Pee Curiam.

Petitioner in No. 69-5003 was convicted of murder in Georgia and was sentenced to death pursuant to Ga. Code Ann. §26-1005 (Supp. 1971) (effective prior to July 1, 1969). 225 Ga. 253, 167 S. E. 2d 628 (1969). Petitioner in No. 69-5030 was convicted of rape in Georgia and was sentenced to death pursuant to Ga. Code Ann. § 26-1302 (Supp. 1971) (effective prior to July 1, 1969). 225 Ga. 790, 171 S. E. 2d 501 (1969). Petitioner in No. 69-5031 was convicted of rape in Texas and was sentenced to death pursuant to Tex. Penal Code, Art. 1189 (1961). 447 S. W. 2d 932 (Ct. Crim. App. 1969). Certiorari was granted limited to the following question: “Does the imposition and carrying out of the death penalty in [these cases] constitute cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments?” 403 U. S. 952 (1971). The Court holds that the imposi*240tion and carrying out of the death penalty in these cases constitute cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments. The judgment in each case is therefore reversed insofar as it leaves undisturbed the death sentence imposed, and the cases are remanded for further proceedings.

So ordered.

Me. Justice Douglas, Mr. Justice Brennan, Mr. Justice Stewart, Mr. Justice White, and Mr. Justice Marshall have filed separate opinions in support of the judgments. The Chief Justice, Mr. Justice Black-mun, Mr. Justice Powell, and Mr. Justice Rehnquist have filed separate dissenting opinions.

Mr. Justice Douglas,

concurring.

In these three cases the death penalty was imposed, one of them for murder, and two for rape. In each the determination of whether the penalty should be death or a lighter punishment was left by the State to the discretion of the judge or of the jury. In each of the three cases the trial was to a jury. They are here on petitions for certiorari which we granted limited to the question whether the imposition and execution of the death penalty constitute “cruel and unusual punishment” within the meaning of the Eighth Amendment as applied to the States by the Fourteenth.1 I vote to vacate each judgment, believing that the exaction of the death penalty does violate the Eighth and Fourteenth Amendments.

*241That the requirements of due process ban cruel and unusual punishment is now settled. Louisiana ex rel. Francis v. Resweber, 329 U. S. 459, 463, and 473-474 (Burton, J., dissenting); Robinson v. California, 370 U. S. 660, 667. It is also settled that the proscription of cruel and unusual punishments forbids the judicial imposition of them as well as their imposition by the legislature. Weems v. United States, 217 U. S. 349, 378-382.

Congressman Bingham, in proposing the Fourteenth Amendment, maintained that “the privileges or immunities of citizens of the United States” as protected by the Fourteenth Amendment included protection against “cruel and unusual punishments:”

“[M]any instances of State injustice and oppression have already occurred in the State legislation of this Union, of flagrant violations of the guarantied privileges of citizens of the United States, for which the national Government furnished and could furnish by law no remedy whatever. Contrary to the express letter of your Constitution, ‘cruel and unusual punishments’ have been inflicted under State laws within this Union upon citizens, not only for crimes committed, but for sacred duty done, for which and against which the Government of the United States had provided no remedy and could provide none.” Cong. Globe, 39th Cong., 1st Sess., 2542.

Whether the privileges and immunities route is followed, or the due process route, the result is the same.

It has been assumed in our decisions that punishment by death is not cruel, unless the manner of execution can be said to be inhuman and barbarous. In re Kemmler, 136 U. S. 436, 447. It is also said in our opinions *242that the proscription of cruel and unusual punishments “is not fastened to the obsolete but may acquire meaning as public opinion becomes enlightened by a humane justice.” Weems v. United States, supra, at 378. A like statement was made in Trop v. Dulles, 356 U. S. 86, 101, that the Eighth Amendment “must draw its meaning from the evolving standards of decency that mark the progress of a maturing society.”

The generality of a law inflicting capital punishment is one thing. What may be said of the validity of a law on the books and what may be done with the law in its application do, or may, lead to quite different conclusions.

It would seem to be incontestable that the death penalty inflicted on one defendant is “unusual” if it discriminates against him by reason of his race, religion, wealth, social position, or class, or if it is imposed under a procedure that gives room for the play of such prejudices.

There is evidence that the provision of the English Bill of Rights of 1689, from which the language of the Eighth Amendment was taken, was concerned primarily with selective or irregular application of harsh penalties and that its aim was to forbid arbitrary and discriminatory penalties of a severe nature: 2

“Following the Norman conquest of England in 1066, the old system of penalties, which ensured equality between crime and punishment, suddenly disappeared. By the time systematic judicial records were kept, its demise was almost complete. With the exception of certain grave crimes for which the punishment was death or outlawry, the arbitrary fine was replaced by a discretionary *243amercement. Although amercement’s discretionary character allowed the circumstances of each case to be taken into account and the level of cash penalties to be decreased or increased accordingly, the amercement presented an opportunity for excessive or oppressive fines.

“The problem of excessive amercements became so prevalent that three chapters of the Magna Carta were devoted to their regulation. Maitland said of Chapter 14 that ‘very likely there was no clause in the Magna Carta more grateful to the mass of the people.’ Chapter 14 clearly stipulated as fundamental law a prohibition of excessiveness in punishments:

“ ‘A free man shall not be amerced for a trivial offence, except in accordance with the degree of the offence; and for a serious offence he shall be amerced according to its gravity, saving his livelihood; and a merchant likewise, saving his merchandise; in the same way a villein shall be amerced saving his wainage; if they fall into our mercy. And none of the aforesaid amercements shall be imposed except by the testimony of reputable men of the neighborhood.’ ”

The English Bill of Rights, enacted December 16,1689, stated that “excessive bail ought not to be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.” 3 These were the words chosen for our Eighth Amendment. A like provision had been in Virginia’s Constitution of 17764 and in the constitutions *244of seven other States.5 The Northwest Ordinance, enacted under the Articles of Confederation, included a prohibition of cruel and unusual punishments.6 But the debates of the First Congress on the Bill of Rights throw little light on its intended meaning. All that appears is the following: 7

“Mr. Smith, of South Carolina, objected to the words ‘nor cruel and unusual punishments;’ the import of them being too indefinite.

“Mr. Livermore: The clause seems to express a great deal of humanity, on which account I have no objection to it; but as it seems to have no meaning in it, I do not think it necessary. What is meant by the terms excessive bail? Who are to be the judges? What is understood by excessive fines? It lies with the court to determine. No cruel and unusual punishment is to be inflicted; it is sometimes necessary to hang a man, villains often deserve whipping, and perhaps having their ears cut off; but are we in future to be prevented from inflicting these punishments because they are cruel? If a more lenient mode of correcting vice and deterring others from the commission of it could be invented, it would be very prudent in the Legislature to adopt it; but until we have some security that this will be done, we ought not to be restrained from making necessary laws by any declaration of this kind.”

The words “cruel and unusual” certainly include pen*245alties that are barbaric. But the words, at least when read in light of the English proscription against selective and irregular use of penalties, suggest that it is “cruel and unusual” to apply the death penalty — or any other penalty — selectively to minorities whose numbers are few, who are outcasts of society, and who are unpopular, but whom society is willing to see suffer though it would not countenance general application of the same penalty across the board.8 Judge Tuttle, indeed, made abundantly clear in Novak v. Beto, 453 F. 2d 661, 673-679 (CA5) (concurring in part and dissenting in part), that solitary confinement may at times be “cruel and unusual” punishment. Cf. Ex parte Medley, 134 U. S. 160; Brooks v. Florida, 389 U. S. 413.

The Court in McGautha v. California, 402 U. S. 183, 198, noted that in this country there was almost from the beginning a “rebellion against the common-law rule imposing a mandatory death sentence on all convicted *246murderers.” The first attempted remedy was to restrict the death penalty to defined offenses such as “premeditated” murder.9 Ibid. But juries “took the *247law into their own hands” and refused to convict on the capital offense. Id., at 199.

“In order to meet the problem of jury nullification, legislatures did not try, as before, to refine further the definition of capital homicides. Instead they adopted the method of forthrightly granting juries the discretion which they had been exercising in fact.” Ibid.

The Court concluded: “In light of history, experience, and the present limitations of human knowledge, we find it quite impossible to say that committing to the untrammeled discretion of the jury the power to pronounce life or death in capital cases is offensive to anything in the Constitution.” Id., at 207.

The Court refused to find constitutional dimensions in the argument that those who exercise their discretion to send a person to death should be given standards by which that discretion should be exercised. Id., at 207-208.

A recent witness at the Hearings before Subcommittee No. 3 of the House Committee on the Judiciary, 92d Cong., 2d Sess., Ernest van den Haag, testifying on H. R. 8414 et al.,10 stated:

“Any penalty, a fine, imprisonment or the death penalty could be unfairly or unjustly applied. The *248vice in this case is not in the penalty but in the process by which it is inflicted. It is unfair to inflict unequal penalties on equally guilty parties, or on any innocent parties, regardless of what the penalty is.” Id., at 116-117. (Emphasis supplied.)

But those who advance that argument overlook McGautha, supra.

We are now imprisoned in the McGautha holding. Indeed the seeds of the present cases are in McGautha. Juries (or judges, as the case may be) have practically untrammeled discretion to let an accused live or insist that he die.11

*249Mr. Justice Field, dissenting in O’Neil v. Vermont, 144 U. S. 323, 340, said, “The State may, indeed, make the drinking of one drop of liquor an offence to be punished by imprisonment, but it would be an unheard-of cruelty if it should count the drops in a single glass and make thereby a thousand offences, and thus extend the punishment for drinking the single glass of liquor to an imprisonment of almost indefinite duration.” What the legislature may not do for all classes uniformly and systematically, a judge or jury may not do for a class that prejudice sets apart from the community.

There is increasing recognition of the fact that the basic theme of equal protection is implicit in “cruel and unusual” punishments. “A penalty . . . should be considered 'unusually’ imposed if it is administered arbitrarily or discriminatorily.” 12 The same authors add that “[t]he extreme rarity with which applicable death penalty provisions are put to use raises a strong inference of arbitrariness.” 13 The President’s Commission on Law Enforcement and Administration of Justice recently concluded: 14

“Finally there is evidence that the imposition of the death sentence and the exercise of dispensing power by the courts and the executive follow discriminatory patterns. The death sentence is disproportionately imposed and carried out on the *250poor, the Negro, and the members of unpopular groups.”

A study of capital cases in Texas from 1924 to 1968 reached the following conclusions:15

“Application of the death penalty is unequal: most of those executed were poor, young, and ignorant.

*251“Seventy-five of the 460 cases involved co-defendants, who, under Texas law, were given separate trials. In several instances where a white and a Negro were co-defendants, the white was sentenced to life imprisonment or a term of years, and the Negro was given the death penalty.

“Another ethnic disparity is found in the type of sentence imposed for rape. The Negro convicted of rape is far more likely to get the death penalty than a term sentence, whereas whites and Latins are far more likely to get a term sentence than the death penalty.”

Warden Lewis E. Lawes of Sing Sing said:16

“Not only does capital punishment fail in its justification, but no punishment could be invented with so many inherent defects. It is an unequal punishment in the way it is applied to the rich and to the poor. The defendant of wealth and position never goes to the electric chair or to the gallows. Juries do not intentionally favour the rich, the law is theoretically impartial, but the defendant with ample means is able to have his case presented with every favourable aspect, while the poor defendant often has a lawyer assigned by the court. Sometimes such assignment is considered part of political patronage; usually the lawyer assigned has had no experience whatever in a capital case.”

Former Attorney General Ramsey Clark has said, “It is the poor, the sick, the ignorant, the powerless and the hated who are executed.” 17 One searches our chron*252icles in vain for the execution of any member of the affluent strata of this society. The Leopolds and Loebs are given prison terms, not sentenced to death.

Jackson, a black, convicted of the rape of a white woman, was 21 years old. A court-appointed psychiatrist said that Jackson was of average education and average intelligence, that he was not an imbecile, or schizophrenic, or psychotic, that his traits were the product of environmental influences, and that he was competent to stand trial. Jackson had entered the house after the husband left for work. He held scissors against the neck of the wife, demanding money. She could find none and a struggle ensued for the scissors, a battle which she lost; and she was then raped, Jackson keeping the scissors pressed against her neck. While there did not appear to be any long-term traumatic impact on the victim, she was bruised and abrased in the struggle but was not hospitalized. Jackson was a convict who had escaped from a work gang in the area, a result of a three-year sentence for auto theft. He was at large for three days and during that time had committed several other offenses — burglary, auto theft, and assault and battery.

Furman, a black, killed a householder while seeking to enter the home at night. Furman shot the deceased through a closed door. He was 26 years old and had finished the sixth grade in school. Pending trial, he was committed to the Georgia Central State Hospital for a psychiatric examination on his plea of insanity tendered by court-appointed counsel. The superintendent reported that a unanimous staff diagnostic conference had concluded “that this patient should retain his present diagnosis of Mental Deficiency, Mild to Moderate, with Psychotic Episodes associated with Convulsive Disorder.” The physicians agreed that “at present the patient is not psychotic, but he is not capable of cooperating with his counsel in the preparation of his *253defense”; and the staff believed “that he is in need of further psychiatric hospitalization and treatment.”

Later, the superintendent reported that the staff diagnosis was Mental Deficiency, Mild to Moderate, with Psychotic Episodes associated with Convulsive Disorder. He concluded, however, that Furman was “not psychotic at present, knows right from wrong and is able to cooperate with his counsel in preparing his defense.”

Branch, a black, entered the rural home of a 65-year-old widow, a white, while she slept and raped her, holding his arm against her throat. Thereupon he demanded money and for 30 minutes or more the widow searched for money, finding little. As he left, Jackson said if the widow told anyone what happened, he would return and kill her. The record is barren of any medical or psychiatric evidence showing injury to her as a result of Branch’s attack.

He had previously been convicted of felony theft and found to be a borderline mental deficient and well below the average IQ of Texas prison inmates. He had the equivalent of five and a half years of grade school education. He had a “dull intelligence” and was in the lowest fourth percentile of his class.

We cannot say from facts disclosed in these records that these defendants were sentenced to death because they were black. Yet our task is not restricted to an effort to divine what motives impelled these death penalties. Rather, we deal with a system of law and of justice that leaves to the uncontrolled discretion of judges or juries the determination whether defendants committing these crimes should die or be imprisoned. Under these laws no standards govern the selection of the penalty. People live or die, dependent on the whim of one man or of 12.

Irving Brant has given a detailed account of the Bloody Assizes, the reign of terror that occupied the *254closing years of the rule of Charles II and the opening years of the regime of James II (the Lord Chief Justice was George Jeffreys):

“Nobody knows how many hundreds of men, innocent or of unproved guilt, Jeffreys sent to their deaths in the pseudo trials that followed Monmouth’s feeble and stupid attempt to seize the throne. When the ordeal ended, scores had been executed and 1,260 were awaiting the hangman in three counties. To be absent from home during the uprising was evidence of guilt. Mere death was considered much too mild for the villagers and farmers rounded up in these raids. The directions to a high sheriff were to provide an ax, a cleaver, 'a furnace or cauldron to boil their heads and quarters, and soil to boil therewith, half a bushel to each traitor, and tar to tar them with, and a sufficient number of spears and poles to fix their heads and quarters’ along the highways. One could have crossed a good part of northern England by their guidance.

“The story of The Bloody Assizes, widely known to Americans, helped to place constitutional limitations on the crime of treason and to produce a bar against cruel and unusual punishments. But in the polemics that led to the various guarantees of freedom, it had no place compared with the tremendous thrust of the trial and execution of Sidney. The hundreds of judicial murders committed by Jeffreys and his fellow judges were totally inconceivable in a free American republic, but any American could imagine himself in Sidney’s place— executed for putting on paper, in his closet, words that later on came to express the basic principles of republican government. Unless barred by fundamental law, the legal rulings that permitted this *255result could easily be employed against any person whose political opinions challenged the party in power.” The Bill of Rights 154-155 (1965).

Those who wrote the Eighth Amendment knew what price their forebears had paid for a system based, not on equal justice, but on discrimination. In those days the target was not the blacks or the poor, but the dissenters, those who opposed absolutism in government, who struggled for a parliamentary regime, and who opposed governments’ recurring efforts to foist a particular religion on the people. Id., at 155-163. But the tool of capital punishment was used with vengeance against the opposition and those unpopular with the regime. One cannot read this history without realizing that the desire for equality was reflected in the ban against “cruel and unusual punishments” contained in the Eighth Amendment.

In a Nation committed to equal protection of the laws there is no permissible “caste” aspect18 of law enforcement. Yet we know that the discretion of judges and juries in imposing the death penalty enables the penalty to be selectively applied, feeding prejudices against the accused if he is poor and despised, and lacking political clout, or if he is a member of a suspect or unpopular minority, and saving those who by social position may be in a more protected position. In ancient Hindu law a Brahman was exempt from capital punishment,19 and under that law, “[g] enerally, in the law books, punishment increased in severity as social status diminished.” 20 We have, I fear, taken in practice the same position, partially as a result of making the death pen*256alty discretionary and partially as a result of the ability of the rich to purchase the services of the most respected and most resourceful legal talent in the Nation.

