V
The Death Penalty
During an oral argument in a capital case in the spring of 1982, a man in the audience stood up and started to take off his clothes. His bizarre behavior was intended to dramatize his opposition to the death penalty, but the Supreme Court police removed him so silently and promptly that the argument proceeded without interruption. A few days later my wife, Maryan, and I were guests at a state dinner at the White House. She sat next to Charles Zwick, director of the United States Information Agency (USIA), and I was seated at President Reagan’s table next to Queen Beatrix of the Netherlands. My assumption that my dinner companions would be interested in a description of the unusual protest at the Court could not have been more misguided; as a raconteur, I really laid an egg. Maryan, on the other hand, got along famously with her dinner partner. As a result, we later received an invitation from the USIA to go to Helsinki to represent the United States at a celebration of the fortieth anniversary of an organization formed to maintain friendly relations between Finland and the United States. Shortly after we arrived in Helsinki, our ambassador arranged a press conference with about a dozen representatives of the Finnish news media. The entire conference consisted of variations on one question: “Why does the United States still have the death penalty?” In my repeated attempts to answer that question I must have laid another egg, because it was obvious that none of the reporters were favorably impressed by anything I said to justify that institution.
I had told them that our decisions upholding the constitutionality of state capital punishment statutes did not mean that the justices agreed with the wisdom of the legislation. Indeed, I think Warren Burger and Harry Blackmun—who were both former residents of Minnesota, which did not permit capital punishment—had both stated publicly that as a matter of policy they opposed the death penalty. Rather, the Court had merely held that under our Constitution it was permissible for state legislators to conclude that the possibility of being sentenced to death might deter some potential murderers from committing that crime, and that community outrage sometimes demanded retribution for especially vicious crimes. Intervening events have convinced me that even if the death penalty was justified in 1982, this is no longer the case.
It was after that visit that I learned why the state of Michigan does not have the death penalty. As David Garland explains in his book Peculiar Institution, in 1846 the state legislature adopted a statute abolishing capital punishment for any crime other than treason. The sponsors of the legislation were concerned about the “fallibility” of capital punishment. As Professor Garland explains, an innocent man had recently been executed in Canada, and the state senator who introduced the bill had apparently presided over the execution of an innocent person while serving as a sheriff in New York. Some years later, Michigan lawmakers concluded that the state constitution should be amended to avoid the risk of such an injustice. Over the years since my visit to Finland I have concluded that the federal Constitution should also either be construed or amended to avoid that risk.
The enactment by most state legislatures of statutes authorizing sentences of life imprisonment without the possibility of parole has eliminated one justification for the death penalty and so reduced the significance of another that it barely passes the rational basis test. It can no longer be argued that execution of a potentially dangerous offender is necessary in order to remove the risk that he will commit further crimes. And the notion that the possible imposition of a death sentence is a significant deterrent on potential murderers must be modified to evaluate the marginal difference between the deterrent effect of that possible sentence and the deterrent effect of a sentence of life imprisonment without the possibility of parole. It is unlikely that criminals contemplating vicious crimes engage in the kind of cost-benefit analysis that would draw a distinction between those sentences. The real justification for preserving capital punishment surely rests on the interest in retribution. “An eye for an eye and a tooth for a tooth” provides an explanation for preserving capital punishment that is both more realistic and more acceptable than any other.
That justification is consistent with the repeated decisions of state legislatures to limit the availability of the death sentence to cases in which statutorily defined aggravating circumstances must be proved in order to establish the defendant’s eligibility for that penalty. It is also consistent with Justice Potter Stewart’s view that “death sentences are cruel and unusual in the same way that being struck by lightning is cruel and unusual.… the Eighth and Fourteenth Amendments cannot tolerate the infliction of a sentence of death under legal systems that permit this unique penalty to be so wantonly and so freakishly imposed.”1 And retribution is surely the only justification for the Supreme Court’s shameful approval of the prosecutor’s use of so-called victim-impact testimony at the penalty phase of capital cases even when the evidence sheds no light on either the guilt or the moral culpability of the defendant. In sum, I am convinced that the question whether we should retain the death penalty depends on the strength of the interest in retribution—the interest in avenging the harms caused by the most vicious criminals. I shall therefore comment on two cases that shed light on the diminishing importance of that interest, and then conclude with a comment on the finality of the death sentence in a criminal justice system that is not infallible, and an amendment to put an end to what has become a wretched arrangement.
