 |
|||||
|
|
|
|
|
||
|
|
|||||
|
|
|
|
|
||
From its fractured origins in Furman v. Georgia, 408 U. S. 238 (1972), this Court's capital punishment jurisprudence has matured into a comparatively stable body of law. Although tensions remain, legislatures now have a good idea how to write constitutional death penalty statutes, and courts know how to apply them.
Accepting the defendant's arguments would needlessly upset this system. In order to categorically exempt from capital punishment any person who tests as mentally retarded, the Court must overrule important parts of Penry v. Lynaugh, 492 U. S. 302 (1989) and Stanford v. Kentucky, 492 U. S. 361 (1989). Nothing in this case warrants such a disruption of the important and complex body of law surrounding capital punishment. Defendant's desired result will provide little additional aid to those for whom the death penalty is clearly unjust, but it would entail considerable cost. Since there is no "special justification" for departing from these precedents, cf. Dickerson v. United States, 530 U. S. 428, 443 (2000) (internal quotation marks omitted), Penry andStanford should not be overruled.
Few areas of constitutional law have been as contested or as complex as modern capital punishment jurisprudence. Starting with Furman v. Georgia,supra, the longest and one of the most divided opinions in the United States Reports, see Steiker & Steiker, Sober Second Thoughts: Reflections on Two Decades of Constitutional Regulation of Capital Punishment, 109 Harv. L. Rev. 355, 362 (1995), this Court's death penalty opinions have followed a labyrinthine path to the current, comparatively settled body of law. These opinions have created a unique set of procedures for capital cases that "are extensive and complex." 5 W. LaFave, J. Israel, & N. King, Criminal Procedure §26.1(b), p. 698 (2d ed. 1999). While these procedures may now strike an appropriate balance between Eighth Amendment interests and respect for state criminal justice systems, see Bilionis, Legitimating Death, 91 Mich. L. Rev. 1643, 1680 (1993), this equilibrium came at considerable cost. In addition to its complexity, the torturous path from Furman has burdened society with the needless retrial or release of some of our worst criminals, burdened federalism in a crucial area, and at times has promoted the very arbitrariness condemned by Furman. Any major expansion of the death penalty defendant's Eighth Amendment rights would reopen this can of worms. At this stage, there is no need to reinflict the wounds of the last 30 years.
A substantial cost of developing modern death penalty law is the difficulty of figuring it out and complying with it. The law's complexity is a weapon for capital defendants to frustrate their deserts. See Richmond v. Lewis, 506 U. S. 40, 54 (1992) (Scalia, J., dissenting). It also harms capital defendants by restricting the pool of available lawyers. "[T]he jurisprudence of death is so complex, so esoteric, so harrowing, this is one area where there aren't nearly enough lawyers willing and able to handle all the current cases." Kozinski & Gallagher, For an Honest Death Penalty, N. Y. Times, Mar. 8, 1995, p. A21, col. 1.
Developing such extensive regulations carries other costs as well. Courts and legislatures have had considerable difficulty in applying these decisions and in anticipating developments in death penalty law. An obvious example is found in the initial reaction to Furman. Furman provided very little guidance to state legislatures on how to write capital punishment statutes that complied with the Eighth Amendment. "But identifying the 'concerns' of Furman is a daunting task . . . . The opinions presented a staggering array of arguments for and against the constitutionality of the death penalty and offered little means, aside from shrewd political prediction of determining which arguments would dominate in the decision of any future cases." Steiker & Steiker,supra, 109 Harv. L. Rev., at 362. "Predictably, the variety of opinions supporting the judgment in Furman engendered confusion as to what was required in order to impose the death penalty in accord with the Eighth Amendment." Lockett v. Ohio, 438 U. S. 586, 599 (1978) (plurality). Responding to the concern with arbitrariness that ran through the Furman opinions, see Graham v. Collins, 506 U. S. 461, 483 (1993) (Thomas, J., concurring) (summarizing opinions), several states enacted mandatory death penalty statutes in order to foreclose arbitrariness. See Lockett, supra, at 599-600. Four years afterFurman, this Court decided that these states were wrong, invalidating their mandatory statutes and striking down numerous death sentences. See Woodsonv. North Carolina, 428 U. S. 280, 305 (1976); Roberts v. Louisiana, 428 U. S. 325, 336 (1976). This result was itself arbitrary. Given no real guidance, several state legislatures simply guessed wrong as to what this Court would require. As a result, numerous deserving defendants avoided their punishment, as all existing death sentences in these states were wiped out for the second time in four years. Each additional layer of complexity added to capital jurisprudence carries the risk of similar arbitrariness as defendants' punishments are determined by the ability of courts and legislatures to predict the next twist in this winding road.