The high service rendered by the “cruel and unusual” punishment clause of the Eighth Amendment is to require legislatures to write penal laws that are evenhanded, nonselective, and nonarbitrary, and to require judges to see to it that general laws are not applied sparsely, selectively, and spottily to unpopular groups.

A law that stated that anyone making more than $50,000 would be exempt from the death penalty would plainly fall, as would a law that in terms said that blacks, those who never went beyond the fifth grade in school, those who made less than $3,000 a year, or those who were unpopular or unstable should be the only people executed. A law which in the overall view reaches that result in practice 21 has no more sanctity than a law which in terms provides the same.

Thus, these discretionary statutes are unconstitutional *257in their operation. They are pregnant with discrimination and discrimination is an ingredient not compatible with the idea of equal protection of the laws that is implicit in the ban on “cruel and unusual” punishments.

Any law which is nondiscriminatory on its face may be applied in such a way as to violate the Equal Protection Clause of the Fourteenth Amendment. Yick Wo v. Hopkins, 118 U. S. 356. Such conceivably might be the fate of a mandatory death penalty, where equal or lesser sentences were imposed on the elite, a harsher one on the minorities or members of the lower castes. Whether a mandatory death penalty would otherwise be constitutional is a question I do not reach.

I concur in the judgments of the Court.

Mr. Justice Brennan,

concurring.

The question presented in these cases is whether death is today a punishment for crime that is “cruel and unusual” and consequently, by virtue of the Eighth and Fourteenth Amendments, beyond the power of the State to inflict.1

*258Almost a century ago, this Court observed that “[d]ifficulty would attend the effort to define with exactness the extent of the constitutional provision which provides that cruel and unusual punishments shall not be inflicted.” Wilkerson v. Utah, 99 U. S. 130, 135-136 (1879). Less than 15 years ago, it was again noted that “[t]he exact scope of the constitutional phrase 'cruel and unusual’ has not been detailed by this Court.” Trop v. Dulles, 356 U. S. 86, 99 (1958). Those statements remain true today. The Cruel and Unusual Punishments Clause, like the other great clauses of the Constitution, is not susceptible of precise definition. Yet we know that the values and ideals it embodies are basic to our scheme of government. And we know also that the Clause imposes upon this Court the duty, when the issue is properly presented, to determine the constitutional validity of a challenged punishment, whatever that punishment may be. In these cases, “[t]hat issue confronts us, and the task of resolving it is inescapably ours.” Id., at 103.

I

We have very little evidence of the Framers’ intent in including the Cruel and Unusual Punishments Clause among those restraints upon the new Government enumerated in the Bill of Rights. The absence of such a restraint from the body of the Constitution was alluded to, so far as we now know, in the debates of only two of the state ratifying conventions. In the Massachusetts convention, Mr. Holmes protested:

“What gives an additional glare of horror to these gloomy circumstances is the consideration, that Congress have to ascertain, point out, and deter*259mine, what kind of punishments shall be inflicted on persons convicted of crimes. They are nowhere restrained from inventing the most cruel and unheard-of punishments, and annexing them to crimes; and there is no constitutional check on them, but that racks and gibbets may be amongst the most mild instruments of their discipline.” 2 J. Elliot’s Debates 111 (2d ed. 1876).

Holmes’ fear that Congress would have unlimited power to prescribe punishments for crimes was echoed by Patrick Henry at the Virginia convention:

“. . . Congress, from their general powers, may fully go into business of human legislation. They may legislate, in criminal cases, from treason to the lowest offence — petty larceny. They may define crimes and prescribe punishments. In the definition of crimes, I trust they will be directed by what wise representatives ought to be governed by. But when we come to punishments, no latitude ought to be left, nor dependence put on the virtue of representatives. What says our [Virginia] bill of rights?— 'that excessive bail ought not to be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.’ Are you not, therefore, now calling on those gentlemen who are to compose Congress, to ... define punishments without this control? Will they find sentiments there similar to this bill of rights? You let them loose; you do more — you depart from the genius of your country. . . .

“In this business of legislation, your members of Congress will loose the restriction of not imposing excessive fines, demanding excessive bail, and inflicting cruel and unusual punishments. These are prohibited by your [Virginia] declaration of rights. What has distinguished our ancestors?— *260That they would not admit of tortures, or cruel and barbarous punishment.” 3 id., at 447.2

These two statements shed some light on what the Framers meant by “cruel and unusual punishments.” Holmes referred to “the most cruel and unheard-of punishments,” Henry to “tortures, or cruel and barbarous punishment.” It does not follow, however, that the Framers were exclusively concerned with prohibiting torturous punishments. Holmes and Henry were objecting to the absence of a Bill of Rights, and they cited to support their objections the unrestrained legislative power to prescribe punishments for crimes. Certainly we may suppose that they invoked the specter of the most drastic punishments a legislature might devise.

In addition, it is quite clear that Holmes and Henry focused wholly upon the necessity to restrain the legislative power. Because they recognized “that Congress have to ascertain, point out, and determine, what kinds of punishments shall be inflicted on persons convicted of crimes,” they insisted that Congress must be limited in its power to punish. Accordingly, they *261called for a “constitutional check” that would ensure that “when we come to punishments, no latitude ought to be left, nor dependence put on the virtue of representatives.” 3

The only further evidence of the Framers’ intent appears from the debates in the First Congress on the adoption of the Bill of Rights.4 As the Court noted in Weems v. United States, 217 U. S. 349, 368 (1910), *262the Cruel and Unusual Punishments Clause “received very little debate.” The extent of the discussion, by two opponents of the Clause in the House of Representatives, was this:

“Mr. Smith, of South Carolina, objected to the words 'nor cruel and unusual punishments;’ the import of them being too indefinite.

“Mr. Livermore. — The [Eighth Amendment] seems to express a great deal of humanity, on which account I have no objection to it; but as it seems to have no meaning in it, I do not think it necessary. ... No cruel and unusual punishment is to be inflicted; it is sometimes necessary to hang a man, villains often deserve whipping, and perhaps having their ears cut off; but are we in future to be prevented from inflicting these punishments because they are cruel? If a more lenient mode of correcting vice and deterring others from the commission of it could be invented, it would be very prudent in the Legislature to adopt it; but until we have some security that this will be done, we ought not to be restrained from making necessary laws by any declaration of this kind.

“The question was put on the [Eighth Amendment], and it was agreed to by a considerable majority.” 1 Annals of Cong. 754 (1789).5

Livermore thus agreed with Holmes and Henry that the Cruel and Unusual Punishments Clause imposed a limitation upon the legislative power to prescribe pun*263ishments. However, in contrast to Holmes and Henry, who were supporting the Clause, Livermore, opposing it, did not refer to punishments that were considered barbarous and torturous. Instead, he objected that the Clause might someday prevent the legislature from inflicting what were then quite common and, in his view, “necessary” punishments — death, whipping, and earcrop-ping.6 The only inference to be drawn from Liver-more’s statement is that the “considerable majority” was prepared to run that risk. No member of the House rose to reply that the Clause was intended merely to prohibit torture.

Several conclusions thus emerge from the history of the adoption of the Clause. We know that the Framers’ concern was directed specifically at the exercise of legislative power. They included in the Bill of Rights a prohibition upon “cruel and unusual punishments” precisely because the legislature would otherwise have had the unfettered power to prescribe punishments for crimes. Yet we cannot now know exactly what the Framers thought “cruel and unusual punishments” were. Certainly they intended to ban torturous punishments, but the available evidence does not support the further conclusion that only torturous punishments were to be outlawed. As Livermore’s comments demonstrate, the Framers were well aware that the reach of the Clause was not limited to the proscription of unspeakable atrocities. Nor did they intend simply to forbid punishments considered “cruel and unusual” at the time. The “import” of the Clause is, indeed, “indefinite,” and for good reason. A constitutional provision “is enacted, it is true, from an experience of evils, but its general lan*264guage should not, therefore, be necessarily confined to the form that evil had theretofore taken. Time works changes, brings into existence new conditions and purposes. Therefore a principle to be vital must be capable of wider application than the mischief which gave it birth.” Weems v. United States, 217 U. S., at 373.

It was almost 80 years before this Court had occasion to refer to the Clause. See Pervear v. The Commonwealth, 5 Wall. 475, 479-480 (1867). These early cases, as the Court pointed out in Weems v. United States, supra, at 369, did not undertake to provide “an exhaustive definition” of “cruel and unusual punishments.” Most of them proceeded primarily by “looking backwards for examples by which to fix the meaning of the clause,” id., at 377, concluding simply that a punishment would be “cruel and unusual” if it were similar to punishments considered “cruel and unusual” at the time the Bill of Rights was adopted.7 In Wilkerson v. Utah, 99 U. S., at 136, for instance, the Court found it “safe to affirm that punishments of torture . . . and all others in the same line of unnecessary cruelty, are forbidden.” The “punishments of torture,” which the Court labeled “atrocities,” were cases where the criminal “was embowelled alive, beheaded, and quartered,” and cases “of public dissection . . . and burning alive.” Id., at 135. Similarly, in In re Kemm-*265ler, 136 U. S. 436, 446 (1890), the Court declared that “if the punishment prescribed for an offence against the laws of the State were manifestly cruel and unusual, as burning at the stake, crucifixion, breaking on the wheel, or the like, it would be the duty of the courts to adjudge such penalties to be within the constitutional prohibition.” The Court then observed, commenting upon the passage just quoted from Wilkerson v. Utah, supra, and applying the “manifestly cruel and unusual” test, that “[pjunishments are cruel when they involve torture or a lingering death; but the punishment of death is not cruel, within the meaning of that word as used in the Constitution. It implies there something inhuman and barbarous, something more than the mere extinguishment of life.” 136 U. S., at 447.

Had this “historical” interpretation of the Cruel and Unusual Punishments Clause prevailed, the Clause would have been effectively read out of the Bill of Rights. As the Court noted in Weems v. United States, supra, at 371, this interpretation led Story to conclude “that the provision ‘would seem to be wholly unnecessary in a free government, since it is scarcely possible that any department of such a government should authorize or justify such atrocious conduct.’ ” And Cooley in his book, Constitutional Limitations, said the Court, “apparently in a struggle between the effect to be given to ancient examples and the inconsequence of a dread of them in these enlightened times, . . . hesitate [d] to advance definite views.” Id., at 375. The result of a judicial application of this interpretation was not surprising. A state court, for example, upheld the constitutionality of the whipping post: “In comparison with the ‘barbarities of quartering, hanging in chains, castration, etc.,’ it was easily reduced to insignificance.” Id., at 377.

*266But this Court in Weems decisively repudiated the “historical” interpretation of the Clause. The Court, returning to the intention of the Framers, “rel[ied] on the conditions which existed when the Constitution was adopted.” And the Framers knew “that government by the people instituted by the Constitution would not imitate the conduct of arbitrary monarchs. The abuse of power might, indeed, be- apprehended, but not that it would be manifested in provisions or practices which would shock the sensibilities of men.” Id., at 375. The Clause, then, guards against “[t]he abuse of power”; contrary to the implications in Wilkerson v. Utah, supra, and In re Kemmler, supra, the prohibition of the Clause is not “confine[d] ... to such penalties and punishment as were inflicted by the Stuarts.” 217 U. S., at 372. Although opponents of the Bill of Rights “felt sure that the spirit of liberty could be trusted, and that its ideals would be represented, not debased, by legislation,” ibid., the Framers disagreed:

“[Patrick] Henry and those who believed as he did would take no chances. Their predominant political impulse was distrust of power, and they insisted on constitutional limitations against its abuse. But surely they intended more than to register a fear of the forms of abuse that went out of practice with the Stuarts. Surely, their [jealousy] of power had a saner justification than that. They were men of action, practical and sagacious, not beset with vain imagining, and it must have come to them that there could be exercises of cruelty by laws other than those which inflicted bodily pain or mutilation. With power in a legislature great, if not unlimited, to give criminal character to the actions of men, with power unlimited to fix terms of imprisonment with what accompaniments they *267might, what more potent instrument of cruelty could be put into the hands of power? And it was believed that power might be tempted to cruelty. This was the motive of the clause, and if we are to attribute an intelligent providence to its advocates we cannot think that it was intended to prohibit only practices like the [Stuarts’,] or to prevent only an exact repetition of history. We cannot think that the possibility of a coercive cruelty being exercised through other forms of punishment was overlooked.” Id., at 372-373.

The Court in Weems thus recognized that this “restraint upon legislatures” possesses an “expansive and vital character” that is “ ‘essential ... to the rule of law and the maintenance of individual freedom.’ ” Id., at 376-377. Accordingly, the responsibility lies with the courts to make certain that the prohibition of the Clause is enforced.8 Referring to cases in which “prominence [was] given to the power of the legislature to define crimes and their punishment,” the Court said:

“We concede the power in most of its exercises. We disclaim the right to assert a judgment *268against that of the legislature of the expediency of the laws or the right to oppose the judicial power to the legislative power to define crimes and fix their punishment, unless that power encounters in its exercise a constitutional prohibition. In such case not our discretion but our legal duty, strictly defined and imperative in its direction, is invoked.” Id., at 378.9

In short, this Court finally adopted the Framers’ view of the Clause as a “constitutional check” to ensure that “when we come to punishments, no latitude ought to be left, nor dependence put on the virtue of representatives.” That, indeed, is the only view consonant with our constitutional form of government. If the judicial conclusion that a punishment is “cruel and unusual” “depend [ed] upon virtually unanimous condemnation of the penalty at issue,” then, “[l]ike no other constitutional provision, [the Clause’s] only function would be to legitimize advances already made by the other departments and opinions already the conventional wisdom.” We know that the Framers did not envision “so narrow a role for this basic guaranty of human rights.” Goldberg & Dershowitz, Declaring the Death Penalty Unconstitutional, 83 Harv. L. Rev. 1773, 1782 (1970). The right to be free of cruel and unusual punishments, like the other guarantees of the Bill of Rights, “may not be submitted to vote; [it] depend [s] on the outcome of no elections.” “The very purpose of a Bill of Rights was to withdraw certain subjects from the vicissitudes of political controversy, to place them beyond the reach of majorities and officials and to establish them as legal principles to be applied *269by the courts." Board of Education v. Barnette, 319 U. S. 624, 638 (1943).

Judicial enforcement of the Clause, then, cannot be evaded by invoking the obvious truth that legislatures have the power to prescribe punishments for crimes. That is precisely the reason the Clause appears in the Bill of Rights. The difficulty arises, rather, in formulating the “legal principles to be applied by the courts" when a legislatively prescribed punishment is challenged as “cruel and unusual." In formulating those constitutional principles, we must avoid the insertion of “judicial conception [s] of . . . wisdom or propriety," Weems v. United States, 217 U. S., at 379, yet we must not, in the guise of “judicial restraint,” abdicate our fundamental responsibility to enforce the Bill of Rights. Were we to do so, the “constitution would indeed be as easy of application as it would be deficient in efficacy and power. Its general principles would have little value and be converted by precedent into impotent and lifeless formulas. Rights declared in words might be lost in reality.” Id., at 373. The Cruel and Unusual Punishments Clause would become, in short, “little more than good advice.” Trop v. Dulles, 366 U. S., at 104.

II

Ours would indeed be a simple task were we required merely to measure a challenged punishment against those that history has long condemned. That narrow and unwarranted view of the Clause, however, was left behind with the 19th century. Our task today is more complex. We know “that the words of the [Clause] are not precise, and that their scope is not static." We know, therefore, that the Clause “must draw its meaning from the evolving standards of decency that mark the prog*270ress of a maturing society.” Id., at 100-101.10 That knowledge, of course, is but the beginning of the inquiry.

In Trop v. Dulles, supra, at 99, it was said that “[t]he question is whether [a] penalty subjects the individual to a fate forbidden by the principle of civilized treatment guaranteed by the [Clause].” It was also said that a challenged punishment must be examined “in light of the basic prohibition against inhuman treatment” embodied in the Clause. Id., at 100 n. 32. It was said, finally, that:

“The basic concept underlying the [Clause] is nothing less than the dignity of man. While the State has the power to punish, the [Clause] stands to assure that this power be exercised within the limits of civilized standards.” Id., at 100.

At bottom, then, the Cruel and Unusual Punishments Clause prohibits the infliction of uncivilized and inhuman punishments. The State, even as it punishes, must treat its members with respect for their intrinsic worth as human beings. A punishment is “cruel and unusual,” therefore, if it does not comport with human dignity.

This formulation, of course, does not of itself yield principles for assessing the constitutional validity of particular punishments. Nevertheless, even though “[t]his Court has had little occasion to give precise content to the [Clause],” ibid., there are principles recognized in our cases and inherent in the Clause sufficient to permit a judicial determination whether a challenged punishment comports with human dignity.