On August 12, 1984, Joe Elton Nixon, then twenty-three years old, committed a brutal murder that would have made him eligible for the death penalty in any jurisdiction that authorizes capital punishment. He had approached Jeanne Bickner in a shopping mall, asked her to help him jump-start his car, and then accepted her offer to drive him home. He directed her to drive to a remote location where he overpowered her, tied her to a tree with jumper cables, torched her personal belongings, and used them to set her on fire. Her charred body made it obvious that she had suffered a gruesome, excruciatingly painful death. The police promptly apprehended Nixon and collected overwhelming evidence establishing his guilt. Before his trial, his appointed counsel offered to have him plead guilty in exchange for the prosecutor’s recommendation of any sentence other than death. When the prosecutor declined that offer, the only issue that separated the parties was the question whether Nixon would be allowed to die a natural death in prison or be executed by the state. Answering that question in Nixon’s case—as in most capital cases—was time consuming and extremely costly for the state of Florida.
In his opening statement Nixon’s lawyer, Michael Corin, acknowledged Nixon’s guilt but advised the jury that at the penalty phase of the trial he would introduce evidence showing that his client’s conduct was the product of mental illness. Nixon himself was excluded from attendance at most of the trial because of his disruptive and violent misbehavior. No evidence was offered on his behalf during the guilt phase. At the penalty phase, however, Corin presented the testimony of eight witnesses, including friends and relatives who described Nixon’s emotional troubles and erratic behavior, and both a psychologist and a psychiatrist who addressed his history of emotional instability, his low IQ, and the possibility that he had suffered brain damage. After three hours of deliberation the jury recommended the death sentence, which the trial court imposed.
On direct appeal to the Florida Supreme Court, Nixon’s new lawyer argued that Corin had rendered ineffective assistance by conceding Nixon’s guilt without his express consent. Given the similarity between that strategy and a guilty plea, and the fact that the law is well settled that counsel may not enter a guilty plea without first obtaining his client’s express consent, there was obviously a substantial basis for the argument. (A similar argument had been endorsed by state supreme courts in Illinois and South Carolina, and by the federal Court of Appeals for the Sixth Circuit.) Instead of either approving or disapproving Corin’s trial strategy in the direct appeal from Nixon’s conviction, the Florida Supreme Court invited him to raise the issue in a state post-conviction proceeding at which a state trial judge could receive whatever further evidence his new lawyer might present. After multiple evidentiary hearings and appeals, in 2003—nineteen years after the crime—the Florida Supreme Court upheld Nixon’s claim because the evidence did not establish that he had made the kind of explicit consent that a valid guilty plea would require.
The elected state attorney general, Charles Crist—who was later elected governor—then filed a petition for certiorari, presenting the United States Supreme Court with the question whether Nixon had the burden of proving that he was prejudiced by his lawyer’s trial strategy. The Court granted the petition and reversed the state supreme court, thus finally ending the litigation. Justice Ginsburg, with her characteristic eloquence, wrote what turned out to be a unanimous opinion. She noted that the renowned advocate Clarence Darrow had “famously employed a similar strategy as counsel for the youthful, cold-blooded killers Richard Loeb and Nathan Leopold.”2
The fact that Justice Ginsburg was so persuasive in explaining the reasonableness of Corin’s trial strategy surely does not raise any question about the competence of the state judges who had come to a different conclusion. Indeed, although I did not agree with my law clerk’s analysis in the case, I find this excerpt from his cert memo interesting. He wrote: “I think the Florida court got it exactly right: While it may on occasion be a legitimate strategy for a lawyer to concede his client’s guilt during the guilt phase of a capital trial, the decision to choose such a strategy belongs to the client. When a lawyer pursues such a strategy without a client’s consent, that lawyer’s actions constitute an impermissible forfeiture of the client’s constitutional right to an adversarial proceeding to determine guilt.”
The expression of that opinion was fully consistent with his qualifications as a law clerk. Surely the earlier expression of the same opinion by judges on the Florida Supreme Court should not provide a basis for removing them from the bench. Nevertheless, in the judicial retention election in 2012—twenty-eight years after the crime—the political party in control of Florida’s government sponsored a campaign to use three justices’ votes in the Nixon case as a basis for removing them from the court. The campaign was not successful—all three justices won retention by substantial majorities—but it illustrates the fact that the political consequences of death penalty litigation may be more important than vindicating the state’s interest in retribution.