Unfortunately, there are many other examples of the disruption caused by the development of the Eighth Amendment death penalty jurisprudence. TheLockett plurality sought to give "the clearest guidance that the Court can provide" through its holding. 438 U. S., at 602. Few promises from this Court have been less fulfilled. Its premise, that the sentencer in capital cases must not be prevented "from giving independent mitigating weight to aspects of the defendant's character and record and to circumstances of the offense," id., at 605, opened a Pandora's box for the legislatures, courts, and prosecutors left to implement it.
The explosive growth of Lockett's individualized sentencing branch of the post-Furman cases is well documented. See, e.g., McAllister, The Problem of Implementing a Constitutional System of Capital Punishment, 43 Kan. L. Rev. 1039, 1057-1060 (1995) (describing cases); id., at 1065 ("the virtually limitless expansion of the individualized sentencing principle"). Lockett, and this Court's numerous interpretations of its principles, see, e.g., 5 W. LaFave, J. Israel, & N. King, supra, §26.1(b), at 700, n. 17, led to the reversal of many death sentences where legislatures and courts failed to anticipate these sudden shifts in the Court's death penalty law. Since the Lockett "Court did not attempt to define the range of mitigating evidence encompassed by the individualization requirement," see Steiker & Steiker, supra, 109 Harv. L. Rev., at 390, years of "intricate litigation over states' fulfillment" of it was inevitable. See ibid. The result was a morass of confusion, reversals, and retrials. CJLF's survey of habeas reversals of state capital cases in the Eleventh Circuit found Lockett to be the single largest source of the reversals. See K. Scheidegger, Rethinking Habeas Corpus 36 (1989), reprinted in Habeas Corpus Issues: Hearings Before the Subcomm. on Civil and Constitutional Rights of the House Comm. on the Judiciary, 102d Cong., 1st Sess. 212, 251 (1991).
These and other reversals led to numerous expensive retrials. Many capital defendants subsequently escaped their sentence, whether from unsuccessful attempts at retrial or the wholesale invalidation of state death penalty schemes. In addition to frustrating the will of many juries, this also cost innocent people their lives. "The death penalty does, however, undeniably serve as a deterrent in one respect: once the sentence is carried out, the recidivism is quite low. And, the simple fact is, people sentenced to life in prison without parole, or even to a death sentence, do, occasionally, get out and do it again." Kozinski & Gallagher, Death: The Ultimate Run-On Sentence, 46 Case W. Res. L. Rev. 1, 25 (1995).
Federalism also paid a heavy price for establishing Furman's legacy. Criminal law is primarily a state matter. See, e.g., Payne v. Tennessee, 501 U. S. 808, 824 (1991); Argersinger v. Hamlin, 407 U. S. 25, 38 (1972). By addressing the most serious crimes with the most serious penalty, capital punishment sits at the apex of a state's criminal justice system. While the Eighth Amendment has some hold over this most important part of criminal justice, seePayne, 501 U. S., at 824, the detailed set of regulations erected by this Court since Furman substantially limits legitimate state sovereignty. Besides the extensive regulations mandated by this Court, federalism is further impaired by the state legislatures' understandable overreaction to these decisions. Thus, "the Court's emerging doctrine has motivated every death penalty jurisdiction to permit the introduction and consideration of 'any' mitigating factor," see Steiker & Steiker, supra, 109 Harv. L. Rev., at 391, even though the Eighth Amendment does not actually require this. See, e.g., Walton v. Arizona, 497 U. S. 639, 652 (1990) ("there is no . . . constitutional requirement of unfettered sentencing discretion in the jury, and States are free to structure and shape consideration of mitigating evidence in an effort to achieve a more rational and equitable administration of the death penalty") (internal quotation marks omitted); Boyde v. California, 494 U. S. 370, 377 (1990). Having been burned many times before, states are afraid to conduct the experimentation in this field that forms the heart of federalism.
Many of these costs have now been paid. The uncertainty and instability that plagued much of the post-Furman era is nearly gone. In its place is a mature, stable body of law. With regard to its narrowing requirement, "[t]he relevant Eighth Amendment law is well-defined." Richmond v. Lewis, 506 U. S. 40, 46 (1992). Similarly, for the individualized sentencing requirement, cases like Boyde v. California, supra, Saffle v. Parks, 494 U. S. 484 (1990), andBlystone v. Pennsylvania, 494 U. S. 299 (1990), have "brought a measure of formulaic simplicity to this concern as well." Bilionis, supra, 91 Mich. L. Rev., at 1653-1654. Since all the states now freely admit mitigating evidence, "virtually all of the current litigation concerning the individualization requirement is backward-looking, gauging the constitutionality of statutory provisions and state practices that are no longer in force." Steiker & Steiker, supra, 109 Harv. L. Rev., at 390. Even the famous tension between these two lines may be abating. Justice Thomas has sketched out a compromise position between these competing requirements. See Graham, 506 U. S., at 498-499 (Thomas, J., concurring). Although it involves an admittedly narrow reading of the Lockett-Eddings line, id., at 490, it is consistent with the spirit of Furman.