*271The primary principle is that a punishment must not be so severe as to be degrading to the dignity of human beings. Pain, certainly, may be a factor in the judgment. The infliction of an extremely severe punishment will often entail physical suffering. See Weems v. United States, 217 U. S., at 366.11 Yet the Framers also knew “that there could be exercises of cruelty by laws other than those which inflicted bodily pain or mutilation.” Id., at 372. Even though “[t]here may be involved no physical mistreatment, no primitive torture,” Trop v. Dulles, supra, at 101, severe mental pain may be inherent in the infliction of a particular punishment. See Weems v. United States, supra, at 366.12 That, indeed, was one of the conclusions underlying the holding of the plurality in Trop v. Dulles that the punishment of expatriation violates the Clause.13 And the *272physical and mental suffering inherent in the punishment of cadena temporal, see nn. 11-12, supra, was an obvious basis for the Court’s decision in Weems v. United States that the punishment was “cruel and unusual.” 14

More than the presence of pain, however, is comprehended in the judgment that the extreme severity of a punishment makes it degrading to the dignity of human beings. The barbaric punishments condemned by history, “punishments which inflict torture, such as the rack, the thumbscrew, the iron boot, the stretching of limbs and the like,” are, of course, “attended with acute pain and suffering.” O’Neil v. Vermont, 144 U. S. 323, 339 (1892) (Field, J., dissenting). When we consider why they have been condemned, however, we realize that the pain involved is not the only reason. The true significance of these punishments is that they treat *273members of the human race as nonhumans, as objects to be toyed with and discarded. They are thus inconsistent with the fundamental premise of the Clause that even the vilest criminal remains a human being possessed of common human dignity.

The infliction of an extremely severe punishment, then, like the one before the Court in Weems v. United States, from which “[n]o circumstance of degradation [was] omitted,” 217 U. S., at 366, may reflect the attitude that the person punished is not entitled to recognition as a fellow human being. That attitude may be apparent apart from the severity of the punishment itself. In Louisiana ex rel. Francis v. Resweber, 329 U. S. 459, 464 (1947), for example, the unsuccessful electrocution, although it caused "mental anguish and physical pain,” was the result of “an unforeseeable accident.” Had the failure been intentional, however, the punishment would have been, like torture, so degrading and indecent as to amount to a refusal to accord the criminal human status. Indeed, a punishment may be degrading to human dignity solely because it is a punishment. A State may not punish a person for being “mentally ill, or a leper, or . . . afflicted with a venereal disease,” or for being addicted to narcotics. Robinson v. California, 370 U. S. 660, 666 (1962). To inflict punishment for having a disease is to treat the individual as a diseased thing rather than as a sick human being. That the punishment is not severe, “in the abstract,” is irrelevant; “[e]ven one day in prison would be a cruel and unusual punishment for the 'crime’ of having a common cold.” Id., at 667. Finally, of course, a punishment may be degrading simply by reason of its enormity. A prime example is expatriation, a “punishment more primitive than torture,” Trop v. Dulles, 356 U. S., at 101, for it necessarily involves a *274denial by society of the individual's existence as a member of the human community.15

In determining whether a punishment comports with human dignity, we are aided also by a second principle inherent in the Clause — that the State must not arbitrarily inflict a severe punishment. This principle derives from the notion that the State does not respect human dignity when, without reason, it inflicts upon some people a severe punishment that it does not inflict upon others. Indeed, the very words “cruel and unusual punishments” imply condemnation of the arbitrary infliction of severe punishments. And, as we now know, the English history of the Clause10 reveals a particular concern with the establishment of a safeguard against arbitrary punishments. See Granucci, “Nor Cruel and Unusual Punishments Inflicted:” The Original Meaning, 57 Calif. L. Rev. 839, 857-860 (1969).17

*275This principle has been recognized in our cases.18 In Wilkerson v. Utah, 99 U. S., at 133 — 134, the Court reviewed various treatises on military law in order to demonstrate that under “the custom of war” shooting was a common method of inflicting the punishment of death. On that basis, the Court concluded:

“Cruel and unusual punishments are forbidden by the Constitution, but the authorities referred to [treatises on military law] are quite sufficient to show that the punishment of shooting as a mode of executing the death penalty for the crime of murder in the first degree is not included in that *276category, within the meaning of the [Clause]. Soldiers convicted of desertion or other capital military offenses are in the great majority of cases sentenced to be shot, and the ceremony for such occasions is given in great fulness by the writers upon the subject of courts-martial.” Id., at 13A-135.

The Court thus upheld death by shooting, so far as appears, solely on the ground that it was a common method of execution.19

As Wilkerson v. Utah suggests, when a severe punishment is inflicted "in the great majority of cases” in which it is legally available, there is little likelihood that the State is inflicting it arbitrarily. If, however, the infliction of a severe punishment is “something different from that which is generally done” in such cases, Trop v. Dulles, 356 U. S., at 101 n. 32,20 there is a sub*277stantial likelihood that the State, contrary to the requirements of regularity and fairness embodied in the Clause, is inflicting the punishment arbitrarily. This principle is especially important today. There is scant danger, given the political processes “in an enlightened democracy such as ours,” id., at 100, that extremely severe punishments will be widely applied. The more significant function of the Clause, therefore, is to protect against the danger of their arbitrary infliction.

A third principle inherent in the Clause is that a severe punishment must not be unacceptable to contemporary society. Rejection by society, of course, is a strong indication that a severe punishment does not comport with human dignity. In applying this principle, however, we must make certain that the judicial determination is as objective as possible.21 *278Thus, for example, Weems v. United States, 217 U. S., at 380, and Trop v. Dulles, 356 U. S., at 102-103, suggest that one factor that may be considered is the existence of the punishment in jurisdictions other than those before the Court. Wilkerson v. Utah, supra, suggests that another factor to be considered is the historic usage of the punishment.22 Trop v. Dulles, supra, at 99, combined present acceptance with past usage by observing that “the death penalty has been employed throughout our history, and, in a day when it is still widely accepted, it cannot be said to violate the constitutional concept of cruelty.” In Robinson v. California, 370 U. S., at 666, which involved the infliction of punishment for narcotics addiction, the Court went a step further, concluding simply that “in the light of contemporary human knowledge, a law which made a criminal offense of such a disease would doubtless be universally thought to be an infliction of cruel and unusual punishment.”

The question under this principle, then, is whether there are objective indicators from which a court can conclude that contemporary society considers a severe punishment unacceptable. Accordingly, the judicial *279task is to review the history of a challenged punishment and to examine society’s present practices with respect to its use. Legislative authorization, of course, does not establish acceptance. The acceptability of a severe punishment is measured, not by its availability, for it might become so offensive to society as never to be inflicted, but by its use.

The final principle inherent in the Clause is that a severe punishment must not be excessive. A punishment is excessive under this principle if it is unnecessary: The infliction of a severe punishment by the State cannot comport with human dignity when it is nothing more than the pointless infliction of suffering. If there is a significantly less severe punishment adequate to achieve the purposes for which the punishment is inflicted, cf. Robinson v. California, supra, at 666; id., at 677 (Douglas, J., concurring); Trop v. Dulles, supra, at 114 (Brennan, J., concurring), the punishment inflicted is unnecessary and therefore excessive.

This principle first appeared in our cases in Mr. Justice Field’s dissent in O’Neil v. Vermont, 144 U. S., at 337.23 He there took the position that:

“[The Clause] is directed, not only against punishments of the character mentioned [torturous punishments], but against all punishments which by *280their excessive length or severity are greatly dispro-portioned to the offences charged. The whole inhibition is against that which is excessive either in the bail required, or fine imposed, or punishment inflicted.” Id., at 339-340.

Although the determination that a severe punishment is excessive may be grounded in a judgment that it is disproportionate to the crime,24 the more significant basis is that the punishment serves no penal purpose more effectively than a less severe punishment. This view of the principle was explicitly recognized by the Court in Weems v. United States, supra. There the Court, reviewing a severe punishment inflicted for the falsification of an official record, found that “the highest punishment possible for a crime which may cause the loss of many thousand [s] of dollars, and to prevent which the duty of the State should be as eager as to prevent the perversion of truth in a public document, is not greater than that which may be imposed for falsifying a single item of a public account.” Id., at 381. Stating that “this contrast shows more than different exercises of legislative judgment,” the Court concluded that the punishment was unnecessarily severe in view of the purposes for which it was imposed. Ibid.25 *281See also Trop v. Dulles, 356 U. S., at 111-112 (Brennan, J., concurring).26

There are, then, four principles by which we may determine whether a particular punishment is “cruel and unusual.” The primary principle, which I believe supplies the essential predicate for the application of the others, is that a punishment must not by its severity be degrading to human dignity. The paradigm violation of this principle would be the infliction of a torturous punishment of the type that the Clause has always prohibited. Yet “ [i] t is unlikely that any State at this moment in history,” Robinson v. California, 370 U. S., at 666, would pass a law providing for the infliction of such a punishment. Indeed, no such punishment has ever been before this Court. The same may be said of the other principles. It is unlikely that this Court will confront a severe punishment that is obviously inflicted in wholly arbitrary fashion; no State would engage in a reign of blind terror. Nor is it likely that this Court will be called upon to review a severe punishment that is clearly and totally rejected throughout society; no legislature would be able even to authorize the infliction of such a punishment. Nor, finally, is it likely that this Court will have to consider a severe punishment that is patently unnecessary; no State today would inflict a severe punishment knowing that there was no reason whatever for doing so. In short, we are unlikely to have occasion to determine that a punishment is fatally offensive under any one principle.

*282Since the Bill of Rights was adopted, this Court has adjudged only three punishments to be within the prohibition of the Clause. See Weems v. United States, 217 U. S. 349 (1910) (12 years in chains at hard and painful labor); Trop v. Dulles, 356 U. S. 86 (1958) (expatriation) ; Robinson v. California, 370 U. S. 660 (1962) (imprisonment for narcotics addiction). Each punishment, of course, was degrading to human dignity, but of none could it be said conclusively that it wasl fatally offensive under one or the other of the principles. Rather, these “cruel and unusual punishments” seriously implicated several of the principles, and it was the application of the principles in combination that supported the judgment. That, indeed, is not surprising. The function of these principles, after all, is simply to provide means by which a court can determine whether a challenged punishment comports with human dignity. They are, therefore, interrelated, and in most cases it will be their convergence that will justify the conclusion that a punishment is “cruel and unusual.” The test, then, will ordinarily be a cumulative one: If a punishment is unusually severe, if there is a strong probability that it is inflicted arbitrarily, if it is substantially rejected by contemporary society, and if there is no reason to believe that it serves any penal purpose more effectively than some less severe punishment, then the continued infliction of that punishment violates the command of the Clause that the State may not inflict inhuman and uncivilized punishments upon those convicted of crimes.

Ill

The punishment challenged in these cases is death. Death, of course, is a “traditional” punishment, Trop v. Dulles, supra, at 100, one that “has been employed throughout our history,” id., at 99, and its constitu*283tional background is accordingly an appropriate subject of inquiry.

There is, first, a textual consideration raised by the Bill of Rights itself. The Fifth Amendment declares that if a particular crime is punishable by death, a person charged with that crime is entitled to certain procedural protections.27 We can thus infer that the Framers recognized the existence of what was then a common punishment. We cannot, however, make the further inference that they intended to exempt this particular punishment from the express prohibition of the Cruel and Unusual Punishments Clause.28 Nor is there any indication in the debates on the Clause that a special exception was to be made for death. If anything, the indication is to the contrary, for Livermore specifically mentioned death as a candidate for future proscription under the Clause. See supra, at 262. Finally, it does not advance analysis to insist that the Framers did not believe that adoption *284of the Bill of Rights would immediately prevent the infliction of the punishment of death; neither did they believe that it would immediately prevent the infliction of other corporal punishments that, although common at the time, see n. 6, supra, are now acknowledged to be impermissible.29

There is also the consideration that this Court has decided three cases involving constitutional challenges to particular methods of inflicting this punishment. In Wilkerson v. Utah, 99 U. S. 130 (1879), and In re Kemmler, 136 U. S. 436 (1890), the Court, expressing in both cases the since-rejected “historical” view of the Clause, see supra, at 264-265, approved death by shooting and death by electrocution. In Wilkerson, the Court concluded that shooting was a common method of execution, see supra, at 275-276;30 in Kemmler, the Court held that the Clause did not apply to the States, 136 U. S., at 447-449.31 *285In Louisiana ex rel. Francis v. Resweber, supra, the Court approved a second attempt at electrocution after the first had failed. It was said that “ [t] he Fourteenth [Amendment] would prohibit by its due process clause execution by a state in a cruel manner,” 329 U. S., at 463, but that the abortive attempt did not make the “subsequent execution any more cruel in the constitutional sense than any other execution,” id., at 464.32 These three decisions thus reveal that the Court, while ruling upon various methods of inflicting death, has assumed in the past that death was a constitutionally permissible punishment.33 Past assumptions, however, are not sufficient to limit the scope of our examination of this punishment today. The constitutionality of death itself under the Cruel and Unusual Punishments Clause is before this Court for the first time; we cannot avoid the question by recalling past cases that never directly considered it.

The question, then, is whether the deliberate infliction of death is today consistent with the command of the Clause that the State may not inflict punishments that do not comport with human dignity. I will analyze the punishment of death in terms of the principles *286set out above and the cumulative test to which they lead: It is a denial of human dignity for the State arbitrarily to subject a person to an unusually severe punishment that society has indicated it does not regard as acceptable, and that cannot be shown to serve any penal purpose more effectively than a significantly less drastic punishment. Under these principles and this test, death is today a “cruel and unusual” punishment.

Death is a unique punishment in the United States. In a society that so strongly affirms the sanctity of life, not surprisingly the common view is that death is the ultimate sanction. This natural human feeling appears all about us. There has been no national debate about punishment, in general or by imprisonment, comparable to the debate about the punishment of death. No other punishment has been so continuously restricted, see infra, at 296-298, nor has any State yet abolished prisons, as some have abolished this punishment. And those States that still inflict death reserve it for the most heinous crimes. Juries, of course, have always treated death cases differently, as have governors exercising their commutation powers. Criminal defendants are of the same view. “As all practicing lawyers know, who have defended persons charged with capital offenses, often the only goal possible is to avoid the death penalty.” Griffin v. Illinois, 351 U. S. 12, 28 (1956) (Burton and Minton, JJ., dissenting). Some legislatures have required particular procedures, such as two-stage trials and automatic appeals, applicable only in death cases. “It is the universal experience in the administration of criminal justice that those charged with capital offenses are granted special considerations.” Ibid. See Williams v. Florida, 399 U. S. 78, 103 (1970) (all States require juries of 12 in death cases). This Court, too, almost *287always treats death cases as a class apart.34 And the unfortunate effect of this punishment upon the functioning of the judicial process is well known; no other punishment has a similar effect.

The only explanation for the uniqueness of death is its extreme severity. Death is today an unusually severe punishment, unusual in its pain, in its finality, and in its enormity. No other existing punishment is comparable to death in terms of physical and mental suffering. Although our information is not conclusive, it appears that there is no method available that guarantees an immediate and painless death.35 Since the discon*288tinuance of flogging as a constitutionally permissible punishment, Jackson v. Bishop, 404 F. 2d 571 (CA8 1968), death remains as the only punishment that may involve the conscious infliction of physical pain. In addition, we know that mental pain is an inseparable part of our practice of punishing criminals by death, for the prospect of pending execution exacts a frightful toll during the inevitable long wait between the imposition of sentence and the actual infliction of death. Cf. Ex parte Medley, 134 U. S. 160, 172 (1890). As the California Supreme Court pointed out, “the process of carrying out a verdict of death is often so degrading and brutalizing to the human spirit as to constitute psychological torture.” People v. Anderson, 6 Cal. 3d 628, 649, 493 P. 2d 880, 894 (1972).36 Indeed, as Mr. Justice Frankfurter noted, “the onset of insanity while awaiting *289execution of a death sentence is not a rare phenomenon.” Solesbee v. Balkcom, 339 U. S. 9, 14 (1950) (dissenting opinion). The “fate of ever-increasing fear and distress” to which the expatriate is subjected, Trop v. Dulles, 356 U. S., at 102, can only exist to a greater degree for a person confined in prison awaiting death.37

The unusual severity of death is manifested most clearly in its finality and enormity. Death, in these respects, is in a class by itself. Expatriation, for example, is a punishment that “destroys for the individual the political existence that was centuries in the development,” that “strips the citizen of his status in the national and international political community,” and that puts “[h]is very existence” in jeopardy. Expatriation thus inherently entails “the total destruction of the individual’s status in organized society.” Id., at 101. “In short, the expatriate has lost the right to have rights.” Id., at 102. Yet, demonstrably, expatriation is not “a fate worse than death.” Id., at 125 (Frankfurter, J., dissenting).38 Although death, like expatriation, destroys the *290individual’s “political existence” and his “status in organized society,” it does more, for, unlike expatriation, death also destroys “[h]is very existence.” There is, too, at least the possibility that the expatriate will in the future regain “the right to have rights.” Death forecloses even that possibility.

