts should defer to the political branches, which presumably do have such expertise.

This move, too, steered around the facts, but in a different way. As Lain establishes, state officials designing and implementing execution protocols in many states manifestly lack expertise, certainly in Oklahoma (pp. 9–13, 125–31, 139–40). Ordinarily, courts premise deference to administrative actors on their expertise.¹¹² The Roberts Court, however, extended such deference to Oklahoma’s correctional department without examining whether such deference was deserved.¹¹³ This reflexive deference would be noteworthy in all events, but it is even more so given that the Roberts Court often denies deference to other administrative actors, including those who likely do have genuine expertise.¹¹⁴ Remarkably, the current Court’s skepticism of administrative agencies does not extend to state departments of corrections implementing lethal injection, even though those departments’ degree of expertise over lethal injection is exceedingly low.¹¹⁵

Finally, in keeping with its hyperdeferential approach, the Court didn’t even seem troubled by recent botched Oklahoma executions. The Court brushed aside the Lockett botch for two entirely unpersuasive reasons.¹¹⁶ First, it noted that the Lockett problems were “primarily” due to difficulties obtaining IV access,¹¹⁷ even though, as we have seen, IV access is a recurrent problem in lethal injection.¹¹⁸ Second, the Court seemed comforted by the fact that Oklahoma had since increased the amount of midazolam,¹¹⁹ even though midazolam at any dose cannot “render[] a person insensate to [the] pain caused by the second and third [lethal injection] drugs.”¹²⁰ Nor did the Court grapple with the problem of botched executions more generally.

That brings us to Bucklew v. Precythe,¹²¹ which added two additional wrinkles discouraging close attention to execution protocol details. First, the Court announced that the Eighth Amendment’s primary concern is “punishments in which ‘terror, pain, or disgrace [were] superadded ’ to the penalty of death.”¹²² In finding no superadded pain, the Court highlighted that “the search for less painful modes of execution led many States to switch to lethal injection.”¹²³ Bucklew thus intimates, albeit inchoately, that a vague intention to improve an execution method is good enough and obviates the need for closer factual inquiry into whether the method under review actually causes pain.

Second, Bucklew required the prisoner to proffer an alternative method of execution with significant detail.¹²⁴ Justice Gorsuch faulted Bucklew’s bare-bones proposal to execute him using nitrogen hypoxia, noting that he failed to present detailed evidence about how such an execution would work.¹²⁵ The Court, in other words, refused to look carefully at the procedure before it but instead required the plaintiff to provide significant detail about the proffered alternative.

Even if prisoners somehow surmount these sizable hurdles, the Court’s precedent invites lower courts to hold that a state’s procedure is not “sure or very likely” to inflict terrible suffering whenever there is conflicting expert testimony.¹²⁶ States are usually able to find an expert to defend their protocols, so this is a low bar.¹²⁷ In some instances, the resulting deference may be defensible, such as in cases about whether pulmonary edema causes suffering, a question for which the evidence appears genuinely equivocal.¹²⁸ Significantly, though, prisoners lose even when the science is overwhelmingly on their side, as in Glossip. Indeed, that case demonstrates that any testimony in favor of the state’s protocol, no matter how inconsistent with widely accepted medical understandings, will apparently suffice for a court to disregard a protocol’s dangers.

The Court even sometimes wipes aside lower courts’ factual concerns in emergency docket cases without even bothering to engage with those facts.¹²⁹ For example, in Landrigan v. Brewer, a federal district court in Arizona granted a temporary stay of execution because the State procured its thiopental from a non-FDA approved foreign source and refused to disclose any evidence regarding the drug.¹³⁰ Despite the trial court’s conclusion “that use of sodium thiopental from a non-FDA-approved source raises issues regarding its efficacy and possible side-effects,”¹³¹ the Supreme Court vacated the order the very next day in a one paragraph opinion that rejected the trial court’s factual concerns as “speculation.”¹³² The trial court had merely stayed the execution to permit it “to fully consider” the challenge,¹³³ but the Supreme Court didn’t want it to delay an execution to dig into those facts. The Court similarly vacated a district court’s preliminary injunction in Barr v. Lee, again without full briefing or oral argument—that is, without any serious factual inquiry.¹³⁴