The long road from Furman has reached a balance between the need to insure that capital punishment is enforced both rationally and equitably and the states' legitimate interest in enforcing the death penalty. This achievement has come at considerable cost, and should not be disturbed without substantial justification. As the next section demonstrates, accepting the defendant's arguments would cause such a disturbance.
Creating a constitutional categorical exemption from capital punishment for mental retardation carries considerable costs. It would involve a disregard for precedent, a disruption of capital sentencing schemes, the creation of additional complexity, and the attendant arbitrariness that is all too similar to the turmoil associated with the aftermath of Furman v. Georgia. Any decision concerning the defendant's request must take these costs into account.
The clearest cost of a per se exemption from capital punishment for mental retardation is the damage to precedent. In Penry v. Lynaugh, 492 U. S. 302 (1989) (Penry I), this Court declined to create this categorical exemption from the death penalty. Since only two states prohibited executing the mentally retarded there was no national consensus that required extending that ban to all other states. See id., at 334.
Penry I's analysis still holds. It is true that more states now bar the execution of the mentally retarded than when Penry I was decided. But even the defendant's claim that 18 states with capital punishment and the federal government now bar the execution of the retarded does not support a different result. While it is possible to forge a new consensus under the "evolving standards of decency" strand of Eighth Amendment analysis, the defendant bears a "heavy burden" of establishing a consensus against a practice. See Stanford v. Kentucky, 492 U. S. 361, 373 (1989).
The defendant has not met that burden. His requested change instead is revolutionary, not evolutionary. Penry I and other cases demonstrate that the consensus of other state legislatures must be overwhelming before the few remaining states are forced to conform. It noted that when this Court held that the Eighth Amendment forbids executing the insane, no state permitted that practice. See Penry I, 492 U. S., at 334; Ford v. Wainwright, 477 U. S. 399, 408, n. 2 (1986). Stanford reinforces the importance of proceeding cautiously when evaluating a claim that national consensus against some punishment has reached constitutional proportions. In deciding that the Eighth Amendment did not bar executing someone who was 16 or 17 when the crime was committed, this Court held that 15 states with capital punishment forbidding this practice did "not establish the degree of national consensus" to support a finding of cruel and unusual punishment. Stanford, 492 U. S., at 371. As the Stanford Court noted, prior examples of striking down a practice under the national consensus theory had an overwhelming majority of states opposed to the relevant practice. The four examples given by the Stanford Court of an appropriate national consensus had either no state applying the relevant punishment, see Ford, 477 U. S., at 408 (insanity), only one state applying it, seeCoker v. Georgia, 433 U. S. 584, 595-596 (1977) (plurality) (rape of adult woman); Solem v. Helm, 463 U. S. 277, 300 (1983) (life without possibility of parole for minor offense), or eight states utilizing the punishment, see Enmund v. Florida, 458 U. S. 782, 792 (1982) (death for robbery in which accomplice kills). See Stanford, 492 U. S., at 371. Similarly, this Court has indicated that even if a majority of states followed a practice, that practice was not necessarily imposed on the other states under the Eighth Amendment. See Spaziano v. Florida, 468 U. S. 447, 464 (1984).
Amicus suggests that any national consensus must be at least as pervasive as in Enmund before it can be considered for being enshrined in the Eighth Amendment. Although this Court makes the final determination under the Eighth Amendment, the judgments of state legislatures and courts will "weigh heavily in the balance . . . ," see Enmund, 458 U. S., at 797, because there are no other appropriate indicators for this amorphous standard. If "emerging national consensus" is to mean something other than a byword for the Justices' own preferences, then the opinions of state legislatures must be listened to as the best objective source available. However, federalism and the integrity of state punishment systems both counsel against the aggressive use of such evidence.
Ordinarily, the Constitution does not impose a uniform approach by the states to any particular problem. Allowing a state to deviate from the majority rule of the other states is a hallmark of our federalism. Thus, "the Due Process Clause has never been perverted so as to force upon the forty-eight States a uniform code of criminal procedure." Carter v. Illinois, 329 U. S. 173, 175 (1946). As Justice Brandeis' famous dissent noted, each individual state must be allowed to serve as "a laboratory; and try novel social and economic experiments without risk to the rest of the country." New State Ice Co. v.Liebmann, 285 U. S. 262, 311 (1932) (Brandeis, J., dissenting).