Death is truly an awesome punishment. The calculated killing of a human being by the State involves, by its very nature, a denial of the executed person’s humanity. The contrast with the plight of a person punished by imprisonment is evident. An individual in prison does not lose “the right to have rights.” A prisoner retains, for example, the constitutional rights to the free exercise of religion, to be free of cruel and unusual punishments, and to treatment as a “person” for purposes of due process of law and the equal protection of the laws. A prisoner remains a member of the human family. Moreover, he retains the right of access to the courts. His punishment is not irrevocable. Apart from the common charge, grounded upon the recognition of human fallibility, that the punishment of death must inevitably be inflicted upon innocent men, we know that death has been the lot of men whose convictions were unconstitutionally secured in view of later, retroactively applied, holdings of this Court. The punishment itself may have been unconstitutionally inflicted, see Witherspoon v. Illinois, 391 U. S. 510 (1968), yet the finality of death precludes relief. An executed person has indeed “lost the right to have rights.” As one 19th century proponent of punishing criminals by death declared, “When a man is hung, there is an end of our relations with him. His execution is a way of saying, ‘You are not fit for this world, take your chance elsewhere.’ ”39

*291In comparison to all other punishments today, then, the deliberate extinguishment of human life by the State is uniquely degrading to human dignity. I would not hesitate to hold, on that ground alone, that death is today a “cruel and unusual” punishment, were it not that death is a punishment of longstanding usage and acceptance in this country. I therefore turn to the second principle — that the State may not arbitrarily inflict an unusually severe punishment.

The outstanding characteristic of our present practice of punishing criminals by death is the infrequency with which we resort to it. The evidence is conclusive that death is not the ordinary punishment for any crime.

There has been a steady decline in the infliction of this punishment in every decade since the 1930’s, the earliest period for which accurate statistics are available. In the 1930’s, executions averaged 167 per year; in the 1940’s, the average was 128; in the 1950’s, it was 72; and in the years 1960-1962, it was 48. There have been a total of. 46 executions since then, 36 of them in 1963-1964.40 Yet our population and the number of capital crimes committed have increased greatly over the past four decades. The contemporary rarity of the infliction of this punishment is thus the end result of a long-continued decline. That rarity is plainly revealed by an examination of the years 1961-1970, the last 10-year period for which statistics are available. During that time, an average of 106 death sentences *292was imposed each year.41 Not nearly that number, however, could be carried out, for many were precluded by commutations to life or a term of years,42 transfers to mental institutions because of insanity,43 resentences to life or a term of years, grants of new trials and orders for resentencing, dismissals of indictments and reversals of convictions, and deaths by suicide and natural causes.44 On January 1, 1961, the death row population was 219; on December 31, 1970, it was 608; during that span, there were 135 executions.45 Consequently, had the 389 additions to death row also been executed, the annual average would have been 52.46 In short, the country *293might, at most, have executed one criminal each week. In fact, of course, far fewer were executed. Even before the moratorium on executions began in 1967, executions totaled only 42 in 1961 and 47 in 1962, an average of less than one per week; the number dwindled to 21 in 1963, to 15 in 1964, and to seven in 1965; in 1966, there was one execution, and in 1967, there were two.47

When a country of over 200 million people inflicts an unusually severe punishment no more than 50 times a year, the inference is strong that the punishment is hot being regularly and fairly applied. To dispel it would indeed require a clear showing of nonarbitrary infliction.

Although there are no exact figures available, we know that thousands of murders and rapes are committed annually in States where death is an authorized punishment for those crimes. However the rate of infliction is characterized — as “freakishly” or “spectacularly” rare, or simply as rare — it would take the purest sophistry to deny that death is inflicted in only a minute fraction of these cases. How much rarer, after all, could the infliction of death be?

When the punishment of death is inflicted in a trivial number of the cases in which it is legally available, the conclusion is virtually inescapable that it is being inr flicted arbitrarily. Indeed, it smacks of little more than a lottery system. The States claim, however, that this rarity is evidence not of arbitrariness, but of informed selectivity: Death is inflicted, they say, only in “extreme” cases.

Informed selectivity, of course, is a value not to be denigrated. Yet presumably the States could make precisely the same claim if there were 10 executions per *294year, or five, or even if tHere were but one. That there may be as many as 50 per year does not strengthen the claim. When, the rate of infliction is at this low level, it is highly implausible that only the worst criminals or the criminals who commit the worst crimes are selected for this punishment. No one has yet suggested a rational basis that could differentiate in those terms the few who die from the many who go to prison. Crimes and criminals simply do not admit of a distinction that can be drawn so finely as to explain, on that ground, the execution of such a tiny sample of those eligible. Certainly the laws that provide for this punishment do not attempt to draw that distinction; all cases to which the laws apply are necessarily “extreme.” Nor is the distinction credible in fact. If, for example, petitioner Furman or his crime illustrates the “extreme,” then nearly all murderers and their murders are also “extreme.” 48 Furthermore, our procedures in death cases, *295rather than resulting in the selection of “extreme” cases for this punishment, actually sanction an arbitrary selection. For this Court has held that juries may, as they do, make the decision whether to impose a death sentence wholly unguided by standards governing that decision. McGautha v. California, 402 U. S. 183, 196—208 (1971). In other words, our procedures are not constructed to guard against the totally capricious selection of criminals for the punishment of death.

Although it is difficult to imagine what further facts would be necessary in order to prove that death is, as my Brother Stewart puts it, “wantonly and . . . freakishly” inflicted, I need not conclude that arbitrary infliction is patently obvious. I am not considering this punishment by the isolated light of one principle. The probability of arbitrariness is sufficiently substantial that it can be relied upon, in combination with the other principles, in reaching a judgment on the constitutionality of this punishment.

When there is a strong probability that an unusually severe and degrading punishment is being inflicted arbitrarily, we may well expect that society will disapprove of its infliction. I turn, therefore, to the third principle. An examination of the history and present operation of the American practice of punishing criminals by death reveals that this punishment has been almost totally rejected by contemporary society.

I cannot add to my Brother Marshall’s comprehensive treatment of the English and American history of *296this punishment. I emphasize, however, one significant conclusion that emerges from that history. From the beginning of our Nation, the punishment of death has stirred acute public controversy. Although pragmatic arguments for and against the punishment have been frequently advanced, this longstanding and heated controversy cannot be explained solely as the result of differences over the practical wisdom of a particular government policy. At bottom, the battle has been waged on moral grounds. The country has debated whether a society for which the dignity of the individual is the supreme value can, without a fundamental inconsistency, follow the practice of deliberately putting some of its members to death. In the United States, as in other nations of the western world, “the struggle about this punishment has been .one between ancient and deeply rooted beliefs in retribution, atonement or vengeance on the one hand, and, on the other, beliefs in the personal value and dignity of the common man that were born of the democratic movement of the eighteenth century, as well as beliefs in the scientific approach to an understanding of the motive forces of human conduct, which are the result of the growth of the sciences of behavior during the nineteenth and twentieth centuries.” 49 It is this essentially moral conflict that forms the backdrop for the past changes in and the present operation of our system of imposing death as a punishment for crime.

Our practice of punishing criminals by death has changed greatly over the years. One significant change has been in our methods of inflicting death. Although this country never embraced the more violent and repulsive methods employed in England, we did for a long time rely almost exclusively upon the gallows and the firing squad. Since the development of the supposedly *297more humane methods of electrocution late in the 19th century and lethal gas in the 20th, however, hanging and shooting have virtually ceased.50 Our concern for decency and human dignity, moreover, has compelled changes in the circumstances surrounding the execution itself. No longer does our society countenance the spectacle of public executions, once thought desirable as a deterrent to criminal behavior by others. Today we reject public executions as debasing and brutalizing to us all.

Also significant is the drastic decrease in the crimes for which the punishment of death is actually inflicted. While esoteric capital crimes remain on the books, since 1930 murder and rape have accounted for nearly 99% of the total executions, and murder alone for about 87%.51 In addition, the crime of capital murder has itself been limited. As the Court noted in McGautha v. California, 402 U. S., at 198, there was in this country a “rebellion against the common-law rule imposing a mandatory death sentence on all convicted murderers.” Initially, that rebellion resulted in legislative definitions that distinguished between degrees of murder, retaining the mandatory death sentence only for murder in the first degree. Yet “[t]h'is new legislative criterion for isolating crimes appropriately punishable by death soon proved as unsuccessful as the concept of ‘malice aforethought,’ ” ibid., the common-law means of separating murder from manslaughter. Not only was the distinction between degrees of murder confusing and uncertain in practice, but even in clear cases of first-degree murder juries continued to take the law into *298their own hands: if they felt that death was an inappropriate punishment, “they simply refused to convict of the capital offense.” Id., at 199. The phenomenon of jury nullification thus remained to counteract the rigors of mandatory death sentences. Bowing to reality, “legislatures did not try, as before, to refine further the definition of capital homicides. Instead they adopted the method of forthrightly granting juries the discretion which they had been exercising in fact.” Ibid. In consequence, virtually all death sentences today are discre-tionarily imposed. Finally, it is significant that nine States no longer inflict the punishment of death under any circumstances,52 and five others have restricted it to extremely rare crimes.53

*299Thus, although “the death penalty has been employed throughout our history,” Trop v. Dulles, 356 U. S., at 99, in fact the history of this punishment is one of successive restriction. What was once a common punishment has become, in the context of a continuing moral debate, increasingly rare. The evolution of this punishment evidences, not that it is an inevitable part of the American scene, but that it has proved progressively more troublesome to the national conscience. The result of this movement is our current system of administering the punishment, under which death sentences are rarely imposed and death is even more rarely inflicted. It is, of course, “We, the People” who are responsible for the rarity both of the imposition and the carrying out of this punishment. Juries, “express [ing] the conscience of the community on the ultimate question of life or death,” Witherspoon v. Illinois, 391 U. S., at 519, have been able to bring themselves to vote for death in a mere 100 or so cases among the thousands tried each year where the punishment is available. Governors, elected by and acting for us, have regularly commuted a substantial number of those sentences. And it is our society that insists upon due process of law to the end that no person will be unjustly put to death, thus ensuring that many more of those sentences will not be carried out. In sum, we have made death a rare punishment today.

The progressive decline in, and the current rarity of, the infliction of death demonstrate that our society seriously questions the appropriateness of this punishment today. The States point out that many legislatures authorize death as the punishment for certain crimes and that substantial segments of the public, as reflected in opinion polls and referendum votes, continue to support it. Yet the availability of this punishment through statutory authorization, as well as the polls and refer-*300enda, which amount simply to approval of that authorization, simply underscores the extent to which our society has in fact rejected this punishment. When an unusually severe punishment is authorized for wide-scale application but not, because of society’s refusal, inflicted save in a few instances, the inference is compelling that there is a deep-seated reluctance to inflict it. Indeed, the likelihood is great that the punishment is tolerated only because of its disuse. The objective indicator of society’s view of an unusually severe punishment is what society does with it, and today society will inflict death upon only a small sample of the eligible criminals. Rejection could hardly be more complete without becoming absolute. At the very least, I must conclude that contemporary society views this punishment with substantial doubt.

The final principle to be considered is that an unusually severe and degrading punishment may not be excessive in view of the purposes for which it is inflicted. This principle, too, is related to the others. When there is a strong probability that the State is arbitrarily inflicting an unusually severe punishment that is subject to grave societal doubts, it is likely also that the punishment cannot be shown to be serving any penal purpose that could not be served equally well by some less severe punishment.

The States’ primary claim is that death is a necessary punishment because it prevents the commission of capital crimes more effectively than any less severe punishment. The first part of this claim is that the infliction of death is necessary to stop the individuals executed from committing further crimes. The sufficient answer to this is that if a criminal convicted of a capital crime poses a danger to society, effective administration of the State’s pardon and parole laws can delay or deny his release from prison, and techniques of isolation can elim*301inate or minimize the danger while he remains confined.

The more significant argument is that the threat of death prevents the commission of capital crimes because it deters potential criminals who would not be deterred by the threat of imprisonment. The argument is not based upon evidence that the threat of death is a superior deterrent. Indeed, as my Brother Marshall establishes, the available evidence uniformly indicates, although it does not conclusively prove, that the threat of death has no greater deterrent effect than the threat of imprisonment. The States argue, however, that they are entitled to rely upon common human experience, and that experience, they say, supports the conclusion that death must be a more effective deterrent than any less severe punishment. Because people fear death the most, the argument runs, the threat of death must be the greatest deterrent.

It is important to focus upon the precise import of this argument. It is not denied that many, and probably most, capital crimes cannot be deterred by the threat of punishment. Thus the argument can apply only to those who think rationally about the commission of capital crimes. Particularly is that true when the potential criminal, under this argument, must not only consider the risk of punishment, but also distinguish between two possible punishments. The concern, then, is with a particular type of potential criminal, the rational person who will commit a capital crime knowing that the punishment is long-term imprisonment, which may well be for the rest of his life, but will not commit the crime knowing that the punishment is death. On the face of it, the assumption that such persons exist is implausible.

In any event, this argument cannot be appraised in the abstract. We are not presented with the theoretical question whether under any imaginable circumstances the *302threat of death might be a greater deterrent to the commission of capital crimes than the threat of imprisonment. We are concerned with the practice of punishing criminals by death as it exists in the United States today. Proponents of this argument necessarily admit that its validity depends upon the existence of a system in which the punishment of death is invariably and swiftly imposed. Our system, of course, satisfies neither condition. A rational person contemplating a murder or rape is confronted, not with the certainty of a speedy death, but with the slightest possibility that he will be executed in the distant future. The risk of death is remote and improbable; in contrast, the risk of long-term imprisonment is near and great. In short, whatever the speculative validity of the assumption that the threat of death is a superior deterrent, there is no reason to believe that as currently administered the punishment of death is necessary to deter the commission of capital crimes. Whatever might be the case were all or substantially all eligible criminals quickly put to death, unverifiable possibilities are an insufficient basis upon which to conclude that the threat of death today has any greater deterrent efficacy than the threat of imprisonment.54

*303There is, however, another aspect to the argument that the punishment of death is necessary for the protection of society. The infliction of death, the States urge, serves to manifest the community’s outrage at the commission of the crime. It is, they say, a concrete public expression of moral indignation that inculcates respect for the law and helps assure a more peaceful community. Moreover, we are told, not only does the punishment of death exert this widespread moralizing influence upon community values, it also satisfies the popular demand for grievous condemnation of abhorrent crimes and thus prevents disorder, lynching, and attempts by private citizens to take the law into their own hands.

The question, however, is not whether death serves these supposed purposes of punishment, but whether death serves them more effectively than imprisonment. There is no evidence whatever that utilization of imprisonment rather than death encourages private blood feuds and other disorders. Surely if there were such a danger, the execution of a handful of criminals each year would not prevent it. The assertion that death alone is a sufficiently emphatic denunciation for capital crimes suffers from the same defect. If capital crimes require the punishment of death in order to provide moral reinforcement for the basic values of the community, those values can only be undermined when death is so rarely inflicted upon the criminals who commit the crimes. Furthermore, it is certainly doubtful that the infliction of death by the State does in fact strengthen the community’s moral code; if the deliberate extin-guishment of human life has any effect at all, it more likely tends to lower our respect for life and brutalize our values. That, after all, is why we no longer carry out public executions. In any event, this claim simply means that one purpose of punishment is to indicate social disapproval of crime. To serve that purpose our *304laws distribute punishments according to the gravity of crimes and punish more severely the crimes society regards as more serious. That purpose cannot justify any particular punishment as the upper limit of severity.

There is, then, no substantial reason to believe that the punishment of death, as currently administered, is necessary for the protection of society. The only other purpose suggested, one that is independent of protection for society, is retribution. Shortly stated, retribution in this context means that criminals are put to death because they deserve it.

Although it is difficult to believe that any State today wishes to proclaim adherence to “naked vengeance,” Trop v. Dulles, 356 U. S., at 112 (Brennan, J., concurring), the States claim, in reliance upon its statutory authorization, that death is the only fit punishment for capital crimes and that this retributive purpose justifies its infliction. In the past, judged by its statutory authorization, death was considered the only fit punishment for the crime of forgery, for the first federal criminal statute provided a mandatory death penalty for that crime. Act of April 30, 1790, § 14, 1 Stat. 115. Obviously, concepts of justice change; no immutable moral order requires death for murderers and rapists. The claim that death is a just punishment necessarily refers to the existence of certain public beliefs. The claim must be that for capital crimes death alone comports with society’s notion of proper punishment. As administered today, however, the punishment of death cannot be justified as a necessary means of exacting retribution from criminals. When the overwhelming number of criminals who commit capital crimes go to prison, it cannot be concluded that death serves the purpose of retribution more effectively than imprisonment. The asserted public belief that murderers and rapists deserve to die is flatly inconsistent with the execution of a random *305few. As the history of the punishment of death in this country shows, our society wishes to prevent crime; we have no desire to kill criminals simply to get even with them.