The post-conviction litigation in the Nixon case was protracted and expensive. Whether those costs were justified depends, I suppose, on the importance of vindicating the state’s interest in retribution by putting Nixon to death. Nevertheless, despite the elapse of over eight years from the date of the United States Supreme Court’s decision—and over twenty-eight years from the date when the Florida prosecutor rejected the offer of a guilty plea that would have required Nixon to spend the rest of his life in prison—he remains alive on death row today. He has not been on death row as long as Gary Eldon Alvord—who was sentenced to death in 1974—but the long delay in cases like Nixon’s and Alvord’s surely demonstrates that it is not necessary to put the defendant to death in order to vindicate the state’s interest in obtaining retribution for a heinous crime. (Alvord, for his part, suffers from disordered thinking and delusions, which under the Court’s decision forbidding capital punishment of the insane in Ford v. Wainwright3 likely make him ineligible for execution. Because the state cannot execute him but also refuses to commute his death sentence, Alvord appears destined to live out the remainder of his natural life under the harsh conditions of death row.)
In 2008, in Baze v. Rees,4 the Supreme Court upheld the constitutionality of Kentucky’s method of executing its prisoners who have been sentenced to death. At the time, the state used a three-step lethal injection procedure similar to that used by the federal government and most other states. The first drug, sodium thiopental, is a barbiturate that makes the prisoner unconscious; the second, pancuronium bromide, causes paralysis; and the third, potassium chloride, causes a fatal heart attack. The purpose of the barbiturate is to prevent the prisoner from feeling the pain caused by the second injection. The second does not serve any therapeutic purpose; it is used to preserve the dignity of the procedure by preventing involuntary muscle movements that observers might incorrectly perceive to be convulsions or seizures caused by severe pain. Because it masks any outward sign of distress, pancuronium bromide creates a risk that the inmate will suffer excruciating pain before death occurs.
There is a general understanding among veterinarians that the risk of pain is sufficiently serious that the use of pancuronium bromide should be prohibited when an animal’s life is being terminated. As a result of that understanding among knowledgeable professionals, Kentucky, like several other states, has enacted legislation prohibiting the use of the drug in animal euthanasia. The inmate petitioners in the Baze case argued that Kentucky should not be permitted to kill them by using a drug that it would not permit to be used on their pets.
The risk that prison personnel may make a mistake that will cause excruciating and undetectable pain is enhanced by the fact that the codes of ethics of the American Medical Association, the American Nurses Association, and the National Association of Emergency Medical Technicians prohibit their members from taking part in executions. Nevertheless, after hearing extensive testimony by experts and by Kentucky corrections officials during a seven-day bench trial, the Kentucky trial judge rejected the inmates’ challenge to the constitutionality of the procedure because he was not convinced that it created an unreasonable risk of causing excruciating pain. The Kentucky Supreme Court affirmed, as did the United States Supreme Court. Nevertheless, more recently, Kentucky has announced that it will join Texas, Arizona, Ohio, and seven other states that now use a single-drug regime for executions and employ a single dose of a barbiturate—either sodium thiopental or pentobarbital—in future executions.
Two features of executions in Elizabethan England and lynchings in Texas as late as 1893 served the public interest in retribution: the events were public spectacles that attracted large crowds, and they were designed to inflict excruciating pain before death occurred. The Baze case is significant because it dramatically illustrates how our society has moved away from public and painful retribution toward ever more humane forms of punishment. Today, instead of a public spectacle, an execution is almost as private as the administration of anesthesia before an operation in a hospital. The use of pancuronium bromide to paralyze the inmate protects those who do witness the event from possible discomfort from witnessing what might otherwise be perceived as a painful event. More important, the law now prohibits the state from deliberately inflicting any pain at all on the offender.
The dispute in Baze was over the factual question whether the state had taken adequate precautions to avoid the risk that improper administration of the three-drug protocol would inadvertently cause pain, not whether it could deliberately do so. The dispute generated seven different opinions by Supreme Court justices, but none suggested that the interest in retribution would allow the state to cause any pain deliberately. Chief Justice Roberts’s opinion apparently assumed that a method of execution would violate the Eighth Amendment if it posed a substantial risk of severe pain that could be reduced by adopting readily available alternatives. Justice Clarence Thomas criticized that view as too lenient because in his view the Constitution contains a bright-line rule that merely prohibits any deliberate infliction of pain. But even under his view the interest in retribution would not justify any attempt to apply an “eye for an eye” standard of punishment. Just as the Nixon case and its aftermath illustrate the waning public support for using the death penalty to avenge serious crimes, the Baze case reminds us that the Court has already developed a rule of law that prohibits states from subjecting the defendant to the kind of pain that he inflicted on his victim.
