Ledell Lee appears in Pulaski County Circuit Court on Thursday for a hearing in which lawyers argued to stop his execution. He was executed Thursday night. (Benjamin Krain/The Arkansas Democrat-Gazette via Associated Press)

Thursday night, the state of Arkansas executed Ledell Lee for the 1993 murder of 26-year-old Debra Reese. It was the first of what the state had initially hoped to be eight executions carried out over the course of 11 days. The reason for the binge is that the state’s supply of one of the drugs it uses in such executions is about to expire.

When death penalty supporters defend state executions, they tend to point to the most heinous of crimes, the most remorseless of killers and the most slam-dunk and irrefutable of convictions. Just today, conservative commentator Jonah Goldberg defended the death penalty in light of the Arkansas executions by referencing “Facebook killer” Steve Stephens.

The suspect was Steve Stephens, the so-called Facebook Killer, who videotaped himself admitting that he was about to murder someone randomly. He then got out of his car, walked up to 74-year-old Robert Godwin, a father of ten and grandfather of 14, and casually executed him. Stephens then posted the video on Facebook.

Stephens killed himself two days later. But say he hadn’t. Obviously, he would have gotten a trial. Let’s suppose he was found guilty and got the death penalty. We would still be subjected to all of the sleight-of-hand rhetoric about the risk of executing innocent people, the costs, etc., even though there would be zero doubt in this instance.

In practice, we know that this isn’t how the death penalty operates. In practice, the factor that overwhelmingly drives the death penalty isn’t the severity of the crime or the strength of the evidence. It’s politics — the politics of the county in which the murder trial takes place, the politics of the prosecutor and the politics of the state. According to a 2013 report by the Death Penalty Information Center, just 2 percent of counties in the United States account for a majority of the death row population. (Those counties also seem to have a disproportionate share of prosecutorial misconduct — and, interestingly, killings by police officers.)

But that is just death sentences. When it comes to actually carrying out executions, the politics of the surrounding state add an additional layer of geographic distortion. For example, according to that 2013 study, five of the 10 counties that produce the most death sentences in the United States are in California. But California hasn’t executed anyone since 2006. If you reside in Riverside or Los Angeles County and are accused of murder, you’re more likely to get a death sentence than anywhere else in the United States. But you’re less likely to actually be executed than those condemned to die in just about any other county in any other state in which the death penalty is still operational.

Arkansas’s unprecedented plan to execute eight men in 11 days this month is as good an illustration of the randomness of the death penalty as any. The men the state wants to kill this month were convicted between the years of 1990 and 2000. In that time, there were about 230,000 murders in the United States. Of those, around 2,800 resulted in death sentences — or a little over 1 percent. They’re unlucky to have been convicted in the 1990s, when death sentences peaked at about 300 per year. They’ve been steadily falling since about 2000. Last year, there were just 30 new death sentences in the entire country. (If you were unlucky enough to get a death sentence in the 1990s but are still alive on death row, your odds of avoiding execution look to be improving, at least outside of Arkansas and a few other states — the number of executions has also been dropping steadily, from a high of 98 in 1999 to just 20 last year.) The 1990s were also the decade that saw the enactment of the the Antiterrorism and Effective Death Penalty Act (AEDPA), which severely limited the ability of federal courts to oversee the states’ administration of the death penalty.

The intersection of time and place provides an additional layer of randomness. These men weren’t just unlucky to have been convicted in the 1990s and in Arkansas, but in the 1990s in Arkansas. Of the state’s 33 death row inmates, 16 were sentenced in the 1990s. The president who signed AEDPA was Bill Clinton, a former governor of Arkansas. It was Clinton who jump-started the death penalty in the state with an execution in 1990, the first in more than two decades. Clinton would then ride his blue-dog-Democrat, law-and-order reputation to the White House. During the 1992 presidential campaign, he made a show of taking a break from the trail to go back to Arkansas to oversee the execution of Ricky Ray Rector. Over the next five years, Arkansas would carry out the modern era’s first “double execution” — two killings in one night. The state followed with triple executions in 1994 and 1997.

But perhaps most random of all is the reason for the state’s plan to kill eight men in April. Arkansas wanted to kill as many people as possible before one of the executions drugs expires at the end of the month. As Justice Stephen Breyer wrote in a dissent last night, that’s about as arbitrary as it gets.

Arkansas set out to execute eight people over the course of 11 days. Why these eight? Why now? The apparent reason has nothing to do with the heinousness of their crimes or with the presence (or absence) of mitigating behavior. It has nothing to do with their mental state. It has nothing to do with the need for speedy punishment. Four have been on death row for over 20 years. All have been housed in solitary confinement for at least 10 years. Apparently the reason the State decided to proceed with these eight executions is that the ‘use by’ date of the State’s execution drug is about to expire. In my view, that factor, when considered as a determining factor separating those who live from those who die, is close to random.