In sum, the punishment of death is inconsistent with all four principles: Death is an unusually severe and degrading punishment; there is a strong probability that it is inflicted arbitrarily; its rejection by contemporary society is virtually total; and there is no reason to believe that it serves any penal purpose more effectively than the less severe punishment of imprisonment. The function of these principles is to enable a court to determine whether a punishment comports with human dignity. Death, quite simply, does not.

IV

When this country was founded, memories of the Stuart horrors were fresh and severe corporal punishments were common. Death was not then a unique punishment. The practice of punishing criminals by death, moreover, was widespread and by and large acceptable to society. Indeed, without developed prison systems, there was frequently no workable alternative. Since that time, successive restrictions, imposed against the background of a continuing moral controversy, have drastically curtailed the use of this punishment. Today death is a uniquely and unusually severe punishment. When examined by the principles applicable under the Cruel and Unusual Punishments Clause, death stands condemned as fatally offensive to human dignity. The punishment of death is therefore “cruel and unusual,” and the States may no longer inflict it as a punishment for crimes. Rather than kill an arbitrary handful of criminals each year, the States will confine them in prison. “The State thereby suffers nothing and loses no power. The purpose of punishment is fulfilled, crime *306is repressed by penalties of just, not tormenting, severity, its repetition is prevented, and hope is given for the reformation of the criminal/’ Weems v. United States, 217 U. S., at 381.

1 concur in the judgments of the Court.

Me. Justice Stewaet,

concurring.

The penalty of death differs from all other forms of criminal punishment, not in degree but in kind. It is unique in its total irrevocability. It is unique in its rejection of rehabilitation of the convict as a basic purpose of criminal justice. And it is unique, finally, in its absolute renunciation of all that is embodied in our concept of humanity.

For these and other reasons, at least two of my Brothers have concluded that the infliction of the death penalty is constitutionally impermissible in all circumstances under the Eighth and Fourteenth Amendments. Their case is a strong one. But I find it unnecessary to reach the ultimate question they would decide. See Ashwander v. Tennessee Valley Authority, 297 U. S. 288, 347 (Brandéis, J., concurring).

The opinions of other Justices today have set out in admirable and thorough detail the origins and judicial history of the Eighth Amendment’s guarantee against the infliction of cruel and unusual punishments,1 and the origin and judicial history of capital punishment.2 There *307is thus no need for me to review the historical materials here, and what I have to say can, therefore, be briefly stated.

Legislatures — state and federal — have sometimes specified that the penalty of death shall be the mandatory punishment for every person convicted of engaging in certain designated criminal conduct. Congress, for example, has provided that anyone convicted of acting as a spy for the enemy in time of war shall be put to death.3 The Rhode Island Legislature has ordained the death penalty for a life term prisoner who commits murder.4 Massachusetts has passed a law imposing the death penalty upon anyone convicted of murder in the commission of a forcible rape.5 An Ohio law imposes the mandatory penalty of death upon the assassin of the President of the United States or the Governor of a State.6

If we were reviewing death sentences imposed under these or similar laws, we would be faced with the need to decide whether capital punishment is unconstitutional for all crimes and under all circumstances. We would need to decide whether a legislature — state or federal— could constitutionally determine that certain criminal conduct is so atrocious that society’s interest in deterrence and retribution wholly outweighs any considerations of reform or rehabilitation of the perpetrator, and that, despite the inconclusive empirical evidence,7 only *308the automatic penalty of death will provide maximum deterrence.

On that score I would say only that I cannot agree that retribution is a constitutionally impermissible ingredient in the imposition of punishment. The instinct for retribution is part of the nature of man, and channeling that instinct in the administration of criminal justice serves an important purpose in promoting the stability of a society governed by law. When people begin to believe that organized society is unwilling or unable to impose upon criminal offenders the punishment they “deserve,” then there are sown the seeds of anarchy — of self-help, vigilante justice, and lynch law.

The constitutionality of capital punishment in the abstract is not, however, before us in these cases. For the Georgia and Texas Legislatures have not provided that the death penalty shall be imposed upon all those who are found guilty of forcible rape.8 And the Georgia Legislature has not ordained that death shall be the automatic punishment for murder.9 In a word, neither State *309has made a legislative determination that forcible rape and murder can be deterred only by imposing the penalty of death upon all who perpetrate those offenses. As Mr. Justice White so tellingly puts it, the “legislative will is not frustrated if the penalty is never imposed.” Post, at 311.

Instead, the death sentences now before us are the product of a legal system that brings them, I believe, within the very core of the Eighth Amendment’s guarantee against cruel and unusual punishments, a guarantee applicable against the States through the Fourteenth Amendment. Robinson v. California, 370 U. S. 660. In the first place, it is clear that these sentences are “cruel” in the sense that they excessively go beyond, not in degree but in kind, the punishments that the state legislatures have determined to be necessary. Weems v. United States, 217 U. S. 349. In the second place, it is equally clear that these sentences are “unusual” in the sense that the penalty of death is infrequently imposed for murder, and that its imposition for rape is extraordinarily rare.10 But I do not rest my conclusion upon these two propositions alone.

These death sentences are cruel and unusual in the same way that being struck by lightning is cruel and unusual. For, of all the people convicted of rapes and murders in 1967 and 1968,11 many just as reprehensible as these, the petitioners are among a capriciously *310selected random handful upon whom the sentence of death has in fact been imposed.12 My concurring Brothers have demonstrated that, if any basis can be discerned for the selection of these few to be sentenced to die, it is the constitutionally impermissible basis of race.13 See McLaughlin v. Florida, 379 U. S. 184. But racial discrimination has not been proved,14 and I put it to one side. I simply conclude that 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.

For these reasons I concur in the judgments of the Court.

Me. Justice White,

concurring.

The facial constitutionality of statutes requiring the imposition of the death penalty for first-degree murder, for more narrowly defined categories of murder, or for rape would present quite different issues under the Eighth Amendment than are posed by the cases before us. In joining the Court's judgments, therefore, I do not at all *311intimate that the death penalty is unconstitutional per se or that there is no system of capital punishment that would comport with the Eighth Amendment. That question, ably argued by several of my Brethren, is not presented by these cases and need not be decided.

The narrower question to which I address myself concerns the constitutionality of capital punishment statutes under which (1) the legislature authorizes the imposition of the death penalty for murder or rape; (2) the legislature does not itself mandate the penalty in any particular class or kind of case (that is, legislative will is not frustrated if the penalty is never imposed), but delegates to judges or juries the decisions as to those cases, if any, in which the penalty will be utilized; and (3) judges and juries have ordered the death penalty with such infrequency that the odds are now very much against imposition and execution of the penalty with respect to any convicted murderer or rapist. It is in this context that we must consider whether the execution of these petitioners would violate the Eighth Amendment.

I begin with what I consider a near truism: that the death penalty could so seldom be imposed that it would cease to be a credible deterrent or measurably to contribute to any other end of punishment in the criminal justice system. It is perhaps true that no matter how infrequently those convicted of rape or murder are executed, the penalty so imposed is not disproportionate to the crime and those executed may deserve exactly what they received. It would also be clear that executed defendants are finally and completely incapacitated from again committing rape or murder or any other crime. But when imposition of the penalty reaches a certain degree of infrequency, it would be very doubtful that any existing general need for retribution would be measurably satisfied. Nor could it be said with confidence that society’s need for specific deterrence justifies death *312for so few when for so many in like circumstances life imprisonment or shorter prison terms are judged sufficient, or that community values are measurably reinforced by authorizing a penalty so rarely invoked.

Most important, a major goal of the criminal law— to deter others by punishing the convicted criminal— would not be substantially served where the penalty is so seldom invoked that it ceases to be the credible threat essential to influence the conduct of others. For present purposes I accept the morality and utility of punishing one person to influence another. I accept also the effectiveness of punishment generally and need not reject the death penalty as a more effective- deterrent than a lesser punishment. But common sense and experience tell us that seldom-enforced laws become ineffective measures for controlling human conduct and that the death penalty, unless- imposed with sufficient frequency, will make little contribution to deterring those crimes for which it may be exacted.

The imposition and execution of the death penalty are obviously cruel in the dictionary sense. But the penalty has not been considered cruel and unusual punishment in the constitutional sense because it was thought justified by the social ends it was deemed to serve. At the moment that it ceases realistically to further these purposes, however, the emerging question is whether its imposition in such circumstances would violate the Eighth Amendment. It is my view that it would, for its imposition would then be the pointless and needless extinction of life with only marginal contributions to any discernible social or public purposes. A penalty with such negligible returns to the State would be patently excessive and cruel and unusual punishment vio-lative of the Eighth Amendment.

It is also my judgment that this point has been reached with respect to capital punishment as it is presently ad*313ministered under the statutes involved in these cases. Concededly, it is difficult to prove as a general proposition that capital punishment, however administered, more effectively serves the ends of the criminal law than does imprisonment. But however that may be, I cannot avoid the conclusion that as the statutes before us are now administered, the penalty is so infrequently imposed that the threat of execution is too attenuated to be of substantial service to criminal justice.

I need not restate the facts and figures that appear in the opinions of my Brethren. Nor can I “prove” my conclusion from these data. But, like my Brethren, I must arrive at judgment; and I can do no more than state a conclusion based' on 10 years of almost daily exposure to the facts and circumstances of hundreds and hundreds of federal and state criminal cases involving crimes for which death is the authorized penalty. That conclusion, as I have said, is that the death penalty is exacted with great infrequency even for the most atrocious crimes and that there is no meaningful basis for distinguishing the few cases in which it is imposed from the many cases in which it is not. The short of it is that the policy of vesting sentencing authority primarily in juries — a decision largely motivated by the desire to mitigate the harshness of the law and to bring community judgment to bear on the sentence as well as guilt or innocence — has so effectively achieved its aims that capital punishment within the confines of the statutes now before us has for all practical purposes run its course.

Judicial review, by definition, often involves a conflict between judicial and legislative judgment as to what the Constitution means or requires. In this respect, Eighth Amendment cases come to us in no different posture. It seems conceded by all that the Amendment imposes some obligations on the judiciary to judge the *314constitutionality of punishment and that there are punishments that the Amendment would bar whether legislatively approved or not. Inevitably, then, there will be occasions when we will differ with Congressi or state legislatures with respect to the validity of punishment. There will also be cases in which we shall strongly disagree among ourselves. Unfortunately, this is one .of them. But as I see it, this case is no different in kind from many others, although it may have wider impact and provoke sharper disagreement.

In this respect, I add only that past and present legislative judgment with respect to the death penalty loses much of its force when viewed in light of the recurring practice of delegating sentencing authority to the jury and the fact that a jury, in its own discretion and without violating its trust or any statutory policy, may refuse to impose the death penalty no matter what the circumstances of the crime. Legislative “policy” is thus necessarily defined not by what is legislatively authorized but by what juries and judges do in exercising the discretion so regularly conferred upon them. In my judgment what was done in these cases violated the Eighth Amendment.

I concur in the judgments of the Court.

Mr. Justice Marshall,

concurring.

These three cases present the question whether the death penalty is a cruel and unusual punishment prohibited by the Eighth Amendment to the United States Constitution.1

*315In No. 69-5003, Furman was convicted of murder for shooting the father of five children when he discovered that Furman had broken into his home early one morning. Nos. 69-5030 and 69-5031 involve state convictions for forcible rape. Jackson was found guilty of rape during the course of a robbery in the victim’s home. The rape was accomplished as he held the pointed ends of scissors at the victim’s throat. Branch also was convicted of a rape committed in the victim’s home. No weapon was utilized, but physical force and threats of physical force were employed.

The criminal acts with which we are confronted are ugly, vicious, reprehensible acts. Their sheer brutality cannot and should not be minimized. But, we are not called upon to condone the penalized conduct; we are asked only to examine the penalty imposed on each of the petitioners and to determine whether or not it violates the Eighth Amendment. The question then is not whether we condone rape or murder, for surely we do not; it is whether capital punishment is “a punishment no longer consistent with our own self-respect” 2 and, therefore, violative of the Eighth Amendment.

The elasticity of the constitutional provision under consideration presents dangers of too little or too much self-restraint.3 Hence, we must proceed with caution to answer the question presented.4 By first examining the historical derivation of the Eighth Amendment and *316the construction given it in the past by this Court, and then exploring the history and attributes of capital punishment in this country, we can answer the question presented with objectivity and a proper measure of self-restraint.

Candor is critical to such an inquiry. All relevant material must be marshaled and sorted and forthrightly examined. We must not only be precise as to the standards of judgment that we are utilizing, but exacting in examining the relevant material in light of those standards.

Candor compels me to confess that I am not oblivious to the fact that this is truly a matter of life and death. Not only does it involve the lives of these three petitioners, but those of the almost 600 other condemned men and women in this country currently awaiting execution. While this fact cannot affect our ultimate decision, it necessitates that the decision be free from any possibility of error.

I

The Eighth Amendment's ban against cruel and unusual punishments derives from English law. In 1583, John Whitgift, Archbishop of Canterbury, turned the High Commission into a permanent ecclesiastical court, and the Commission began to use torture to extract confessions from persons suspected of various offenses.5 Sir Robert Beale protested that cruel and barbarous torture violated Magna Carta, but his protests were made in vain.6

*317Cruel punishments were not confined to those accused of crimes, but were notoriously applied with even greater relish to those who were convicted. Blackstone described in ghastly detail the myriad of inhumane forms of punishment imposed on persons found guilty of any of a large number of offenses.7 Death, of course, was the usual result.8

The treason trials of 1685 — the “Bloody Assizes”— which followed an abortive rebellion by the Duke of Monmouth, marked the culmination of the parade of horrors, and most historians believe that it was this event that finally spurred the adoption of the English Bill of Rights containing the progenitor of our prohibition against cruel and unusual punishments.9 The conduct of Lord Chief Justice Jeffreys at those trials has been described as an “insane lust for cruelty” which was “stimulated by orders from the King” (James II).10 The assizes received wide publicity from Puritan pamphleteers and doubtless had some influence on the adoption of a cruel and unusual punishments clause. But, *318the legislative history of the English Bill of Rights of 1689 indicates that the assizes may not have been as critical to the adoption of the clause as is widely thought. After William and Mary of Orange crossed the channel to invade England, James II fled. Parliament was summoned into session and a committee was appointed to draft general statements containing “such things as are absolutely necessary to be considered for the better securing of our religion, laws and liberties.”11 An initial draft of the Bill of Rights prohibited “illegal” punishments, but a later draft referred to the infliction by James II of “illegal and cruel” punishments, and declared “cruel and unusual” punishments to be prohibited.12 The use of the word “unusual” in the final draft appears to be inadvertent.

This legislative history has led at least one legal historian to conclude “that the cruel and unusual punishments clause of the Bill of Rights of 1689 was, first, an objection to the imposition of punishments that were unauthorized by statute and outside the jurisdiction of the sentencing court, and second, a reiteration of the English policy against disproportionate penalties,” 13 and not primarily a reaction to the torture of the High Commission, harsh sentences, or the assizes.

*319Whether the English Bill of Rights prohibition against cruel and unusual punishments is properly read as a response to excessive or illegal punishments, as a reaction to barbaric and objectionable modes of punishment, or as both, there is no doubt whatever that in borrowing the language and in including it in the Eighth Amendment, our Founding Fathers intended to outlaw torture and other cruel punishments.14

The precise language used in the Eighth Amendment first appeared in America on June 12, 1776, in Virginia’s “Declaration of Rights,” § 9 of which read: “That excessive bail ought not to be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.” 15 This language was drawn verbatim from the English Bill of Rights of 1689. Other States adopted similar clauses,16 and there is evidence in the debates of the various state conventions that were *320called upon to ratify the Constitution of great concern for the omission of any prohibition against torture or other cruel punishments.17

The Virginia Convention offers some clues as to what the Founding Fathers had in mind in prohibiting cruel and unusual punishments. At one point George Mason advocated the adoption of a Bill of Rights, and Patrick Henry concurred, stating:

“By this Constitution, some of the best barriers of human rights are thrown away. Is there not an additional reason to have a bill of rights? . . . Congress, from their general powers, may fully go into business of human legislation. They may legislate, in criminal cases, from treason to the lowest offence — petty larceny. They may define crimes and prescribe punishments. In the definition of crimes, I trust they will be directed by what wise representatives ought to be governed by. But, when we come to punishments, no latitude ought to be left, nor dependence put on the virtue of representatives. What says our bill of rights? — ‘that excessive bail ought not to be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.’ Are you not, therefore, now calling on those gentlemen who are to compose Congress, to prescribe trials and define punishments without this control? Will they find sentiments there similar to this bill of rights? You let them loose; you do more — you depart from the genius of your country. . . .