Counting noses among the state legislatures and routinely forbidding those practices which come a few votes short threatens innovation by punishing states for varying from the norm. Except for extraordinary circumstances like the reaction to Furman, see Scheidegger, Capital Punishment in 1987: The Puzzle Nears Completion, 15 West. St. L. Rev. 95, 107 (1987), any finding by this Court that there is an emerging consensus against a particular practice is likely to be irreversible. A state wishing to reverse such a finding by forming a new consensus will be in a bind, as any sentences imposed contrary to the newly found consensus would be quickly struck down. Unless a considerable number of states act in concert to reverse the consensus, this Court's decision to strike down a particular practice is likely to stick.
This problem is compounded by the fact that the emerging national consensus doctrine can only be used to strike down sentences. States may freely vary from the consensus and forbid a punishment or procedure favored by most states. The reaction of some state legislatures to Penry I is one example. While the Eighth Amendment may forbid a punishment, it will not mandate states to act more harshly against offenders. Therefore, an aggressive use of the national consensus doctrine would lead to a rapid race to minimize punishments and impose procedural restraints upon the states. While punishments could be easily struck down, innovations favoring punishment would be hard to introduce and previously struck-down punishments would be almost impossible to rehabilitate. For example, Nevada's attempt to test Woodson's limits by enacting a mandatory death sentence for murder by a life-term inmate was struck down in Sumner v. Shuman, 483 U. S. 66, 77-78 (1987). Any attempt to restore this innovation would require numerous states to act in concert. A state that wishes to act alone in having a mandatory death sentence for repeat murderers will be stymied in court. Such a system is thus not simply a burden to federalism, but also threatens the entire capital punishment system. Requiring an overwhelming consensus to invalidate a practice places an important limit on a potentially dangerous doctrine.
The defendant has not found this consensus. According to his figures, only 18 of the 38 states with capital punishment exempt the mentally retarded. See Brief for Petitioner 39.(2) Even adding the federal government to the "anti"side does not change the fact that a majority of the jurisdictions with capital punishment do not categorically exempt the mentally retarded. Even this addition is dubious. The federal standard for mental retardation is "lacks the mental capacity to understand the death penalty and why it was imposed on that person." 18 U. S. C. §3596(c). This is the Ford competency standard applied to mental retardation. See Penry I, 492 U. S., at 333. Accepting the defendant's argument would require another standard. Since the defendant cannot even muster a majority of the states against his sentence, let alone a supermajority, his argument fails under Penry and Stanford.
Implicitly recognizing the weakness of his position, the defendant attempts to bolster his cause by invoking an argument previously rejected by this Court. His attempt to add the 12 states banning the death penalty to his total in order to find a majority for a categorical exclusion, see Brief for Petitioner 39, contradicts Stanford. The Stanford Court specifically rejected the relevance of counting the anti-death penalty states in any consensus concerning the administration of capital punishment. See 492 U. S., at 370, n. 2. Try as he might, the defendant cannot properly characterize this rejection as dicta. See Brief for Petitioner 39, n. 44. The difference between Stanford and the present case is small, with 15 states opposing executing juvenile offenders inStanford and 18 opposed to executing the retarded in the present case. Adding the states opposed to the death penalty shrinks the difference betweenStanford and this case, since two fewer states now prohibit the death penalty than when Stanford was decided. Compare Stanford, 492 U. S., at 371, n. 2, with Brief for Petitioner 39. Since the jurisdiction count in the two cases is essentially the same, finding a consensus against executing the retarded in this case would at the very least require overruling the analysis used in Stanford.
Accepting the defendant's claim would also cause considerable disruption to state capital punishment systems. Penry I held that a categorical exemption from punishment for mental retardation would apply retroactively on collateral review under the first exception to Teague v. Lane, 489 U. S. 288 (1989), even though it would be a new rule. See Penry I, 492 U. S., at 330. Unless this holding of Penry I were also overruled, every death row inmate who presented at the penalty phase at least some evidence that he was retarded would be entitled to a new trial to determine whether he was now exempt from his death sentence. This only begins the disruption. Because retardation claims would apply retroactively on federal habeas, they would also be exempt from limits on successive habeas petitions or on default for failing to develop facts in state court. See 28 U. S. C. §§2244(b)(2)(A), 2254(e)(2)(A)(i). Therefore, any death row inmate could try to raise a retardation claim on federal habeas. Since a diagnosis of retardation is more subjective and therefore more subject to falsification than a diagnosis of the measles, the federal and state courts risk being inundated by the claims of death row inmates utilizing a new weapon in their arsenal of delay. As in the aftermath of Furman, many well-deserved death sentences would be set aside. See supra, at 10. Even in cases where the malingerers are identified as such, the process will further delay the already overdue process of capital punishment.
|
|
||||
|
|
|
|
||
Footnote 2. Virginia properly contests this number. See Brief for Respondent, Part II B. Even if the defendant's count is accepted, he still has not forged a
consensus as defined in Stanford.