The Executioner's Dissonant Song: On Capital Punishment and American Legal Values

This essay concerns the implications of the practice of capital punishment on American legal values and procedures. Any death penalty, in my view, conflicts with substantive principles of human dignity, but my argument here addresses a narrower theme: the ways in which attempts to reduce delays in death penalty cases are undermining general legal norms of fairness and due process. The problem I describe is not a temporary phenomenon associated with the transition to a more efficient regime. Instead, the clash betweeen the operational needs of an execution system and the principles and procedures of American legal culture is fundamental. Either the basic rules and values will change or the practice of execution will remain infrequent, conflict-laden, and problematic. Capital punishment can only be regarded as normal state behavior by reimagining basic principles of fairness in criminal justice.

In exploring the wide shadow that death penalty cases cast over American legal values, I am obviously rejecting a discrete conception of capital punishment held by some proponents of execution. In their view, capital punishment is a question of the appropriate sanction to be imposed on the most serious form of murder, a matter of principal importance to one part of the administration of criminal justice. In contrast, this essay reflects a belief that the conduct of death penalty cases has serious implications for the larger realm of criminal procedure. Two recent cases before the U.S. Supreme Court illustrate the pressure that facilitating executions places on legal sentiments and values. They suggest that capital punishment, though it is different in a qualitative way from other criminal sanctions, may subvert due process and fairness values broadly throughout the criminal justice system.

When the Supreme Court in McClesky v. Kemp (1987) found that a disproportionate rate of death sentences in cases involving white victims was not unconstitutional, the difficult and divisive wholesale constitutional challenges to capital punishment were thought to have been exhausted. Eight years of Reagan appointments had created a Court more favorably disposed toward sustaining capital sanctions than had been seen in Washington for a generation. The Bush years produced even more movement toward a Court sympathetic to what has been called (by Robert Weissberg) "deregulating death." The conventional wisdom has been that the legal challenge to capital punishment lacks both a broad issue to bring before the Court and a judicial constituency on the Court willing to entertain objections to executions. But major threats to the operational efficiency of executions do continue, and the Supreme Court has been forced to embrace rather extreme doctrine to protect the execution policies of the states from debilitating legal requirements.

What makes capital punishment a particular problem for criminal justice is that any pending legal challenge delays the imposition of the punishment. By contrast, a term of imprisonment normally begins right after sentencing, and jail confinement often starts after arrest. Even the most protracted appeal process in such circumstances usually does not postpone punishment. So allowing the legal process to go on is costless with respect to ensuring the imposition of imprisonment. The prisoner is not winning a battle against the state as a function of the appeal.

But no meaningful legal appeal can proceed without delay in the schedule of an execution. Even though the prisoner remains in custody, the punishment provided for the crime will not be imposed as long as any part of the review of the case is not complete. This situation creates two instrumental incentives in the appellate legal process. First, it encourages any defendant who wishes to avoid execution to do so by prolonging the process. Second, it makes the state representatives anxious to bring the legal review process to a close. As long as the appeal process is active, the state's penal purposes are frustrated. In such circumstances, the objective of a capital punishment regime must be to minimize the scrutiny of the legal system and to make the review process as short as possible.

All of this is a necessary introduction to the Supreme Court decision in the case of Penry v. Lynaugh, decided in June 1989. Johnny Paul Penry was convicted of rape murder in Texas and sentenced to death. Though twenty-two years old at the time of the crime, the defendant's mental age – representing both his ability to learn and the extent of his knowledge – was that of a 6 1/2-year-old. Penry's social maturity, or ability to function in the world, was that of a nine- or ten-year-old.

The defendant pressed two claims in the Supreme Court. First, he argued that it was unconstitutional to refuse his request for a jury instruction that his mental retardation could be considered as a mitigating circumstance. Second, the defendant asked the Court to rule that it would be cruel and unusual punishment under the Eighth Amendment to execute a mentally retarded person with Penry's level of impaired function.

The Penry case split the Court into three camps. Four Justices would have rejected both the need for an instruction and the defendant's per se Eighth Amendment claim. Four of their colleagues would have held that the execution of retarded persons like Penry would violate the Eighth Amendment. Justice O'Connor broke this deadlock by accepting the constitutional need for a mitigation instruction to the jury, but rejecting Penry's Eighth Amendment claim.