“In this business of legislation, your members of Congress will loose the restriction of not imposing excessive fines, demanding excessive bail, and in*321flicting cruel and unusual punishments. These are prohibited by your declaration of rights. What has distinguished our ancestors? — That they would not admit of tortures, or cruel and barbarous punishment. But Congress may introduce the practice of the civil law, in preference to that of the common law. They may introduce the practice of France, Spain, and Germany — of torturing, to extort a confession of the crime. They will say that they might as well draw examples from those countries as from Great Britain, and they will tell you that there is such a necessity of strengthening the arm of government, that they must have a criminal equity, and extort confession by torture, in order to punish with still more relentless severity. We are then lost and undone.” 18

Henry’s statement indicates that he wished to insure that “relentless severity” would be prohibited by the Constitution. Other expressions with respect to the proposed Eighth Amendment by Members of the First Congress indicate that they shared Henry’s view of the need for and purpose of the Cruel and Unusual Punishments Clause.19

*322Thus, the history of the clause clearly establishes that it was intended to prohibit cruel punishments. We must now turn to the case law to discover the manner in which courts have given meaning to the term “cruel.”

II

This Court did not squarely face the task of interpreting the cruel and unusual punishments language for the first time until Wilkerson v. Utah, 99 U. S. 130 (1879), although the language received a cursory examination in several prior cases. See, e. g., Pervear v. Commonwealth, 5 Wall. 475 (1867). In Wilkerson, the Court unanimously upheld a sentence of public execution by shooting imposed pursuant to a conviction for premeditated murder. In his opinion for the Court, Mr. Justice Clifford wrote:

“Difficulty would attend the effort to define with exactness the extent of the constitutional provision which provides that cruel and unusual punishments shall not be inflicted; but it is safe to affirm that punishments of torture, . . . and all others in the same line of unnecessary cruelty, are forbidden by that amendment to the Constitution.” 99 U. S., at 135-136.

Thus, the Court found that unnecessary cruelty was no more permissible than torture. To determine whether the punishment under attack Was unnecessarily cruel, the Court examined the history of the Utah Territory and the then-current writings on capital punishment, and compared this Nation’s practices with those of other countries. It is apparent that the Court felt it could not dispose of the question simply by referring to traditional practices; instead, it felt bound to examine developing thought.

Eleven years passed before the Court again faced a challenge to a specific punishment under the Eighth *323Amendment. In the case of In re Kemmler, 136 U. S. 436 (1890), Chief Justice Fuller wrote an opinion for a unanimous Court upholding electrocution as a permissible mode of punishment. While the Court ostensibly held that the Eighth Amendment did not apply' to the States, it is very apparent that the nature of the punishment involved was examined under the Due Process Clause of the Fourteenth Amendment. The Court held that the punishment was not objectionable. Today, Kemmler stands primarily for the proposition that a punishment is not necessarily unconstitutional simply because it is unusual, so long as the legislature has a humane purpose in selecting it.20

Two years later in O’Neil v. Vermont, 144 U. S. 323 (1892), the Court reaffirmed that the Eighth Amendment was not applicable to the States. O’Neil was found guilty on 307 counts of selling liquor in violation of Vermont law. A fine of $6,140 ($20 for each offense) and the costs of prosecution ($497.96) were imposed. O’Neil was committed to prison until the fine and the costs were paid; and the court provided that if they were not paid before a specified date, O’Neil was to be confined in the house of corrections for 19,914 days (approximately 54 years) at hard labor. Three Justices — Field, Harlan, and Brewer — dissented. They maintained not only that the Cruel and Unusual Punishments Clause was applicable to the States, but that in O’Neil’s case it had been violated. Mr. Justice Field wrote:

“That designation [cruel and unusual], it is true, is usually applied to punishments which inflict torture, such as the rack, the thumbscrew, the iron boot, the stretching of limbs and the like, which *324are attended with acute pain and suffering. . . . The inhibition is directed, not only against punishments of the character mentioned, but against all punishments which by their excessive length or severity are greatly disproportioned to the offences charged. The whole inhibition is against that which is excessive . . . Id., at 339-340.

In Howard v. Fleming, 191 U. S. 126 (1903), the Court, in essence, followed the approach advocated by the dissenters in O’Neil. In rejecting the claim that 10-year sentences for conspiracy to defraud were cruel and unusual, the Court (per Mr. Justice Brewer) considered the nature of the crime, the purpose of the law, and the length of the sentence imposed.

The Court used the same approach seven years later in the landmark case of Weems v. United States, 217 U. S. 349 (1910). Weems, an officer of the Bureau of Coast Guard and Transportation of the United States Government of the Philippine Islands, was convicted of falsifying a “public and official document.” He was sentenced to 15 years’ incarceration at hard labor with chains on his ankles, to an unusual loss of his civil rights, and to perpetual surveillance. Called upon to determine whether this was a cruel and unusual punishment, the Court found that it was.21 The Court emphasized that the Constitution was not an “ephemeral” enactment, or one “designed to meet passing occasions.” 22 Recognizing that “[t]ime works changes, [and] brings into existence new conditions and purposes,” 23 the Court commented that “[i]n the application of a constitu*325tion . . . our contemplation cannot be only of what has been but of what may be.” 24

In striking down the penalty imposed on Weems, the Court examined the punishment in relation to the offense, compared the punishment to those inflicted for other crimes and to those imposed in other jurisdictions, and concluded that the punishment was excessive.25 Justices White and Holmes dissented and argued that the cruel and unusual prohibition was meant to prohibit only those things that were objectionable at the time the Constitution was adopted.26

Weems is a landmark case because it represents the first time that the Court invalidated a penalty prescribed by a legislature for a particular offense. The Court made it plain beyond any reasonable doubt that excessive punishments were as objectionable as those that were inherently cruel. Thus, it is apparent that the dissenters’ position in O’Neil had become the opinion of the Court in Weems.

Weems was followed by two cases that added little to our knowledge of the scope of the cruel and unusual language, Badders v. United States, 240 U. S. 391 (1916), and United States ex rel. Milwaukee Social Democratic Publishing Co. v. Burleson, 255 U. S. 407 (1921).27 Then *326came another landmark case, Louisiana ex rel. Francis v. Resweber, 329 U. S. 459 (1947).

Francis had been convicted of murder and sentenced to be electrocuted. The first time the current passed through him, there was a mechanical failure and he did not die. Thereafter, Francis sought to prevent a second electrocution on the ground that it would be a cruel and unusual punishment. Eight members of the Court assumed the applicability of the Eighth Amendment to the States.28 The Court was virtually unanimous in agreeing that “[t]he traditional humanity of modern Anglo-American law forbids the infliction of unnecessary pain,” 29 but split 5-4 on whether Francis would, under the circumstances, be forced to undergo any excessive pain. Five members of the Court treated the case like In re Kemmler and held that the legislature adopted electrocution for a humane purpose, and that its will should not be thwarted because, in its desire to reduce pain and suffering in most cases, it may have inadvertently increased suffering in one particular case.30 *327The four dissenters felt that the case should be remanded for further facts.

As in Weems, the Court was concerned with excessive punishments. Resweber is perhaps most significant because the analysis of cruel and unusual punishment questions first advocated by the dissenters in O’Neil was at last firmly entrenched in the minds of an entire Court.

Trop v. Dulles, 356 U. S. 86 (1958), marked the next major cruel and unusual punishment case in this Court. Trop, a native-born American, was declared to have lost his citizenship by reason of a conviction by court-martial for wartime desertion. Writing for himself and Justices Black, Douglas, and Whittaker, Chief Justice Warren concluded that loss of citizenship amounted to a cruel and unusual punishment that violated the Eighth Amendment.31

Emphasizing the flexibility inherent in the words “cruel and unusual,” the Chief Justice wrote that “{t]he Amendment must draw its meaning from the evolving standards of decency that mark the progress of a maturing society.” 32 His approach to the problem was that utilized by the Court in Weems: he scrutinized the severity of the penalty in relation to the offense, examined the practices of other civilized nations of the world, and concluded that involuntary statelessness was an excessive and, therefore, an unconstitutional punishment. Justice Frankfurter, dissenting, urged that expatriation was not punishment, and that even if it were, it was not excessive. While he criticized the conclusion arrived at by the Chief Justice, his approach to the Eighth Amendment question was identical.

*328Whereas in Trop a majority of the Court failed to agree on whether loss of citizenship was a cruel and unusual punishment, four years later a majority did agree in Robinson v. California, 370 U. S. 660 (1962), that a sentence of 90 days’ imprisonment for violation of a California statute making it a crime to “be addicted to the use of narcotics” was cruel and unusual. Mr. Justice Stewart, writing the opinion of the Court, reiterated what the Court had said in Weems and what Chief Justice Warren wrote in Trop — that the cruel and unusual punishment clause was not a static concept, but one that must be continually re-examined “in the light of contemporary human knowledge.” 33 The fact that the penalty under attack was only 90 days evidences the Court’s willingness to carefully examine the possible excessiveness of punishment in a given case even where what is involved is a penalty that is familiar and widely accepted.34

We distinguished Robinson in Powell v. Texas, 392 U. S. 514 (1968), where we sustained a conviction for drunkenness in a public place and a fine of $20. Four Justices dissented on the ground that Robinson was controlling. The analysis in both cases was the same; only the conclusion as to whether or not the punishment was excessive differed. Powell marked the last time prior to today’s decision that the Court has had occasion to construe the meaning of the term “cruel and unusual” punishment.

Several principles emerge from these prior cases and serve as a beacon to an enlightened decision in the instant cases.

*329III

Perhaps the most important principle in analyzing “cruel and unusual” punishment questions is one that is reiterated again and again in the prior opinions of the Court: i. e., the cruel and unusual language “must draw its meaning from the evolving standards of decency that mark the progress of a maturing society.” 35 Thus, a penalty that was permissible at one time in our Nation’s history is not necessarily permissible today.

The fact, therefore, that the Court, or individual Justices, may have in the past expressed an opinion that the death penalty is constitutional is not now binding on us. A fair reading of Wilkerson v. Utah, supra; In re Kemmler, supra; and Louisiana ex rel. Francis v. Resweber, supra, would certainly indicate an acceptance sub silentio of capital punishment as constitutionally permissible. Several Justices have also expressed their individual opinions that the death penalty is constitutional.36 Yet, some of these same Justices and others have at times expressed concern over capital punishment.37 *330There is no holding directly in point, and the very nature of the Eighth Amendment would dictate that unless a very recent decision existed, stare decisis would bow to changing values, and the question of the constitutionality of capital punishment at a given moment in history would remain open.

Faced with an open question, we must establish our standards for decision. The decisions discussed in the previous section imply that a punishment may be deemed cruel and unusual for any one of four distinct reasons.

First, there are certain punishments that inherently involve so much physical pain and suffering that civilized people cannot tolerate them — e. g., use of the rack, the thumbscrew, or other modes of torture. See O’Neil v. Vermont, 144 U. S., at 339 (Field, J., dissenting). Regardless of public sentiment with respect to imposition of one of these punishments in a particular case or at any one moment in history, the Constitution prohibits it. These are punishments that have been barred since the adoption of the Bill of Rights.

*331Second, there are punishments that are unusual, signifying that they were previously unknown as penalties for a given offense. Cf. United States ex rel. Milwaukee Social Democratic Publishing Co. v. Burleson, 255 U. S., at 435 (Brandéis, J., dissenting). If these punishments are intended to serve a humane purpose, they may be constitutionally permissible. In re Kemmler, 136 U. S., at 447; Louisiana ex rel. Francis v. Resweber, 329 U. S., at 464. Prior decisions leave open the question of just how much the word “unusual” adds to the word “cruel.” I have previously indicated that use of the word “unusual” in the English Bill of Rights of 1689 was inadvertent, and there is nothing in the history of the Eighth Amendment to give flesh to its intended meaning. In light of the meager history that does exist, one would suppose that an innovative punishment would probably be constitutional if no more cruel than that punishment which it superseded. We need not decide this question here, however, for capital punishment is certainly not a recent phenomenon.

Third, a penalty may be cruel and unusual because it is excessive and serves no valid legislative purpose. Weems v. United States, supra. The decisions previously discussed are replete with assertions that one of the primary functions of the cruel and unusual punishments clause is to prevent excessive or unnecessary penalties, e. g., Wilkerson v. Utah, 99 U. S., at 134; O’Neil v. Vermont, 144 U. S., at 339-340 (Field, J., dissenting); Weems v. United States, 217 U. S., at 381; Louisiana ex rel. Francis v. Resweber, supra; these punishments are unconstitutional even though popular sentiment may favor them. Both The Chief Justice and Mr. Justice Powell seek to ignore or to minimize this aspect of the Court’s prior decisions. But, since Mr. Justice Field first suggested that “[t]he whole inhibition [of the prohibition against cruel and unusual punish*332ments] is against that which is excessive,” O’Neil v. Vermont, 144 U. S., at 340, this Court has steadfastly maintained that a penalty is unconstitutional whenever it is unnecessarily harsh or cruel. This is what the Founders of this country intended; this is what their fellow citizens believed the Eighth Amendment provided; and this was the basis for our decision in Robinson v. California, supra, for the plurality opinion by Mr. Chief Justice Warren in Trop v. Dulles, supra, and for the Court’s decision in Weems v. United States, supra. See also W. Bradford, An Enquiry How Far the Punishment of Death is Necessary in Pennsylvania (1793), reprinted in 12 Am. J. Legal Hist. 122, 127 (1968). It should also be noted that the "cruel and unusual” language of the Eighth Amendment immediately follows language that prohibits excessive bail and excessive fines. The entire thrust of the Eighth Amendment is, in short, against “that which is excessive.”

Fourth, where a punishment is not excessive and serves a valid legislative purpose, it still may be invalid if popular sentiment abhors it. For example, if the evidence clearly demonstrated that capital punishment served valid legislative purposes, such punishment would, nevertheless, be unconstitutional if citizens found it to be morally unacceptable. A general abhorrence on the part of the public would, in effect, equate a modern punishment with those barred since the adoption of the Eighth Amendment. There are no prior cases in this Court striking down a penalty on this ground, but the very notion of changing values requires that we recognize its existence.

It is immediately obvious, then, that since capital punishment is not a recent phenomenon, if it violates the Constitution, it does so because it is excessive or *333unnecessary, or because it is abhorrent to currently existing moral values.

We must proceed to the history of capital punishment in the United States.

IV

Capital punishment has been used to penalize various forms of conduct by members of society since the beginnings of civilization. Its precise origins are difficult to perceive, but there is some evidence that its roots lie in violent retaliation by members of a tribe or group, or by the tribe or group itself, against persons committing hostile acts toward group members.38 Thus, infliction of death as a penalty for objectionable conduct appears to have its beginnings in private vengeance.39

As individuals gradually ceded their personal prerogatives to a sovereign power, the sovereign accepted the authority to punish wrongdoing as part of its “divine right” to rule. Individual vengeance gave way to the vengeance of the state, and capital punishment became a public function.40 Capital punishment worked its way into the laws of various countries,41 and was inflicted in a variety of macabre and horrific ways.42

It was during the reign of Henry II (1154-1189) that English law first recognized that crime was more than a personal affair between the victim and the per*334petrator.43 The early history of capital punishment in England is set forth in McGautha v. California, 402 U. S. 183, 197-200 (1971), and need not be repeated here.

By 1500, English law recognized eight major capital crimes: treason, petty treason (killing of husband by his wife), murder, larceny, robbery, burglary, rape, and arson.44 Tudor and Stuart kings added many more crimes to the list of those punishable by death, and by 1688 there were nearly 50.45 George II (1727-1760) added nearly 36 more, and George III (1760-1820) increased the number by 60.46

By shortly after 1800, capital offenses numbered more than 200 and not only included crimes against person and property, but even some against the public peace. While England may, in retrospect, look particularly brutal, Blackstone points out that England was fairly civilized when compared to the rest of Europe.47

*335Capital punishment was not as common a penalty in the American Colonies. “The Capitall Lawes of New-England,” dating from 1636, were drawn by the Massachusetts Bay Colony and are the first written expression of capital offenses known to exist in this country. These laws make the following crimes capital offenses: idolatry, witchcraft, blasphemy, murder, assault in sudden anger, sodomy, buggery, adultery, statutory rape, rape, man-stealing, perjury in a capital trial, and rebellion. Each crime is accompanied by a reference to the Old Testament to indicate its source.48 It is not known with any certainty exactly when, or even if, these laws were enacted as drafted; and, if so, just how vigorously these laws were enforced.49 We do know that the other Colonies had a variety of laws that spanned the spectrum of severity.50

By the 18th century, the list of crimes became much less theocratic and much more secular. In the average colony, there were 12 capital crimes.51 This was far fewer than existed in England, and part of the reason was that there was a scarcity of labor in the Colonies.52 Still, there were many executions, because “[w]ith county jails inadequate and insecure, the criminal population seemed best controlled by death, mutilation, and fines.” 53

Even in the 17th century, there was some opposition *336to capital punishment in some of the colonies. In his “Great Act” of 1682, William Penn prescribed death only for premeditated murder and treason,54 although his reform was not long lived.55

In 1776 the Philadelphia Society for Relieving Distressed Prisoners organized, and it was followed 11 years later by the Philadelphia Society for Alleviating the Miseries of Public Prisons.56 These groups pressured for reform of all penal laws, including capital offenses. Dr. Benjamin Rush soon drafted America’s first reasoned argument against capital punishment, entitled An Enquiry into the Effects of Public Punishments upon Criminals and upon Society.57 In 1793, William Bradford, the Attorney General of Pennsylvania and later Attorney General of the United States, conducted “An Enquiry How Ear the Punishment of Death is Necessary in Pennsylvania.”58 He concluded that it was doubtful whether capital punishment was at all necessary, and that until more information could be obtained, it should be immediately eliminated for all offenses except high treason and murder.59

The “Enquiries” of Rush and Bradford and the Pennsylvania movement toward abolition of the death *337penalty had little immediate impact on the practices of other States.60 But in the early 1800’s, Governors George and DeWitt Clinton and Daniel Tompkins unsuccessfully urged the New York Legislature to modify or end capital punishment. During this same period, Edward Livingston, an American lawyer who later became Secretary of State and Minister to France under President Andrew Jackson, was appointed by the Louisiana Legislature to draft a new penal code. At the center of his proposal was “the total abolition of capital punishment.” 61 His Introductory Report to the System of Penal Law Prepared for the State of Louisiana62 contained a systematic rebuttal of all arguments favoring capital punishment. Drafted in 1824, it was not published until 1833. This work was a tremendous impetus to the abolition movement for the next half century.

