 |
|||||
|
|
|
|
|
||
|
|
|||||
|
|
|
|
|
||
Apprendi v. New Jersey, 530 U. S. 466, 478 (2000) states, "Any possible distinction between an 'element' of a felony offense and a 'sentencing factor' was unknown to the practice of criminal indictment, trial by jury, and judgment by court as it existed during the years surrounding our Nation's founding." (Footnote omitted). That is not entirely correct. The opinion goes on to say, "The defendant's ability to predict with certainty the judgment from the face of the felony indictment flowed from the invariable linkage of punishment with crime." Ibid. (emphasis added). The overstatement of this sentence is revealed by the citation which follows: "See 4 Blackstone 369-370 (after verdict, and barring a defect in the indictment, pardon, or benefit of clergy, 'the court must pronounce that judgment, which the law hath annexed to the crime' . . . )." Id., at 478-479 (emphasis added, Apprendi's emphasis deleted). Apprendi thus skipped lightly over a major feature of common law sentencing.
The benefit of clergy began, as the name implies, as an exemption from punishment by secular courts for members of the clergy, as very broadly defined. See 4 Blackstone, Commentaries 358-359 (1st ed. 1769); 1 J. Stephen, History of the Criminal Law of England 459 (1883). Persons so exempted were tried in ecclesiastical courts and punished very lightly. See G. Dalzell, Benefit of Clergy in America 11 (1955). Later, any man who could read was assumed to be a clergyman, a fiction which drifted further from reality as learning spread. 4 Blackstone, at 360. In 1487, by statute, literate laymen were distinguished from actual clergymen, and the former could only claim the benefit once. 1 Stephen, at 462.
Offenders allowed the benefit of clergy were branded on the thumb to more readily identify them on the next offense. Ibid. After 1576, they were not turned over to the church, but rather jailed for up to one year. Ibid. The privilege was extended to peers without a reading requirement in 1547, and to women in all cases in 1692. Ibid. The reading test was routinely faked. See Dalzell, at 24-25. In 1705, it was abolished by statute. See 1 Stephen, at 462. As Blackstone notes, punishing the illiterate more severely than those who had received the benefit of education was backwards. 4 Blackstone, at 363. A further innovation in England was to punish offenders by sending them to America, id., at 363-364, but this enhancement was obviously unavailable to colonial courts.
While the benefit of clergy was being extended to an expanding class of felons, it was also being restricted to a shrinking class of felonies. See 1 Stephen, at 469. The most important of these restrictions was among the first, eliminating the benefit in cases of "murder of malice prepensed." See 1 Stephen, at 464-465 (discussing statutes of 1531 and 1547); Acker & Lanier, Capital Murder from Benefit of Clergy to Bifurcated Trials: Narrowing the Class of Offenses Punishable by Death, 29 Crim. L. Bull. 291, 293, and n. 8 (1993). This became the distinction between murder and manslaughter. Acker & Lanier, at 294.
Defendant asserts that at common law murder was capital homicide and manslaughter was not, i.e., that "the death penalty was only available for" murder. Brief for Petitioner 25. He is mistaken. Manslaughter was a "clergyable felony," (2) while murder was a "felony without benefit of clergy." Both classes of felony were "death-eligible," to use a modern term, but the latter class meant death in every case, subject only to executive clemency. Repeat offenders could be and were hanged in America for clergyable felonies in the years just before the American Revolution, see Dalzell, at 152, and in some states into the nineteenth century. See id., at 259.
In a nutshell, sentencing for homicide at the time of the American Revolution worked as follows. First, the jury would convict of either murder or manslaughter. "Malice aforethought" was an element of the crime of murder, which distinguished it from manslaughter. As an element, malice aforethought had to be specified in the indictment and proved to the trial jury. See 2 W. Hawkins, Pleas of the Crown, ch. 33, §25, p. 342 (2d ed. 1726).
If the defendant was convicted of manslaughter, then the trial judge had to determine whether he was eligible for benefit of clergy. This meant determining whether a defendant claiming to be a clergyman really was, see 2 Hawkins, ch. 33, §113, at 360, whether defendant was eligible as a peer, see Dalzell, at 22, and whether the defendant was ineligible as a repeat offender. See id., at 152.
"Who is to judge whether a Person who demands the Benefit of the Clergy, has a Right to it or not: I take it, That in all Cases the Temporal Judge is to determine both, whether the Crime be within the Benefit of the Clergy, and also whether the Person who demands it be qualified to demand it or not . . . ." 2 Hawkins, ch. 33, §113, at 359.