The O'Connor opinion is a puzzle. This Justice had earlier led the Court in concluding that imposition of the death penalty for a crime committed when the perpetrator was not yet sixteen would be an Eighth Amendment violation. Why not a per se rule for a defendant with abilities and controls less than those of the average ten-year-old? Even though the concept of "mental age" is rather arbitrary, the gap between Johnny Penry and normal seems wide enough for a per se rule. And Justice O'Connor's opinion in Penry is not unsympathetic to the arguments for an Eighth Amendment ban. Indeed, Justice Stevens, who dissented on this issue, saw no need to outline the reason for his conclusion because he believed that Justice O'Connor's opinion (to the opposite result) "adequately and fairly states the competing arguments respecting capital punishment of mentally retarded persons."

What, then, is the special danger of per se rules on this subject to Justice O'Connor? One answer to this question may be the potential of delays in large numbers of cases if an Eighth Amendment ban on executing the retarded were announced. This potential certainly distinguishes retardation from objectively incontrovertible factors such as chronological age, where only manifestly eligible defendants could benefit from a new legal rule. With no bright line between borderline and profoundly retarded adults and many hundreds of death row inmates who suffer from substantial cognitive deficits, the threat of a per se exemption for the retarded is by no means trivial. Excluding defendants who had not raised the issue at trial would seem morally objectional. Sorting out deserving from undeserving claims on the merits would take years. This possibility is nowhere discussed in Justice O'Connor's opinion, but its role as a subtext is one of few plausible explanations for the O'Connor vote.

Allowing the Eighth Amendment exemption would certainly produce delay and extra litigation. This is one important way in which a rule on chronological age is less troublesome than a rule on mental retardation. But not allowing such an exemption compromises the substantive claim that this ultimate penalty is reserved for only the most blameworthy criminal defendants. Nowhere in the opinions of the Penry case is there any argument to the effect that Johnny Penry is more culpable than a fifteen-year-old who murders, or than a fully competent rapist who is exempted by the Eighth Amendment from the death penalty because no death occurred as a result of his acts.

From the perspective of a capital punishment regime, the operational danger of setting up a new categorical target for postconviction litigators to aim for is quite severe. The substantive problem of allowing the ultimate penalty for a person profoundly disabled in judgment and cognitive ability is equally severe. A judicial system that recognizes the exemption will bog down in the extra hearings many defendants will gain. An operationally efficient capital punishment system will be morally compromised by its failure to protect those undeserving of death. These are the kind of hard choices that produce sharp and close divisions among the Justices in cases such as Penry.

Executing the Innocent

The clash between operational necessity and moral legitimacy is even more evident in the 1993 decision of Herrera v. Collins.The petitioner in Herrera had been convicted in the murder of a police officer and sentenced to death in 1982. Ten years later, after exhausting state remedies and fully prosecuting one federal habeas corpus petition, Herrera brought another federal habeas action alleging that he had proof of his actual innocence of the crime for which he was sentenced to death. He argued that executing a person for a crime he did not commit would constitute cruel and unusual punishment under the Eighth Amendment, and that he was entitled to one evidentiary hearing to establish his innocence. The federal district court ruled that a hearing was in order on this claim.

The Court's opinion, by Chief Justice Rehnquist, rejected Herrera's claim without deciding the question of whether the execution of an innocent person would constitute an Eighth Amendment violation: "We may assume, for the sake of argument in deciding this case, that in a capital case a truly persuasive demonstration of 'actual innocence' made after a trial would render the execution of the defendant unconstitutional, and warrant federal habeas corpus relief if there were no state avenue open to process such a claim." The opinion then holds that the proof offered by the defendant falls short of such a "persuasive demonstration" and therefore does not require the Court to decide whether the assumed constitutional standard should become an actual constitutional rule.

Two concurring opinions joined in by three Justices go further in supporting some constitutional protection against the execution of the probably innocent defendant, so that one might count as many as six of the nine Justices as supporting some form of Eighth Amendment protection in extreme cases. But why is the concession that executing the innocent might be a constitutional error so grudgingly made by all but the three dissenters? Why are the proper procedural channels and waiver rules so important not only to Justices Scalia and Thomas (who reject any constitutional protection), but also to the Chief Justice, who will only "assume for the sake of argument" that there is some constitutional problem with hanging the wrong man?

What we have here is a head-on collision between the operating needs of the capital punishment system and the sentiments and norms of Anglo-American criminal justice. As a substantive matter, nothing could be worse for a criminal justice system in the United States than the execution of an innocent person. No matter the procedural history that might precede such an event, no set of circumstances would seem to justify or excuse the outcome. The strongest reason for last-gasp hearings and procedures is to save the innocent from execution. Yet a rule that provided a right to a fresh hearing when new evidence might cast doubt on guilt would provide a procedural avenue of delay that many death row inmates would try to use. The few reversals that might result would be accompanied by scores of two-year delays tacked on at the end of a long and convoluted process.