In fact, of the three drugs the state uses to execute the condemned, one is set to expire and one was “donated” by a mysterious supplier. The third was supplied by a company that says the state acquired it under false pretenses, and has stated in court fillings that had it known the drug was to be used it executions, it would never have sold it to state officials. Not to mention that we know little to nothing about how these drugs actually work, as evidenced by the series of horrifying botched executions over the past few years. Proponents of lethal injection say it’s the most humane method of killing available, but given that we don’t know what actually happens to the condemned after the paralytic agent sets in, they seem to be more concerned about the appearance of humaneness — that these executions look as little like what they are as possible — than minimizing the pain and suffering of those who are executed.

But then there are the cases themselves. Far from the open-and-shut, worst-of-the-worst stories death penalty advocates often point to, you’d be hard-pressed to find a better example of the biases, deficiencies and utter randomness with which the death penalty is applied.

The state’s plan to execute eight men in 11 days took a hit early in the month when the Arkansas Parole Board recommended clemency for Jason McGehee, who was convicted of abducting and killing a teenager in 1996. McGehee’s own case shows how unevenly death is meted out as a sentence. He and two friends carried out the crime. One of his accomplices was sentenced to life without parole. The other, a juvenile at the time, received an adjusted sentence of 40 years. Only McGehee received a death sentence. The judge in McGehee’s case was among those who recommended clemency. He told the New York Times earlier this year, “I tried a lot of capital murder cases in my years, and I saw people that I thought were much worse individuals get life without parole as opposed to the death penalty. I didn’t see him as the worst of the worst. As a matter of fact, he was a very young man.”

But the same parole board also denied clemency to Marcel Williams. As Liliana Segura reported at the Intercept*, while — as with McGehee — there’s little doubt about Williams’s guilt, the details of his background and upbringing are just devastating. At his clemency hearing, Williams’ attorney showed a video about Williams’ life.

In interview clips, two of Williams’s cousins described Williams’s childhood home in a violent part of North Arkansas. “The first thing you would notice would be the odor,” Shannon Carthron said. There was a stench of sewage and spoiled food. “There were roaches. Most of the time they didn’t have electricity.” Williams never knew his father; his mother was often absent. When she was at home, she was brutally abusive. One time, Williams’s mother made him strip naked and beat him with an extension cord. His cousins saw him burst through the door screaming; he ran into the backyard, naked and bleeding. “I was scared,” Carthron said. “I thought that he was gonna die.” In another incident described in Williams’s clemency petition, his mother burned him with a pot of boiling water. “The scars are still visible on Mr. Williams’s body,” according to Dr. David Lisak, a clinical psychologist and expert on childhood trauma. Lisak reviewed Williams’s history and interviewed Williams’s mother. In his subsequent statement, he said she began pimping her son out to older women in exchange for food stamps and other necessities, which he described as a form of incest. By the time Williams was 12, she did so “routinely.” He was also sexually abused by men who his mother brought home.

By the time the video ended, Williams was weeping.

But it wasn’t just family who failed Williams. The state did, too.

If the odds were stacked against Williams, at least one former board member had known him as something other than a sexual predator. Longtime board chairman Leroy Brownlee, who retired last year, had known Williams when he was a counselor at the Pine Bluff Training School, a juvenile reform facility. In the 1980s, he met Williams, who was 12 or 13 at the time. As Brownlee recalled in a 2006 statement, it became clear he was not used to receiving attention; he thrived in the structured environment and was even reluctant to leave. “The downfall for Marcel was the environment that he was released to go back to,” Brownlee wrote. According to his clemency petition, Williams “immediately committed a robbery,” intending to be sent back to the training school. But at 17, he was sent to an adult prison, where he was gang raped by older men. “As Marcel’s prison record show, he began committing infractions so he could be placed in solitary confinement, as a way of protecting himself.”

In death penalty cases, defense attorneys are supposed to present this sort of mitigating information to juries at the sentencing portion of the trial. Williams’s attorney, who was just two years out of law school and handling his first murder case, failed to do so. He told the board, “I mean, we completely and absolutely dropped the ball. I think there was huge amount of mitigation that could have been brought forward.”