During the 1830’s, there was a rising tide of sentiment against capital punishment. In 1834, Pennsylvania abolished public executions,63 and two years later, The Report on Capital Punishment Made to the Maine Legislature was published. It led to a law that prohibited the executive from issuing a warrant for execution within one year after a criminal was sentenced by the courts. The totally discretionary character of the law was at odds with almost all prior practices. The “Maine Law” resulted in little enforcement of the death penalty, which was not surprising since the legislature’s idea in passing the law was that the affirmative burden placed on the governor to issue a warrant one full year *338or more after a trial would be an effective deterrent to exercise of his power.64 The law spread throughout New England and led to Michigan's being the first State to abolish capital punishment in 1846.65

Anti-capital-punishment feeling grew in the 1840’s as the literature of the period pointed out the agony of the condemned man and expressed the philosophy that repentance atoned for the worst crimes, and that true repentance derived, not from fear, but from harmony with nature.66

By 1850, societies for abolition existed in Massachusetts, New York, Pennsylvania, Tennessee, Ohio, Alabama, Louisiana, Indiana, and Iowa.67 New York, Massachusetts, and Pennsylvania constantly had abolition bills before their legislatures. In 1852, Rhode Island followed in the footsteps of Michigan and partially abolished capital punishment.68 Wisconsin totally abolished the death penalty the following year.69 Those States that did not abolish the death penalty greatly reduced its scope, and “[f]ew states outside the South had more than one or two . . . capital offenses” in addition to treason and murder.70

But the Civil War halted much of the abolition furor. One historian has said that “[a]fter the Civil War, men’s finer sensibilities, which had once been revolted by the execution of a fellow being, seemed hardened and *339blunted.”71 Some of the attention previously given to abolition was diverted to prison reform. An abolitionist movement still existed, however. Maine abolished the death penalty in 1876, restored it in 1883, and abolished it again in 1887; Iowa abolished capital punishment from 1872-1878; Colorado began an erratic period of de jacto abolition and revival in 1872; and Kansas also abolished it de jacto in 1872, and by law in 1907.72

One great success of the abolitionist movement in the period from 1830-1900 was almost complete elimination of mandatory' capital punishment. Before the legislatures formally gave juries discretion to refrain from imposing the death penalty, the phenomenon of “jury nullification,” in which juries refused to convict in cases in which they believed that death was an inappropriate penalty, was experienced.73 Tennessee was the first State to give juries discretion, Tenn. Laws 1837-1838, c. 29, but other States quickly followed suit. Then, Rep. Curtis of New York introduced a federal bill that ultimately became law in 1897 which reduced the number of federal capital offenses from 60 to 3 (treason, murder, and rape) and gave the jury sentencing discretion in murder and rape cases.74

By 1917 12 States had become abolitionist jurisdictions.75 But, under the nervous tension of World War I, *340four of these States reinstituted capital punishment and promising movements in other States came grinding to a halt.76 During the period following the First World War, the abolitionist movement never regained its momentum.

It is not easy to ascertain why the movement lost its vigor. Certainly, much attention was diverted from penal reform during the economic crisis of the depression and the exhausting years of struggle during World War II. Also, executions, which had once been frequent public spectacles, became infrequent private affairs. The manner of inflicting death changed, and the horrors of the punishment were, therefore, somewhat diminished in the minds of the general public.77

In recent years there has been renewed interest in modifying capital punishment. New York has moved toward abolition,78 as have several other States.79 In 1967, a bill was introduced in the Senate to abolish *341capital punishment for all federal crimes, but it died in committee.80

At the present time, 41 States, the District of Columbia, and other federal jurisdictions authorize the death penalty for at least one crime. It would be fruitless to attempt here to categorize the approach to capital punishment taken by the various States.81 It is sufficient to note that murder is the crime most often punished by death, followed by kidnaping and treason.82 Rape is a capital offense in 16 States and the federal system.83

The foregoing history demonstrates that capital punishment was carried from Europe to America but, once here, was tempered considerably. At times in our history, strong abolitionist movements have existed. But, they have never been completely successful, as no more than one-quarter of the States of the Union have, at any one time, abolished the death penalty. They have had partial success, however, especially in reducing the number of capital crimes, replacing mandatory death sentences with jury discretion, and developing more humane methods of conducting executions.

This is where our historical foray leads. The question now to be faced is whether American society has *342reached a point where abolition is not dependent on a successful grass roots movement in particular jurisdictions, but is demanded by the Eighth Amendment. To answer this question, we must first examine whether or not the death penalty is today tantamount to excessive punishment.

V

In order to assess whether or not death is an excessive or unnecessary penalty, it is necessary to consider the reasons why a legislature might select it as.punishment for one or more offenses, and examine whether less severe penalties would satisfy the legitimate legislative wants as well as capital punishment. If they would, then the death penalty is unnecessary cruelty, and, therefore, unconstitutional.

There are six purposes conceivably served by capital punishment: retribution, deterrence, prevention of repetitive criminal acts, encouragement of guilty pleas and confessions, eugenics, and economy. These are considered seriatim, below.

A. The concept of retribution is one of the most misunderstood in all of our criminal jurisprudence. The principal source of confusion derives from the fact that, in dealing with the concept, most people confuse the question “why do men in fact punish?” with the question “what justifies men in punishing?” 84 Men may punish for any number of reasons, but the one reason that punishment is morally good or morally justifiable is that someone has broken the law. Thus, it can correctly be said that breaking the law is the sine qua non of punishment, or, in other words, that we only *343tolerate punishment as it is imposed on one who deviates from the norm established by the criminal law.

The fact that the State may seek retribution against those who have broken its laws does not mean that retribution may then become the State’s sole end in punishing. Our jurisprudence has always accepted deterrence in general, deterrence of individual recidivism, isolation of dangerous persons, and rehabilitation as proper goals of punishment. See Trop v. Dulles, 356 U. S., at 111 (Brennan, J., concurring). Retaliation, vengeance, and retribution have been roundly condemned as intolerable aspirations for a government in a free society.

Punishment as retribution has been condemned by scholars for centuries,85 and the Eighth Amendment itself was adopted to prevent punishment from becoming synonymous with vengeance.

In Weems v. United States, 217 U. S., at 381, the Court, in the course of holding that Weems’ punishment violated the Eighth Amendment, contrasted it with penalties provided for other offenses and concluded:

“[Tjhis contrast shows more than different exercises of legislative judgment. It is greater than that. It condemns the sentence in this case as cruel and unusual. It exhibits a difference between unrestrained power and that which is exercised under the spirit of constitutional limitations formed to establish justice. The State thereby suffers nothing and loses no power. The purpose of punishment is fulfilled, crime is repressed by penalties of just, not tormenting, severity, its repetition is prevented, and hope is given for the reformation of the criminal.” (Emphasis added.)

*344It is plain that the view of the Weems Court was that punishment for the sake of retribution was not permissible under the Eighth Amendment. This is the only view that the Court could have taken if the “cruel and unusual” language were to be given any meaning. Retribution surely underlies the imposition of some punishment on one who commits a criminal act. But, the fact that some punishment may be imposed does not mean that any punishment is permissible. If retribution alone could serve as a justification for any particular penalty, then all penalties selected by the legislature would by definition be acceptable means for designating society’s moral approbation of a particular act. The “cruel and unusual” language would thus be read out of the Constitution and the fears of Patrick Henry and the other Pounding Fathers would become realities.

To preserve the integrity of the Eighth Amendment, the Court has consistently denigrated retribution as a permissible goal of punishment.86 It is undoubtedly correct that there is a demand for vengeance on the part of many persons in a community against one who is convicted of a particularly offensive act. At times a cry is heard that morality requires vengeance to evi*345dence society’s abhorrence of the act.87 But the Eighth Amendment is our insulation from our baser selves. The “cruel and unusual” language limits the avenues through which vengeance can be channeled. Were this not so, the language would be empty and a return to the rack and other tortures would be possible in a given case.

Mr. Justice Story wrote that the Eighth Amendment’s limitation on punishment “would seem to be wholly unnecessary in a free government, since it is scarcely possible that any department of such a government should authorize or justify such atrocious conduct.”88

I would reach an opposite conclusion — that only in a free society would men recognize their inherent weaknesses and seek to compensate for them by means of a Constitution.

The history of the Eighth Amendment supports only the conclusion that retribution for its own sake is improper.

B. The most hotly contested issue regarding capital punishment is whether it is better than life imprisonment as a deterrent to crime.89

While the contrary position has been argued,90 it is my firm opinion that the death penalty is a more severe sanction than life imprisonment. Admittedly, there are *346some persons who would rather die than languish in prison for a lifetime. But, whether or not they should be able to choose death as an alternative is a far different question from that presented here — i. e., whether the State can impose death as a punishment. Death is irrevocable; life imprisonment is not. Death, of course, makes rehabilitation impossible; life imprisonment does not. In short, death has always been viewed as the ultimate sanction, and it seems perfectly reasonable to continue to view it as such.91

It must be kept in mind, then, that the question to be considered is not simply whether capital punishment is *347a deterrent, but whether it is a better deterrent than life imprisonment.92

There is no more complex problem than determining the deterrent efficacy of the death penalty. “Capital punishment has obviously failed as a deterrent when a murder is committed. We can number its failures. But we cannot number its successes. No one can ever know how many people have refrained from murder because of the fear of being hanged.” 93 This is the nub of the problem and it is exacerbated by the paucity of useful data. The United States is more fortunate than most countries, however, in that it has what are generally considered to be the world’s most reliable statistics.94

The two strongest arguments in favor of capital punishment as a deterrent are both logical hypotheses devoid of evidentiary support, but persuasive nonetheless. The first proposition was best stated by Sir James Stephen in 1864:

“No other punishment deters men so effectually from committing crimes as the punishment of death. This is one of those propositions which it is difficult to prove, simply because they are in themselves more obvious than any proof can make them. It is possible to display ingenuity in arguing against it, but that is all. The whole experience of mankind is in the other direction. The threat of instant death is the one to which resort has always been made when there was an absolute necessity for producing some result. . . . No one goes to certain *348inevitable death except by compulsion. Put the matter the other way. Was there ever yet a criminal who, when sentenced to death and brought out to die, would refuse the offer of a commutation of his sentence for the severest secondary punishment? Surely not. Why is this? It can only be because 'All that a man has will he give for his life.’ In any secondary punishment, however terrible, there is hope; but death is death; its terrors cannot be described more forcibly.” 95

This hypothesis relates to the use of capital punishment as a deterrent for any crime. The second proposition is that "if life imprisonment is the maximum penalty for a crime such as murder, an offender who is serving a life sentence cannot then be deterred from murdering a fellow inmate or a prison officer.”96 This hypothesis advocates a limited deterrent effect under particular circumstances.

Abolitionists attempt to disprove these hypotheses by amassing statistical evidence to demonstrate that there is no correlation between criminal activity and the existence or nonexistence of a capital sanction. Almost all of the evidence involves the crime of murder, since murder is punishable by death in more jurisdictions than are other offenses,97 and almost 90%. of all executions since 1930 have been pursuant to murder convictions.98

Thorsten Sellin, one of the leading authorities on capital punishment, has urged that if the death penalty *349deters prospective murderers, the following hypotheses should be true:

“(a) Murders should be less frequent in states that have the death penalty than in those that have abolished it, other factors being equal. Comparisons of this nature must be made among states that are as alike as possible in all other respects— character of population, social and economic condition, etc. — in order not to introduce factors known to influence murder rates in a serious manner but present in only one of these states.

“(b) Murders should increase when the death penalty is abolished and should decline when it is restored.

“(c) The deterrent effect should be greatest and should therefore affect murder rates most powerfully in those communities where the crime occurred and its consequences are most strongly brought home to the population.

“(d) Law enforcement officers would be safer from murderous attacks in states that have the death penalty than in those without it.”99 (Footnote omitted.)

Sellin’s evidence indicates that not one of these propositions is true. This evidence has its problems, however. One is that there are no accurate figures for capital murders; there are only figures on homicides and they, of course, include noncapital killings.100 A second problem is that certain murders undoubtedly are misinterpreted as accidental deaths or suicides, and there *350is no way of estimating the number of such undetected crimes. A third problem is that not all homicides are reported. Despite these difficulties, most authorities have assumed that the proportion of capital murders in a State’s or nation’s homicide statistics remains reasonably constant,101 and that the homicide statistics are therefore useful.

Sellin’s statistics demonstrate that there is no correlation between the murder rate and the presence or absence of the capital sanction. He compares States that have similar characteristics and finds that irrespective of their position on capital punishment, they have similar murder rates. In the New England States, for example, there is no correlation between executions102 and homicide rates.103 The same is true for Midwestern States,104 and for all others studied. Both the United Nations105 and Great Britain 100 have acknowledged the validity of Sellin’s statistics.

Sellin also concludes that abolition and/or reintroduction of the death penalty had no effect on the homicide rates of the various States involved.107 This conclusion is borne out by others who have made similar *351inquiries108 and by the experience of other countries.109 Despite problems with the statistics,110 Sellin’s evidence has been relied upon in international studies of capital punishment.111

Statistics also show that the deterrent effect of capital punishment is no greater in those communities where executions take place than in other communities.112 In fact, there is some evidence that imposition of capital punishment may actually encourage crime, rather than deter it.113 And, while police and law enforcement offi*352cers are the strongest advocates of capital punishment,114 the evidence is overwhelming that police are no safer in communities that retain the sanction than in those that have abolished it.115

There is also a substantial body of data showing that the existence of the death penalty has virtually no effect on the homicide rate in prisons.110 Most of the persons sentenced to death are murderers, and murderers tend to be model prisoners.117

*353In sum, the only support for the theory that capital punishment is an effective deterrent is found in the hypotheses with which we began and the occasional stories about a specific individual being deterred from doing a contemplated criminal act.118 These claims of specific deterrence are often spurious,119 however, and may be more than counterbalanced by the tendency of capital punishment to incite certain crimes.120

The United Nations Committee that studied capital punishment found that “'[i]t is generally agreed between the retentionists and abolitionists, whatever their opinions about the validity of comparative studies of deterrence, that the data which now exist show no correlation between the existence of capital punishment and lower rates of capital crime.”121

Despite the fact that abolitionists have not proved non-deterrence beyond a reasonable doubt, they have succeeded in showing by clear and convincing evidence that capital punishment is not necessary as a deterrent to crime in our society. This is all that they must do. We would shirk our judicial responsibilities if we failed to accept the presently existing statistics and demanded more proof. It may be that we now possess all the proof that anyone could ever hope to assemble on the subject. But, even if further proof were to be forthcoming, I believe there is more than enough evidence presently available for a decision in this case.

In 1793 William Bradford studied the utility of the death penalty in Pennsylvania and found that it probably had no deterrent effect but that more evidence *354was needed.122 Edward Livingston reached a similiar conclusion with respect to deterrence in 1833 upon completion of his study for Louisiana.123 Virtually every study that has since been undertaken has reached the same result.124

In light of the massive amount of evidence before us, I see no alternative but to conclude that capital punishment cannot be justified on the basis of its deterrent effect.125

*355C. Much of what must be said about the death penalty as a device to prevent recidivism is obvious — if a murderer is executed, he cannot possibly commit another offense. The fact is, however, that murderers are extremely unlikely to commit other crimes either in prison or upon their release.126 For the most part, they are first offenders, and when released from prison they are known to become model citizens.127 Furthermore, most persons who commit capital crimes are not executed. With respect to those who are sentenced to die, it is critical to note that the jury is never asked to determine whether they are likely to be recidivists. In light of these facts, if capital punishment were justified purely on the basis of preventing recidivism, it would have to be considered to be excessive; no general need to obliterate all capital offenders could have been demonstrated, nor any specific need in individual cases.