Jury participation in the post-verdict eligibility determination was limited to one specific factual question. If the prosecution alleged a prior crime and prior use of the benefit of clergy, and the defendant denied he was the same person as the one tried in the earlier proceeding, a jury was empaneled "to try the question of identity in issue." 1 J. Chitty, Criminal Law *688 (1819); State v. Carroll, 24 N. C. 183, 184 (1842). Even so, recidivism was not an element of the offense, as it was not specified in the indictment or determined by the guilt-phase jury, and no question but identity was determined by a jury at all.
At the time of the American Revolution, benefit of clergy was the law in 12 of the 13 states. Dalzell, at 233. The most famous application was in the Boston Massacre case. Two of the British soldiers were convicted of manslaughter, granted benefit of clergy, burned on the hand, and released. See id., at 204. This was accepted practice, even in a case as controversial as this one. Ibid.
Benefit of clergy did not survive long after the adoption of the Bill of Rights in most states. Broad penal reforms, including the building of state prisons, soon swept it into the dustbin of history. See id., at 249 (Virginia, 1796); id., at 252 (Maryland, 1810); id., at 254 (New York, 1788); id., at 255 (Massachusetts, 1785); id., at 267 (Georgia, 1817). Even so, it was a part of the common law heritage and was a law in effect in most states at the time the Sixth Amendment was adopted. (3) A Sixth Amendment argument based on the law as it existed in 1791, see Apprendi, 530 U. S., at 478-480, necessarily fails if the rule asserted would have been inconsistent with the practice of benefit of clergy.
Certainly it is not true that every fact that went into the determination of punishment was an element of the offense. The fact that determined whether a homicide was clergyable manslaughter or nonclergyable murder was an element, but the facts that determined whether the manslaughter convict actually received the benefit were not. Common law practice was consistent with the distinction between elements and sentencing factors expressed in Justice Scalia's dissent in Almendarez-Torres v. United States, 523 U. S. 224, 257, n. 2 (1998), quoted infra, at 17.
First-degree murder under Ariz. Rev. Stat. §13-1105(C) is analytically similar to a "clergyable felony" in 1791. It is punishable by death or a lesser punishment, with the choice to be made in post-verdict proceedings. Those proceedings are not entirely discretionary but involve some fact-finding. While the law may assign some of the fact-finding to a jury, as the common law did with disputed identity in prior conviction cases, that does not make the fact an element of the offense or bring it within the Sixth Amendment. These sentencing factors may be found by judges, so long as the maximum punishment is within the range established for the offense.
A. Rejection of the Purely Functional Definition.
Defendant argues for a purely functional definition of elements of the offense for constitutional purposes. See Brief for Petitioner 10. Under this view, it would be completely irrelevant how the legislature defines the elements of the offense or what it specifies as the maximum punishment for the base offense. Instead, courts would look only at what facts lead to what punishment and apply the constitutional protections of burden of proof, jury trial, and, in federal cases, indictment to the facts needed for the punishment. This purely functional approach cannot be reconciled with Mullaney v. Wilbur, 421 U. S. 684 (1975), Patterson v. New York, 432 U. S. 197 (1977), and Edwards v. United States, 523 U. S. 511 (1998).
Mullaney arose in a state with the traditional definitions of murder and manslaughter. "Malice aforethought" was the element of murder which distinguished it from manslaughter, but it was defined in the negative, as the absence of "heat of passion on sudden provocation." 421 U. S., at 686-687; see also id., at 701-702 (discussing "proving a negative"). In these circumstances, placing the burden of proof on the defendant violated the Due Process Clause. See id., at 703-704.
The other bookend of this pair is Patterson v. New York. The New York law at issue in that case did not make "malice" an element of second-degree murder, but instead required only intentional killing. 432 U. S., at 198. In place of the common law "malice" element, New York provided an affirmative defense of "extreme emotional disturbance," which would reduce the crime to manslaughter if proved by the defendant by a preponderance of the evidence. Id., at 199-200. The Court upheld this system, rejecting the argument that Mullaney controlled. See id., at 214-216.
Mullaney and Patterson cannot be reconciled with a purely functional analysis. In both Maine and New York, the state could rest after proving only intentional killing. In both states, the defendant could not reduce the murder to manslaughter by raising only a reasonable doubt as to the existence of heat of passion or extreme emotional distress; instead, he had to affirmatively prove it. Yet one system was constitutional and the other was not.