There were also enormous problems with how the state prosecuted Ledell Lee, the man Arkansas executed last night. During one of Lee’s two trials, the judge was having an affair with the prosecutor. His post-conviction attorney was an admitted alcoholic who showed up to at least one hearing intoxicated. Lee, too, was accused of at least one other murder and two other rapes, so perhaps he isn’t the most sympathetic of those on death row, but in the case for which he was sentenced to die, the physical evidence against him was weak. He was implicated, for example, by a method of hair fiber analysis that has since been discredited. He was also implicated by testimony from a serologist about a blood spot on his shoe. Unfortunately, the analyst used up the sample, preventing Lee’s own attorneys from having it tested. There was a wealth of other physical evidence at the scene, none of which matched Lee, and which his attorneys tried to get tested for DNA. Such a test could have exonerated Lee, or it could well have gone a long way toward confirming his guilt. The state and the courts refused.

Untested DNA is also an issue in the case of Stacey Johnson, another of the eight men Arkansas wants to kill this month. Johnson’s innocence claim is even stronger. He was convicted primarily due to the eyewitness testimony of a girl who was 6 at the time of the crime. Again from Segura:

Though she was found incompetent to testify at Johnson’s first trial, in 1994, Ashley was deemed ready to take the stand for the retrial in 1997. The 10-year-old delivered testimony that seemed heavily influenced by relatives and prosecutors — a fact that alarmed members of the Arkansas Supreme Court who reviewed the case years later. In a 4-3 ruling leaving the conviction intact, the dissenting judges noted that Johnson’s defense attorneys had been denied access to therapist records that showed “Ashley’s stories were profoundly inconsistent and that she had been under considerable pressure from her family and the prosecutor to convict Stacey Johnson.” Among passages they quoted: “The DA says she’s the only one who can ‘keep him behind bars’”; “Her grandmother told Ashley that she ‘has to keep him behind bars,’ because if he gets out he’ll try to kill Ashley next.”

The only physical evidence against Johnson was a hair found at the crime scene. DNA testing at the time could determine only that Johnson couldn’t be excluded as the source of the hair. Today’s more sophisticated technology could come up with a more definitive conclusion, but here again, the state refuses to order testing, and the courts have refused Johnson’s request to compel it.

There are more problems with just these eight cases. The Fair Punishment Project recently reviewed the case files. Among the group’s findings:

  • Jason McGehee has bipolar disorder and comes from a family with a history of mental illness. His own mother thought he was “possessed by the devil.” His father was physically abusive with both Jason and his mother, and when Jason was young, slit the throats of Jason’s dogs as the boy watched. At one point, Jason’s mother forced him to live outside in a dog coop. When he was 7, his stepfather kicked Jason’s dog to death as he looked on in horror.
  • Bruce Ward has been diagnosed with paranoid schizophrenia. A psychiatrist once testified that he believed his death sentence was “harassment by evil or demonic forces which God has temporarily allowed [in order] to prepare him for a special mission as an evangelist.” Ward, too, was abused. His mother once covered him in tar and forced him to sit in ice water. When the roof to their home began to leak, she forced him to lay under the leak for an extended period of time as the water dripped onto him.
  • Kenneth Williams has an IQ of 70. He stayed at six different foster homes as a child. His father once abducted his mother at gunpoint for several days. He also cut open her stomach with a broken bottle.
  • Jack Jones suffers from bipolar disorder, depression and hallucinations. He has tried to kill himself several times.
  • In most of these cases, this sort of information came out only in post-conviction. In most cases, the defense attorneys presented appallingly little mitigation evidence at trial.

Importantly, these eight cases were not cherry-picked by death penalty opponents to illustrate the deficiencies in how we apply capital punishment in the United States. They were selected by the state of Arkansas for the relatively arbitrary reason that these men happened to be close to execution at a time when one of the state’s execution drugs happened to be near its expiration date, and the state has enough of the drug on hand for eight executions.

In other words, this pool of eight death penalty cases was selected nearly at random. Yet included among them are disturbing deficiencies such as severe mental illness and disability that were never presented at trial, ineffective defense counsel, prosecutorial and judicial malfeasance, and untested but possibly exculpatory DNA. Most of these problems pervaded several of the eight cases. If this pool of eight cases thrust into the spotlight only by the state of Arkansas’s rush to carry out executions could be so rife with problems, one could imagine you could pick any eight cases from the population of death row inmates and find a similar proportion of deficiencies.

Factor in the arbitrariness of when, where and by whom a defendant was convicted — and that the likelihood of a conviction and death sentences seems to be affected by the race of not only the accused but also the victim — and you begin to see a system that metes out death not on merits of the evidence or the severity of the crime, and certainly not in a way that demonstrates any measured and equal distribution of justice.

Instead, we see a system that appears to impose death on a whim.

(*Disclosure: Segura and I are married.)