For the most part, the Court’s doctrinal moves have had their intended effect. In the years since Baze, plaintiffs challenging execution protocols almost always lose.¹³⁵ They occasionally score a modest win, such as a preliminary injunction or a stay of execution,¹³⁶ and the occasional lower court does examine the facts.¹³⁷ However, prisoners usually lose these cases regardless of the facts they present. Indeed, due to state secrecy statutes, plaintiffs often lose without ever gaining access to important details about the states’ protocols.¹³⁸

The Supreme Court, of course, has its reasons for relegating lethal injection facts to the sidelines. Capital inmates bring many legal claims, and some of them are frivolous.¹³⁹ Even plaintiffs with legitimate claims are usually hoping to delay their executions. The Court knows this and has designed its Eighth Amendment doctrine to defeat what it broadly considers bad faith litigation.¹⁴⁰

Given that the Constitution permits capital punishment,¹⁴¹ the Court’s concerns here are legitimate. From the Court’s perspective, it can’t let strategically timed, substantively frivolous lawsuits grind executions to a halt. The problem is that the Court has created the doctrine with only these interests in mind. It’s as though the prisoners’ Eighth Amendment rights don’t exist at all.

The Court instead could have tried to draw a line between frivolous and meritorious cases. Admittedly, there would be some difficult cases on the margins, and a more fact-dependent analysis would require courts’ closer attention. On the other hand, this more careful approach would give weight to both states’ and prisoners’ interests, and it would focus attention appropriately on the degrees of risk and pain.

That isn’t the Court’s line, though. Rather, the Court has treated all method-of-execution challenges as effectively frivolous, even though, as Professor Lain meticulously demonstrates, many state protocols create profound risks of excruciating pain. The Supreme Court doesn’t want to consider those facts.

III. Fact-Lite Constitutional Law

It is striking how relentlessly the Court avoids the grisly facts of lethal injection, but it is not anomalous. To the contrary, in a variety of constitutional cases, the Court pushes aside facts that are inconvenient to the majority’s preferred outcome.¹⁴² That has long been true in other Eighth Amendment contexts, such as litigation over prison conditions, where the Court also frequently defers to state prison systems with little attention to conditions that are sometimes truly appalling.¹⁴³ Significantly, though, this phenomenon extends well beyond the Eighth Amendment.

The Roberts Court, in particular, has made a specialty of fact-lite constitutional rulings. Of course, a comprehensive discussion and rigorous taxonomy of the Roberts Court’s use of facts in constitutional cases is far beyond this Review’s scope, but a short exploration helps demonstrate that the Court’s fact-lite approach extends well beyond lethal injection decisions. ¹⁴⁴ It would be an overstatement to call these rulings entirely fact-less—they do engage with facts to some extent—but these opinions are notable for their avoidance of facts that are crucial for an honest appraisal of the constitutional question presented. To be clear, in some of these cases, the Court might have reached the same outcome anyway; constitutional cases often involve complicated factors cutting in multiple directions.¹⁴⁵ That said, in some cases, it was easier for the Court to arrive at its outcome by weaving around factual complications that seemed to favor the losing party.

The Court’s fact-lite constitutional law relies on a variety of artful dodges. One of its strategies is to deem legally irrelevant facts that are difficult to quantify—even if those facts are central to the case.¹⁴⁶ For example, in effectively holding that affirmative action violated the Equal Protection Clause, Students for Fair Admissions v. Presidents and Fellows of Harvard College ¹⁴⁷ refused to weigh the educational benefits of diversity because they were “not sufficiently coherent for purposes of strict scrutiny.”¹⁴⁸ “How,” asked the Court, “is a court to know whether leaders have been adequately ‘train[ed]’; whether the exchange of ideas is ‘robust’; or whether ‘new knowledge’ is being developed?”¹⁴⁹ The majority is correct that diversity’s educational benefits are difficult to measure, but that does not mean that they are not real. Almost anyone who has taught law can attest to the pedagogical benefits of diverse classroom perspectives.¹⁵⁰

The Court reasoned similarly in Dobbs v. Jackson Women’s Health Organization,¹⁵¹ discounting women’s reliance interest in the right to abortion under Roe v. Wade ¹⁵² because that interest was too intangible.¹⁵³ Measuring that reliance, the Court wrote, “depends on an empirical question that is hard for anyone—and in particular, for a court—to assess, namely, the effect of the abortion right on society and in particular on the lives of women.”¹⁵⁴ Despite those empirical difficulties, in the half century after Roe, some women did rely on the right to abortion in ordering their lives.¹⁵⁵ In both Students for Fair Admissions and Dobbs, then, the Court discounted as constitutionally irrelevant real-world phenomena because they were difficult to measure empirically.