The difference, says Patterson, was that in Maine malice "was part of the definition of [the] crime" of murder. Id., at 216. The legislative definition of the elements does matter. Once the legislature declares an element, it cannot presume the element and shift the burden of proof, even though, within limits, it can functionally accomplish the same result by redefining the crime and its associated affirmative defenses.
Patterson noted that "there are obviously constitutional limits beyond which the States may not go" in changing elements into affirmative defenses. Id., at 210. While the existence of those limits may be obvious, the extent of them is not. Since Patterson, the Court has not needed to define the limits because the legislatures have not pushed them. See Brief for Criminal Justice Legal Foundation as Amicus Curiae in Harris v. United States, No. 00-10666, p. 19; Hoffman, Apprendi v. New Jersey: Back to the Future?, 38 Am. Crim. L. Rev. 255, 272 (2001).
Patterson held that the legislature may, within limits, redefine a degree-fixing element as an affirmative defense and place the burden of proof on the defendant. If so, there would seem to be no logical reason why it cannot, within limits, set a range of punishments for a single offense and structure the choice within that range based on facts to be determined at sentencing. That is, in essence, what the Sentencing Guidelines do, and this Court rejected a constitutional attack on them just four years ago.
In Edwards v. United States, 523 U. S. 511, 512-513 (1998), the defendants were convicted of a cocaine conspiracy. The sentences were within the statutory range for any kind of cocaine. See id., at 515. However, the sentence determination under the Guidelines required the judge, not the jury, to determine that the defendant's "illegal conduct had involved both cocaine and crack." Id., at 513 (emphasis in original).
Under a purely functional analysis of the type proposed by defendant, the maximum legal sentence in Edwards would have been the Guidelines maximum based on the facts found by the jury verdict, which did not specify crack as opposed to powder. See id., at 513; cf. Brief for Petitioner 10. Functionally, the defendants in Edwards could not be sentenced to more than the Guidelines provide for powder without a finding that crack was involved, and this finding, under defendant's theory, would have to be made by the jury and not the judge. Yet the Edwards Court unanimously rejected the Sixth Amendment argument, noting that the actual sentence did not "exceed[ ] the maximum that the statutes permit for cocaine-only conspiracy." 523 U. S., at 515 (emphasis added). Apprendi reaffirmed this holding. Apprendi v. New Jersey, 530 U. S. 466, 497, n. 21 (2000).
Patterson noted that the ability of the legislature to redefine elements as defenses was not unlimited, and that drastic restructuring to evade the basic burden of proof requirement would be unconstitutional. See 432 U. S., at 210. Citing this portion of Patterson, Apprendi noted that a similar constraint could apply in the event of a hypothetical "extensive revision of the State's entire criminal code . . . ." 530 U. S., at 491, n. 16. The situation with capital punishment and "eligibility" factors is the antithesis of this hypothetical.
By requiring the finding of an aggravating factor to narrow the class of defendants eligible for capital punishment, Congress and the state legislatures have not removed constitutional protections from elements that traditionally had those protections. Instead, they created an entirely new requirement purely for the benefit of the defendant, compared to what the law was before. See Dobbert v. Florida, 432 U. S. 282, 294-295 (1977) (post-Furman changes "ameliorative"). Degrees of murder were invented in 1794 to narrow the eligible class, see McGautha v. California, 402 U. S. 183, 198 (1971), but no further narrowing occurred through 1971. See id., at 199. The American Law Institute had proposed "aggravating circumstances" in substantially their present form in 1959, see id., at 202, 223-224, but no state voluntarily adopted them. See id., at 203. After Furman v. Georgia, 408 U. S. 238 (1972), they were widely adopted to meet the constitutional narrowing requirement. See Tuilaepa v. California, 512 U. S. 967, 971-972 (1994).