The requirement that guilt of a criminal charge be established by proof beyond a reasonable doubt has been a part of our law from our early years as a nation, but it was not until 1970 that the Court finally held that it was an aspect of “due process” protected by the Fifth and Fourteenth Amendments to the Constitution. In re Winship5 was a case in which a New York family court judge found that a twelve-year-old boy had entered a locker and stolen $112 from a woman’s pocketbook. The judge acknowledged that the evidence was not sufficient to establish the boy’s guilt beyond a reasonable doubt, but held that that standard of proof did not apply to delinquency proceedings under the New York Family Court Act, even though the finding justified the boy’s placement in a training school until he reached his eighteenth birthday.
When the case was reviewed in the United States Supreme Court, the judgment was reversed on the ground that the trial judge should have applied the “proof beyond a reasonable doubt” standard. In his concurring opinion Justice Harlan explained that the requirement of proof beyond a reasonable doubt in a criminal case is “bottomed on a fundamental value determination of our society that it is far worse to convict an innocent man than to let a guilty man go free.”6 That value determination increases in importance as the severity of the sentence increases.
Despite requiring proof of guilt that satisfies the beyond-a-reasonable-doubt standard, there is general agreement that juries do occasionally make mistakes and convict defendants who are actually innocent. This disturbing conclusion has been confirmed during the period after 1989, when DNA testing became widely available. The large number of exonerations includes a significant number of defendants who had been sentenced to death. Whether any innocent defendants have actually been put to death, and if so how many, are subjects bitterly debated by scholars.
The character of that debate is illustrated by the dissenting opinion of Justice Souter, and the response by Justice Scalia, in Kansas v. Marsh.7 In that case the Kansas Supreme Court had set aside a death sentence because the jury had been instructed that if they concluded in the penalty phase of the trial that the aggravating evidence favoring a death sentence and the mitigating evidence supporting a lesser sentence were in equipoise, the jurors should sentence the defendant to death. By a vote of five to four the U.S. Supreme Court held that the instruction was proper and reversed the judgment of the state supreme court.
In his dissent Justice Souter characterized the instruction that placed “a thumb on death’s side of the scale” as “morally absurd” under precedent dating back to 1972, and then included a section discussing “a new body of fact” that must be accounted for in deciding what the Eighth Amendment should tolerate because “the period starting in 1989 has seen repeated exonerations of convicts under death sentences, in numbers never imagined before the development of DNA tests.”8 Responding to that dissent, Justice Scalia argued at length that the studies cited by Justice Souter were unreliable and inaccurate. Justice Scalia cited another study in which, for the period between 1989 and 2003, the authors identified “340 ‘exonerations’ nationwide—not just for capital cases, mind you, nor even just for murder convictions, but for various felonies.” Given the fact that during that period there were more than 15 million felony convictions across the country, the error rate was .027 percent and the success rate was 99.973 percent. Moreover, in his view “none of the cases included in the .027 error rate for American verdicts involved a capital defendant erroneously executed.”9
He then summarized his view of the relevant evidence in these three sentences: “Like other human institutions, courts and juries are not perfect. One cannot have a system of criminal punishment without accepting the possibility that someone will be punished mistakenly. That is a truism, not a revelation. But with regard to the punishment of death in the current American system, that possibility has been reduced to an insignificant minimum.”10
For me, the question that cannot be avoided is whether the execution of only an “insignificant minimum” of innocent citizens is tolerable in a civilized society. Given the availability of life imprisonment without the possibility of parole as an alternative method of preventing the defendant from committing further crimes and deterring others from doing so, and the rules that prevent imposing an “eye for an eye” form of retributive punishment, I find the answer to that question pellucidly clear. When it comes to state-mandated killings of innocent civilians, there can be no “insignificant minimum.”
In 1846 evidence that two innocent men had been put to death by hanging convinced the Michigan legislature that the criminal justice system is not infallible and that a civilized society should not preserve a criminal sanction that risks the repetition of such an intolerable injustice. We may never know how many innocent prisoners have actually been put to death. We do know, however, that the risk of such injustice arises whenever a defendant is sentenced to death. Moreover, we also know that the risk is significant and that the finality of state action terminating the life of one of its citizens precludes any possible redress if a mistake does occur.
That risk can, and should, be eliminated by adding five words to the text of the Eighth Amendment, which already prohibits the states as well as the federal government from imposing cruel and unusual punishments. The inclusion of the words “such as the death penalty” in the text of that amendment would make it read:
Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments such as the death penalty inflicted.