In other constitutional cases, the Court casts aside facts found by Congress without careful consideration of whether Congress’s fact-finding merited respect in the given circumstances.¹⁵⁶ Notably, those cases sometimes involve the mechanics of democracy, a topic Congress presumably understands much better than courts. In Citizens United v. Federal Election Commission, the Court ignored congressional fact-finding that restrictions on independent expenditures by corporations’ general treasury funds for election-related speech were necessary to guard against corruption and the appearance of corruption.¹⁵⁷ As the dissent pointed out, Congress had crafted the campaign-finance statute with “a virtual mountain of research on the corruption that previous legislation had failed to avert,”¹⁵⁸ but the Court didn’t look at it. Of course, even with closer factual examination, the Court still may have struck the law down given the importance of the First Amendment interests at stake.¹⁵⁹ Nevertheless, it was striking that the Court simply brushed aside the facts that Congress believed justified the law.

Likewise, in Shelby County v. Holder the Court disregarded fifteen thousand pages of congressional findings when it struck down the coverage formula for the Voting Rights Act’s preclearance provision.¹⁶⁰ Congress had concluded that racial discrimination, while more subtle than in the past, still “diminish[ed] . . . the minority community’s ability to fully participate in the electoral process . . . .”¹⁶¹ The Court, however, deemed those findings legally irrelevant because voting discrimination was no longer as “pervasive” and “flagrant” as it once had been.¹⁶² The Court, in other words, crafted its opinion so that evidence of continuing discrimination didn’t matter.¹⁶³

In another more recent case with implications for democracy, the Court in Trump v. United States disregarded factual allegations that President Trump had tried to overturn the results of the 2020 election.¹⁶⁴ Rather than contemplating the issue of presidential immunity in the context of those alleged facts, the Court instead fixated on the possibility that presidents will prosecute their predecessors.¹⁶⁵ Of course, the Court ought not have treated allegations as fact, but it could have considered the legal question in light of the actual allegations before it. The Court didn’t, though; instead, it resolved a monumental case about presidential power without considering the alleged facts or their potential implications for democracy.

This context-lite jurisprudence is bad for both justice and democracy. As Professor Jamal Greene has argued in the rights context, rather than crafting a nuanced, contextual jurisprudence, the Supreme Court prefers bright-line rules.¹⁶⁶ In American constitutional law, some asserted rights usually prevail, and others usually lose.¹⁶⁷ Under this approach, courts can more easily shove facts aside. The Court’s approach discourages the exercise of “judgment.”¹⁶⁸ Whereas many high courts in other countries engage in context-specific proportionality review, American courts usually don’t.¹⁶⁹ Facts matter far less when courts aren’t interested in balancing the interests on both sides of the ledger.¹⁷⁰

But facts should matter. If executions by lethal injection were always painless, the outcomes of Baze, Glossip, and Bucklew would be justifiable. In light of the facts, though, these decisions appear questionable or even (in Glossip’s case) crazy.

From the Court’s perspective, it’s much easier to write opinions if you don’t have to worry about pesky facts that complicate the bottom line. It’s harder to uphold an execution protocol if you concede its dangers. What the Court gains in rhetorical convenience, though, it loses in other respects. The Court’s fact-lite approach categorically privileges one set of interests over another. Even strong evidence on the “loser’s” side—for instance, that a state’s method of execution inflicts excruciating pain—likely won’t affect the outcome, despite the Eighth Amendment’s prohibition against “cruel and unusual punishments.”¹⁷¹ The resulting messages are that the Court is indifferent to state-imposed cruelty and that the losing party is a constitutional outsider whose interests are invisible to the law.¹⁷²

This fact-lite approach also undermines an important objective of Article III. One advantage of the case-or-controversy requirement is that it presents courts with legal questions not in a theoretical vacuum but with real-world facts.¹⁷³ By deciding some cases without the benefit of a comprehensive view of the facts, the Court denies itself the advantage of context that Article III supposedly confers.¹⁷⁴