As an innovation in the law which operates purely in favor of the defendant relative to the traditional definition, death-eligibility sentencing factors are well within the constitutional limits indicated by Patterson and Apprendi. See Patterson, 432 U. S., at 207 (noting the new defense was "a substantially expanded version of the older heat-of-passion concept"). Under these decisions, legislatures have the power to define the elements of crimes and the range of punishments for them. Affirmative defenses and sentencing factors which reduce the offense or punishment or structure the sentencing decision with the statutory range are also within the legislative power. There is no danger that states will take advantage to remove from the jury the elements that traditionally defined first-degree murder and recharacterize them as "aggravating factors." It has been 26 years since Proffitt, with no movement in that direction yet. The principal legislative innovation since then has been to add new categorical exclusions, usually with the burden of proof on the defendant. See Cal. Penal Code §190.5(a) (under 18, burden on defendant); Tr. of Oral Arg. in Atkins v. Virginia, No. 00-8452, p. 13 (retardation, burden on defendant in every state). Ameliorative innovations may be stifled if a brickload of burdens automatically attaches to them. See Patterson, 432 U. S., at 207-208.
Apprendi is squarely premised on the consistency of its rule with all precedents except, possibly, Almendarez-Torres v. United States, 523 U. S. 224 (1998). The Apprendi Court says a contrary ruling would have been a "rejection of the otherwise uniform course of decision during the entire history of our jurisprudence." 530 U. S., at 490 (emphasis added). To accept defendant's expansive interpretation of the Apprendi rule would require repudiation ofApprendi's premise. Not only does he ask for overruling of Walton v. Arizona, 497 U. S. 639 (1990), as well as Spaziano v. Florida, 468 U. S. 447 (1984) and Hildwin v. Florida, 490 U. S. 638 (1989) (per curiam), see Brief for Petitioner 23, n. 16, defendant's interpretation of Apprendi is also inconsistent with a pair of well-established precedents that have been part of the fabric of the law for a quarter century and with a unanimous decision that predatesApprendi by only two years. Nor does it end there, as we will show in part IV, infra. Such an interpretation cannot be correct.
In Apprendi v. New Jersey, 530 U. S. 466, 501 (2000) (concurring opinion), Justice Thomas suggested a constitutional distinction between facts that increase punishment and facts that mitigate punishment. Drawing the line in this manner would distinguish Patterson v. New York, 432 U. S. 197 (1977) on its facts, but it would be inconsistent with the reasoning of that decision, as discussed in part II, supra. In addition, it would have the effect of inhibiting reforms, as Patterson warned against. See supra, at 11-12.
The traditional felony murder rule was that a participant in a robbery in which a person was killed was guilty of murder, even though he neither committed nor intended the killing, and 20 states even permitted the death penalty in such cases in post-Furman statutes. See Enmund v. Florida, 458 U. S. 782, 820 (1982) (O'Connor, J., dissenting). Under Enmund and Tison v. Arizona, 481 U. S. 137, 138 (1987), states may still impose the death penalty on a nontriggerman who intends to kill or acts with reckless disregard of human life.
Many would consider it a salutary reform to narrow the definition to exclude all persons lacking intent to kill. However, the state may be unwilling to shoulder the burden of proof. See Patterson, 432 U. S., at 207. The perpetrators of the felony are frequently the only living witnesses to the murder. Indeed, the victims are often killed precisely for the purpose of preventing their testimony. As one notorious criminal explained his cold-blooded murder of two teenage boys, " 'I couldn't have no punks running around that could do that [identify him], so I wasted them.' " People v. Harris, 28 Cal. 3d 935, 945, 623 P. 2d 240, 245 (1981).
California dropped its felony-murder special circumstance down to the Tison minimum precisely because of the difficulty of proof in the multiple perpetrators situation. See Cal. Penal Code §190.2(d). (4) A partial affirmative defense of lack of intent would be far easier to enact if the burden of proof could be placed on the defendant. Yet under a straight aggravating/mitigating definition of "elements," intent to kill would surely be aggravating, and burden-shifting would be impermissible.
The Patterson Court had it right the first time. It is better to leave the specification of the elements of crimes to the legislative branch and permit innovation, so long as wholesale evasions are not attempted. See also Schad v. Arizona, 501 U. S. 624, 639 (1991) (plurality opinion); Montana v.Egelhoff, 518 U. S. 37, 58 (1996) (Ginsburg, J., concurring in the judgment).
|
|
||||
|
|
|
|
||
2. Except that stabbing another to death was not clergyable, even if it was manslaughter rather than murder. 4 Blackstone, at 193. This exception was
construed so narrowly as to virtually repeal it. See ibid.
3. In 1791, it had been abolished in Massachusetts and New York, as noted above, and it had never been the law in Connecticut. See id., at 257.
4. The author of Proposition 115, which added this subdivision to §190.2, so informed counsel for amicus at the time of the campaign.