On April 17Arkansas is scheduled to execute seven men over a period of 11 days. If carried out, that will be the most executions performed in such a short time since the modern death-penalty era began in 1976.

The reason: Arkansas’ supply of the controversial drug it is using for executions, midazolam, is set to expire April 30Midazolam is medically used as an anti-anxiety sedative, not an anesthetic. Experts have concerns about the drug’s ability to render a person fully unconscious, heightening the risk of an unconstitutionally cruel punishment. The lawyers defending the men scheduled for death are arguing that the short time will limit their ability to provide effective counsel and that the execution team will be so stressed that they will probably make mistakes.

We have tracked the changing methods and challenges of the death penalty in the United States in a forthcoming book.

A brief U.S. history of the rise and fall of various execution methods

Over the past 100 years, states have sought the most humane execution methods, each supposedly “guaranteed” to eliminate the gruesome errors of previous uncivilized methods. At the beginning of the 19th century, hanging was the universally accepted execution method.

Around the turn of the 20th century, the electric chair was introduced and quickly spread, thanks to the Gerry Commission (named after its chairman, Elbridge Gerry, grandson of the early Massachusetts governor who bequeathed the gerrymander to American politics). The Gerry Commission reviewed and rejected all known execution methods as barbaric and uncivilized — except the brand-new electric chair, then guaranteed to kill the inmate “in the ten-thousandth part of a second.” Thomas Edison vouched for this, the courts went along, and electrocution was soon the main method of state killing.

In 1924, Nevada adopted the gas chamber. Few other states joined in, partly because of the association with Nazi extermination camps, and partly because it was so difficult to seal the deadly gas within the chamber or to vent it safely after the prisoner was dead.

In 1982, Texas was the first to use lethal injection when it executed Charles Brooks Jr. Lethal injection thus became the most recent in a series of “institutional fads.” As you can see in the chart above, since the decline of hanging, no method of execution has remained popular for long.

How lethal injections became the method of choice

In 1977, Oklahoma developed the three-drug protocol that most states quickly adopted. Jay Chapman, the Oklahoma state medical examiner at the time, designed the procedure to improve on what he had seen occur during use of the electric chair. In writing the laws for that procedure, state Sen. Bill Dawson and Rep. Bill Wiseman had little or no consultations with doctors or scientists. The protocol was never subjected to any serious testing or evaluation. They didn’t consider any of the available evidence assessing the risks of lethal injection. The law left all critical decisions to the prison officials in charge of the execution, who often have little medical training or experience. And even the best procedures can go wrong if carried out by inexperienced, stressed and untrained personnel.

The three-drug execution protocol calls for executioners to inject an anesthetic, followed by a paralytic agent, and finally a drug that stops the heart. But which drugs, and in what quantities? Sodium thiopental was originally used in the first step to numb the patient. Attorneys argued that their defendants had been given the wrong dosage, but — rendered paralyzed by the second drug — couldn’t communicate with onlookers. In 2005, an independent study headed by physician Leonidas Koniaris studied 49 executions and concluded:

Protocol information from Texas and Virginia showed that executioners had no anesthesia training, drugs were administered remotely with no monitoring for anesthesia, data were not recorded and no peer-review was done. Toxicology reports from Arizona, Georgia, North Carolina and South Carolina showed that post-mortem concentrations of thiopental in the blood were lower than that required for surgery in 43 of 49 executed inmates (88 percent); 21 (43 percent) inmates had concentrations consistent with awareness.

In 2011, Hospira Pharmaceuticals, the only U.S. manufacturer of sodium thiopental, stopped making the drug because of its use in executions. Shortly after, the European Union banned exportation of any drug used for executions. States have tried various alternative anesthetics, usually pentobarbital and midazolam. But these attempts moved further into the territory of human experimentation. Midazolam, now commonly used in the first step, has resulted in a number of botched attempts, in which the prisoner visibly and audibly struggled. Those include:

  • In January 2014, Ohio administered a two-drug cocktail (including midazolam) to Dennis McGuire. He gasped for air for 25 minutes while the drugs took effect.
  • In April 2014, Oklahoma administered only 100 milligrams of midazolam, instead of the appropriate 500 mgs, to Clayton Lockett. As the second two drugs were being administered, the execution team noticed that something had gone wrong. Clayton writhed and clenched his teeth in pain, and the team canceled the execution before the second two drugs had been fully administered. Forty minutes later, Clayton died on the gurney from cardiac arrest.
  • In 2015, Oklahoma used the wrong combination of drugs on Charles Warner. After the midazolam was administered, Warner said, “My body is on fire.” The autopsy found that potassium acetate was used instead of potassium chloride, which was required by the protocol.
  • In 2016, Alabama injected midazolam to execute Ronald Bert Smith Jr., who for 13 minutes gasped and coughed for air, and didn’t die until 34 minutes after the injections.

These stories are not rare. In a 2014 book, Professor Austin Sarat documented 75 cases of botched execution by injection, and there have been a handful more since publication. He found that with each innovation in the death penalty, there is an increase in the rate of errors.

The U.S. Supreme Court has upheld lethal injection’s constitutionality under the logic expressed best by Justice Antonin Scalia’s opening remarks in Glossip v. Gross: “Because capital punishment is constitutional, then there must be a constitutional means of carrying it out.”

Each botched case of lethal injection has been seen by the court to be an isolated incident. Experts, including former prison wardens and executioners, have written a letter of concern about the “assembly line” plan, which they argue makes mistakes more likely, given the stress and trauma that executions impose on corrections staff.

Arkansas has not carried out an execution in 12 years. If one of the upcoming seven goes awry, we may come to the end of the lethal injection cycle.

Frank R. Baumgartner is the Richard J. Richardson distinguished professor of political science at the University of North Carolina at Chapel Hill, and co-author of the forthcoming book “Deadly Justice: A Statistical Portrait of the Death Penalty.”

Kaneesha Johnson is co-author of “Deadly Justice” and a graduate student in government at Harvard University.