The Court’s penchant for ignoring facts is especially concerning today. One disheartening takeaway from the Trump immunity case was that the Supreme Court sidestepped the case’s context—that is, it ignored allegations that the President had tried to steal an election and undermine democracy.¹⁷⁵ One consequence of that contextless approach is that the Court seemed oddly impervious to the possibility that its broad grant of immunity would embolden future presidents to abuse their power.¹⁷⁶ As President Trump now asserts his increasingly aggressive vision of executive power,¹⁷⁷ the Court is hearing many more cases with profound implications for American democracy.¹⁷⁸ The Court ought not decide those legal questions in a factual vacuum.¹⁷⁹

In the first year of the second Trump presidency, though, the Court has done exactly this all too often. Time and again, it has issued important rulings on the emergency docket with curtailed briefing, no oral argument, and little to no explanation.¹⁸⁰ Such fast-track shadow docket rulings do not give the Court time to engage carefully with the facts (or, for that matter, the law).¹⁸¹

The issues facing the Court and country are too important to decide in a vacuum without factual context. For example, in an era where the government has whisked away lawful residents—some apparently in retaliation for nothing more than political speech the administration finds objectionable¹⁸²—courts must distinguish between policies that genuinely protect national security and those that are a pretext for other agendas. Courts cannot do so effectively without closely and fairly examining the facts. Likewise, as courts consider issues like mass firings of civil servants, the dismissal of government watchdogs, the weaponization of the military and ICE against U.S. citizens, and the use of presidential power to try to cow universities, law firms, and businesses into submission, it is imperative that they examine whether these policies serve a legitimate public function or rather constitute efforts by a would-be autocrat to expand power, normalize abuses of law, and desensitize the public to the erosion of their rights.¹⁸³

When considering challenges to the Trump Administration’s policies, then, courts should consider the implications for the Constitution and democracy within the broader factual context—including the fact that these cases do not arise in isolation but rather as part of a barrage of executive actions that seek to bend both public and private institutions to the President’s will.¹⁸⁴ In most of the emergency docket cases involving the second Trump Administration, the Court has ignored this context and preserved administration policies that the lower courts usually found were likely illegal or unconstitutional.¹⁸⁵ Indeed, the Court has granted the administration numerous emergency stays without considering the traditional equitable factors that ordinarily invite judicial consideration of important facts, such as the harms that might result from granting or denying a stay.¹⁸⁶ Thus, as in the lethal injection sphere, the Court has extended so much deference that it has declined to examine the facts.

Conclusion

Standing next to the fate of American democracy and the rule of law, lethal injection cases may seem like small potatoes. Method-of-execution challenges implicate an esoteric subbranch of constitutional law and, in recent years, have affected only about twenty people annually.¹⁸⁷ Of course, the way we treat society’s most loathed members shines an uncomfortable light on our country’s values; the death penalty celebrates violence and legitimates vengeance.¹⁸⁸ Even so, the republic’s fate does not hinge on Eighth Amendment litigation.

Lethal injection cases, though, might also be the canary in the coal mine. Professor Lain has performed a great service, shining a harsh light on all the facts the killing state doesn’t want us to see. We live, however, in a postfactual age.¹⁸⁹ Sadly, even courts sometimes cherry pick “rogue” facts.¹⁹⁰ If the Supreme Court refuses to look at inconvenient facts in lethal injection cases, we should ask what else it ignores. Indeed, if the Court steers around inconvenient facts where the societal stakes are comparatively low, we might wonder whether it will honestly consider facts and context when the stakes for democracy and the rule of law are much higher.

* Earl Dunlap Distinguished Professor of Law, University of Nebraska College of Law. Thanks to Megan McCracken, Jen Moreno, and the participants in the University of Nebraska College of Law workshop series for helpful suggestions. Thanks also to John Greer for very helpful research assistance and to Sabriya Imami, Lila Nazarian, Amy Dayton, Emi Owens, Jude Goossens, Raisha Waller, Lazar Gueorguiev, Caleb Rummel, and the other editors of the Michigan Law Review for excellent editorial assistance. In the interest of full disclosure, in the mid-2000s, I was a member of teams challenging lethal injection procedures on behalf of Clarence Hill in the U.S. Supreme Court and Michael Taylor in the federal trial and appellate courts. The views expressed here are my own. A McCollum Grant supported the writing of this Review. All errors are mine.