In Herrera, the Court could not forthrightly endorse a basic sentiment of American justice without placing the efficient performance of the capital punishment regime in substantial jeopardy. This is the only plausible explanation for the ambivalent and indecisive analysis of most of the majority justices in Herrera on a basic issue of principle.

An Inevitable Conflict

Is the conflict between fairness and certainty, on the one hand, and operational efficiency, on the other, inevitable in a legal system with capital punishment? As a logical matter, the answer to this question is no. One path away from the conflict would be to reduce the amount of time necessary to assure just outcomes in capital trials and appeals. Given that the passage of time prejudices the capital punishment system, why not invest resources in excellent lawyers, quick trials, and expeditious appeals in capital cases?

In practice, however, such a formula is unlikely to be implemented. The political units that maintain criminal justice systems are unwilling to make the heavy investment in defense services that faster, high-quality justice would require, and this is particularly true of the states where death sentences and executions are most common. Moreover, limiting the acceleration of the process to procedures that would ensure fairness would not reduce the anger and hostility of prosecutors toward judicial review. Even the perfunctory version of appeal that has emerged in California during the last decade takes many years to accomplish and dominates the agenda of the state supreme court. With more than three thousand persons on death row nationwide, even the best case trial and appeal systems would produce a ten-year lag between trial and execution if a defendant wished to delay death. A delay of that length would generate the same frustration and tension observable in the current system.

Legal rights for persons accused of crime are never politically popular, and due process is a particular source of contention when fear of crime is a conspicuous part of the urban social landscape. Rules of substantive and procedural legality are cumbersome and inconvenient for those charged with the investigation and prosecution of crime. Francis Allen has recently reviewed the manifold threats to what he calls the "habits of legality" that currently operate in American society and government. I would add to his list of dangers the possibility that compromises of due process in capital cases will undermine procedural guarantees throughout the criminal justice system. Even though capital cases are infrequent, and even though the special tension betweeen judicial review and punitive closure only occurs in capital cases, insensitivity to considerations of legality can carry over from capital cases to the rest of criminal justice.

Such insensitivity is, to some extent, contagious no matter where in the justice system it occurs. But compromising principles in capital cases carries more than general contagion. Because death is the system's largest punishment, when rights are forfeited in capital cases, there is a momentum toward permitting the same compromises when the penal stakes are more modest. If the capital defendant cannot claim a right, who can?

Hostility to the delay produced in capital cases may also result in broad curtailment of legal remedies because capital cases cannot permissibly be singled out for special prejudice. In habeas corpus, for example, while the real target of both legislative and judicial restriction was death cases, the curtailment of the federal great writ in all state criminal cases was the means employed to achieve the restriction in death cases. This was a process of letting the tail wag the dog in collateral review of state criminal justice.

The destructive influence of capital cases on rights and remedies in other criminal cases is a historical process come full circle. The capital case was always a leading indicator of the direction in which due process guarantees would be extended. The constitutional right to counsel was extended in state capital cases a full generation before it was extended to state felonies. But even with that time gap, the flow from Powell v. Alabama in 1935 to Gideon v. Wainwright in 1963 was obvious. So there is symmetry if not poetic justice when capital cases now serve as leading indicators of a contraction in defendants' rights and the scope of judicial review.

Hostile reactions to delay are by no means the sole reason for restrictions on due process protections in the criminal justice system. But the contribution of the death penalty cases to the contraction of the scope of judicial review in habeas corpus, the growth of "harmless error" doctrine, and recent enthusiasms for enforcing procedural default rules is evident and very important. Further, when formalism neutralizes moral claims for procedural protection in capital cases, there can be no persuasive moral claim to protection in lesser cases. The behavior of the Supreme Court in capital cases thus becomes a portent of moral regression in criminal justice generally.

–Franklin E. Zimring

Franklin E. Zimring is William G. Simon Professor of Law at the University of California, Berkeley. This essay has been adapted from his chapter in The Killing State: Capital Punishment in Law, Politics, and Culture, edited by Austin Sarat, copyright 1999 by Oxford University Press. Published by arrangement with Oxford University Press, New York. Sources: Robert Burt, "Disorder in the Court," Michigan Law Review, vol. 85 (1987); Robert Weissberg, "Deregulating Death," Supreme Court Review (1983); Francis Allen, The Habits of Legality (Oxford University Press, 1996).