D. The three final purposes which may underlie utilization of a capital sanction — encouraging guilty pleas and confessions, eugenics, and reducing state "expenditures — may be dealt with quickly. If the death penalty is used to encourage guilty pleas and thus to deter suspects from exercising their rights under the Sixth Amendment to jury trials, it is unconstitutional. United States *356v. Jackson, 390 U. S. 570 (1968).128 Its elimination would do little to impair the State’s bargaining position in criminal cases, since life imprisonment remains a severe sanction which can be used as leverage for bargaining for pleas or confessions in exchange either for charges of lesser offenses or recommendations of leniency.

Moreover, to the extent that capital punishment is used to encourage confessions and guilty pleas, it is not being used for punishment purposes. A State that justifies capital punishment on its utility as part of the conviction process could not profess to rely on capital punishment as a deterrent. Such a State’s system would be structured with twin goals only: obtaining guilty pleas and confessions and imposing imprisonment as the maximum sanction. Since life imprisonment is sufficient for bargaining purposes, the death penalty is excessive if used for the same purposes.

In light of the previous discussion on deterrence, any suggestions concerning the eugenic benefits of capital punishment are obviously meritless.129 As I pointed out above, there is not even any attempt made to discover which capital offenders are likely to be recidivists, let alone which are positively incurable. No test or procedure presently exists by which incurables can be screened from those who would benefit from treatment. On the one hand, due process would seem to require that we have some procedure to demonstrate incurability before execution; and, on the other hand, equal protection would then seemingly require that all incurables be executed, cf. Skinner v. Oklahoma, 316 U. S. 535 (1942). In addition, the “cruel and unusual” language *357would require that life imprisonment, treatment, and sterilization be inadequate for eugenic purposes. More importantly, this Nation has never formally professed eugenic goals, and the history of the world does not look kindly on them. If eugenics is one of our purposes, then the legislatures should say so forthrightly and design procedures to serve this goal. Until such time, I can only conclude, as has virtually everyone else who has looked at the problem,130 that capital punishment cannot be defended on the basis of any eugenic purposes.

As for the argument that it is cheaper to execute a capital offender than to imprison him for life, even assuming that such an argument, if true, would support. a capital sanction, it is simply incorrect. A disproportionate amount of money spent on prisons is attributable to death row.131 Condemned men are not productive members of the prison community, although they could be,132 and executions are expensive.133 Appeals are often automatic, and courts admittedly spend more time with death cases.134

*358At trial, the selection of jurors is likely to become a costly, time-consuming problem in a capital case,135 and defense counsel will reasonably exhaust every possible means to save his client from execution, no matter how long the trial takes.

During the period between conviction and execution, there are an inordinate number of collateral attacks on the conviction and attempts to obtain executive clemency, all of which exhaust the time, money, and effort of the State. There are also continual assertions that the condemned prisoner has gone insane.136 Because there is a formally established policy of not executing insane persons,137 great sums of money may be spent on detecting and curing mental illness in order to perform the execution.138 Since no one wants the responsibility for the execution, the condemned man is likely to be passed back and forth from doctors to custodial officials to courts like a ping-pong ball.139 The entire process is very costly.

When all is said and done, there can be no doubt that it costs more to execute a man than to keep him in prison for life.140

E. There is but one conclusion that can be drawn from all of this — i. e., the death penalty is an excessive and unnecessary punishment that violates the Eighth *359Amendment. The statistical evidence is not convincing beyond all doubt, but it is persuasive. It is not improper at this point to take judicial notice of the fact that for more than 200 years men have labored to demonstrate that capital punishment serves no purpose that life imprisonment could not serve equally well. And they have done so with great success. Little, if any, evidence has been adduced to prove the contrary. The point has now been reached at which deference to the legislatures is tantamount to abdication of our judicial roles as factfinders, judges, and ultimate arbiters of the Constitution. We know that at some point the presumption of constitutionality accorded legislative acts gives way to a realistic assessment of those acts. This point comes when there is sufficient evidence available so that judges can determine, not whether the legislature acted wisely, but whether it had any rational basis whatsoever for acting. We have this evidence before us now. There is no rational basis for concluding that capital punishment is not excessive. It therefore violates the Eighth Amendment.141

*360VI

In addition, even if capital punishment is not excessive, it nonetheless violates the Eighth Amendment because it is morally unacceptable to the people of the United States at this time in their history.

In judging whether or not a given penalty is morally acceptable, most courts have said that the punishment is valid unless “it shocks the conscience and sense of justice of the people.” 142

*361Judge Frank once noted the problems inherent in the use of such a measuring stick:

“[The court,] before it reduces a sentence as 'cruel and unusual/ must have reasonably good assurances that the sentence offends the 'common conscience.’ And, in any context, such a standard — the community’s attitude — is usually an unknowable. It resembles a slithery shadow, since one can seldom learn, at all accurately, what the community, or a majority, actually feels. Even a carefully-taken 'public opinion poll’ would be inconclusive in a case like this.” 143

While a public opinion poll obviously is of some assistance in indicating public acceptance or rejection of a specific penalty,144 its utility cannot be very great. This is because whether or not a punishment is cruel and unusual depends, not on whether its mere mention “shocks the conscience and sense of justice of the people,” but on whether people who were fully informed as to the purposes of the penalty and its liabilities would find the penalty shocking, unjust, and unacceptable.145

*362In other words, the question with which we must deal is not whether a substantial proportion of American citizens would today, if polled, opine that capital punishment is barbarously cruel, but whether they would find it to be so in the light of all information presently available.

This is not to suggest that with respect to this test of unconstitutionality people are required to act rationally; they are not. With respect to this judgment, a violation of the Eighth Amendment is totally dependent on the predictable subjective, emotional reactions of informed citizens.146

It has often been noted that American citizens know almost nothing about capital punishment.147 Some of the conclusions arrived at in the preceding section and the supporting evidence would be critical to an informed judgment on the morality of the death penalty: e. g., that the death penalty is no more effective a deterrent than life imprisonment, that convicted murderers are *363rarely executed, but are usually sentenced to a term in prison; that convicted murderers usually are model prisoners, and that they almost always become law-abiding citizens upon their release from prison; that the costs of executing a capital offender exceed the costs of imprisoning him for life; that while in prison, a convict under sentence of death performs none of the useful functions that life prisoners perform; that no attempt is made in the sentencing process to ferret out likely recidivists for execution; and that the death penalty may actually stimulate criminal activity.

This information would almost surely convince the average citizen that the death penalty was unwise, but a problem arises as to whether it would convince him that the penalty was morally reprehensible. This problem arises from the fact that the public’s desire for retribution, even though this is a goal that the legislature cannot constitutionally pursue as its sole justification for capital punishment, might influence the citizenry’s view of the morality of capital punishment. The solution to the problem lies in the fact that no one has ever seriously advanced retribution as a legitimate goal of our society. Defenses of capital punishment are always mounted on deterrent or other similar theories. This should not be surprising. It is the people of this country who have urged in the past that prisons rehabilitate as well as isolate offenders, and it is the people who have injected a sense of purpose into our penology. I cannot believe that at this stage in our history, the American people would ever knowingly support purposeless vengeance. Thus, I believe that the great mass of citizens would conclude on the basis of the material already considered that the death penalty is immoral and therefore unconstitutional.

But, if this information needs supplementing, I believe that the following facts would serve to convince *364even the most hesitant of citizens to condemn death as a sanction: capital punishment is imposed discrim-inatorily against certain identifiable classes of people; there is evidence that innocent people have been executed before their innocence can be proved; and the death penalty wreaks havoc with our entire criminal justice system. Each of these facts is considered briefly below.

Regarding discrimination, it has been said that “[i]t is usually the poor, the illiterate, the underprivileged, the member of the minority group — the man who, because he is without means, and is defended by a court-appointed attorney — who becomes society’s sacrificial lamb 148 Indeed, a look at the bare statistics regarding executions is enough to betray much of the discrimination. A total of 3,859 persons have been executed since 1930, of whom 1,751 were white and 2,066 were Negro.149 Of the executions, 3,334 were for murder; 1,664 of the executed murderers were white and 1,630 were Negro; 150 455 persons, including 48 whites and 405 Negroes, were executed for rape.151 It is immediately apparent that Negroes were executed far more often than whites in proportion to their percentage of the population. Studies indicate that while the higher rate of execution among Negroes is partially due to a higher rate of crime, there is evidence of racial discrimination.152 *365Racial or other discriminations should not be surprising. In McGautha v. California, 402 U. S., at 207, this Court held “that committing to the untrammeled discretion of the jury the power to pronounce life or death in capital cases is [not] offensive to anything in the Constitution.” This was an open invitation to discrimination.

There is also overwhelming evidence that the death penalty is employed against men and not women. Only 32 women have been executed since 1930, while 3,827 men have met a similar fate.153 It is difficult to understand why women have received such favored treatment since the purposes allegedly served by capital punishment seemingly are equally applicable to both sexes.154

It also is evident that the burden of capital punishment falls upon the poor, the ignorant, and the under*366privileged members of society.155 It is the poor, and the members of minority groups who are least able to voice their complaints against capital punishment. Their impotence leaves them victims of a sanction that the wealthier, better-represented, just-as-guilty person can escape. So long as the capital sanction is used only against the forlorn, easily forgotten members of society, legislators are content to maintain the status quo, because change would draw attention to the problem and concern might develop. Ignorance is perpetuated and apathy soon becomes its mate, and we have today’s situation.

Just as Americans know little about who is executed and why, they are unaware of the potential dangers of executing an innocent man. Our “beyond a reasonable doubt” burden of proof in criminal cases is intended to protect the innocent, but we know it is not foolproof. Various studies have shown that people whose innocence is later convincingly established are convicted and sentenced to death.156

*367Proving one’s innocence after a jury finding of guilt is almost impossible. While reviewing courts are willing to entertain all kinds of collateral attacks where a sentence of death is involved, they very rarely dispute the jury’s interpretation of the evidence. This is, perhaps, as it should be. But, if an innocent man has been found guilty, he must then depend on the good faith of the prosecutor’s office to help him establish his innocence. There is evidence, however, that prosecutors do not welcome the idea of having convictions, which they labored hard to secure, overturned, and that their cooperation is highly unlikely.157

No matter how careful courts are, the possibility of perjured testimony, mistaken honest testimony, and human error remain all too real.158 We have no way of *368judging how many innocent persons have been executed but we can be certain that there were some. Whether there were many is an open question made difficult by the loss of those who were most knowledgeable about the crime for which they were convicted. Surely there will be more as long as capital punishment remains part of our penal law.

While it is difficult to ascertain with certainty the degree to which the death penalty is discriminatorily imposed or the number of innocent persons sentenced to die, there is one conclusion about the penalty that is universally accepted — i. e., it “tends to distort the course of the criminal law.”159 As Mr. Justice Frankfurter said:

“I am strongly against capital punishment . . . . When life is at hazard in a trial, it sensationalizes the whole thing almost unwittingly; the effect on juries, the Bar, the public, the Judiciary, I regard as very bad. I think scientifically the claim of deterrence is not worth much. Whatever proof there may be in my judgment does not outweigh the social loss due to the inherent sensationalism of a trial for fife.” 160

*369The deleterious effects of the death penalty are also felt otherwise than at trial. For example, its very existence “inevitably sabotages a social or institutional program of reformation.” 161 In short “;[t]he presence of the death penalty as the keystone of our penal system bedevils the administration of criminal justice all the way down the line and is the stumbling block in the path of general reform and of the treatment of crime and criminals.” 162

Assuming knowledge of all the facts presently available regarding capital punishment, the average citizen would, in my opinion, find it shocking to his conscience and sense of justice.163 For this reason alone capital punishment cannot stand.

*370YII

To arrive at the conclusion that the death penalty violates the Eighth Amendment, we have had to engage in a long and tedious journey. The amount of information that we have assembled and sorted is enormous. *371Yet, I firmly believe that we have not deviated in the slightest from the principles with which we began.

At a time in our history when the streets of the Nation’s cities inspire fear and despair, rather than pride and hope, it is difficult to maintain objectivity and concern for our fellow citizens. But, the measure of a country’s greatness is its ability to retain compassion in time of crisis. No nation in the recorded history of man has a greater tradition of revering justice and fair treatment for all its citizens in times of turmoil, confusion, and tension than ours. This is a country which stands tallest in troubled times, a country that clings to fundamental principles, cherishes its constitutional heritage, and rejects simple solutions that compromise the values that lie at the roots of our democratic system.

In striking down capital punishment, this Court does not malign our system of government. On the contrary, it pays homage to it. Only in a free society could right triumph in difficult times, and could civilization record its magnificent advancement. In recognizing the humanity of our fellow beings, we pay ourselves the highest tribute. We achieve “a major milestone in the long road up from barbarism” 164 and join the approximately 70 other jurisdictions in the world which celebrate their regard for civilization and humanity by shunning capital punishment.165

I concur in the judgments of the Court.

[Appendices I, II, and III follow.]

*372APPENDIX I TO OPINION OF MARSHALL, J., CONCURRING

ABOLITION OF THE DEATH PENALTY IN THE UNITED STATES: 1846-1968

(States are listed according to year most recent action was taken)

Year of Year of restoration reabolition State Year of partial abolition Year of complete abolition

New York. 19651

Vermont. 19652

West Virginia. — 1965

Iowa . — 1872 1 — l co 05 C71 00 CO 1 — 1

Oregon . — 1914 l_* co 05 o <M 05 1 — Í

Michigan . 18473 1963

1961 Delaware . — 1958

Alaska . — 1957

Hawaii . — 1957

South Dakota. — Kansas . — Missouri. — Tennessee. 1915 4_ Washington . — Arizona . 19165 North Dakota. 19156 1915 1907 1917 1913 SO CO CO CO C0 CO MMMMWCO GO co co co C* CO

Minnesota . — 1911

1883 1887 Colorado. — Maine. — Wisconsin . — 1897 1876 1853 o 05

Rhode Island. 18527

*373APPENDIX II TO OPINION OF MARSHALL, J., CONCURRING

CRUDE HOMICIDE DEATH RATES, PER 100,000 POPULATION, AND NUMBER OF EXECUTIONS IN CERTAIN AMERICAN STATES: 1920-1955

*374APPENDIX III TO OPINION OF MARSHALL, J., CONCURRING

CRUDE HOMICIDE DEATH RATES, PER 100,000 POPULATION, AND NUMBER OF EXECUTIONS IN CERTAIN AMERICAN STATES: 1920-1955

*375Mr. Chief Justice Burger,

with whom Mr. Justice Blackmun, Mr. Justice Powell, and Mr. Justice Rehnquist join, dissenting.

At the outset it is important to note that only two members of the Court, Mr. Justice Brennan and Mr. Justice Marshall, have concluded that the Eighth Amendment prohibits capital punishment for all crimes and under all circumstances. Mr. Justice Douglas has also determined that the death penalty contravenes the Eighth Amendment, although I do not read his opinion as necessarily requiring final abolition of the penalty.1 For the reasons set forth in Parts I-IV of this opinion, I conclude that the constitutional prohibition against “cruel and unusual punishments” cannot be construed to bar the imposition of the punishment of death.

Mr. Justice Stewart and Mr. Justice White have concluded that petitioners’ death sentences must be set aside because prevailing sentencing practices do not comply with the Eighth Amendment. For the reasons set forth in Part V of this opinion, I believe this approach fundamentally misconceives the nature of the Eighth Amendment guarantee and flies directly in the face of controlling authority of extremely recent vintage.

I

If we were possessed of legislative power, I would either join with Mr. Justice Brennan and Mr. Justice Marshall or, at the very least, restrict the use of capital punishment to a small category of the most heinous crimes. Our constitutional inquiry, however, must be divorced from personal feelings as to the morality and efficacy of the death penalty, and be confined to the meaning and applicability of the uncertain language of the Eighth Amendment. There is no novelty in being called upon to interpret a constitutional provision that is less than *376self-defining, but, of all our fundamental guarantees, the ban on “cruel and unusual punishments” is one of the most difficult to translate into judicially manageable terms. The widely divergent views of the Amendment expressed in today’s opinions reveal the haze that surrounds this constitutional command. Yet it is essential to our role as a court that we not seize upon the enigmatic character of the guarantee as an invitation to enact our personal predilections into law.

Although the Eighth Amendment literally reads as prohibiting only those punishments that are both “cruel” and “unusual,” history compels the conclusion that the Constitution prohibits all punishments of extreme and barbarous cruelty, regardless of how frequently or infrequently imposed.

The most persuasive analysis of Parliament’s adoption of the English Bill of Rights of 1689 — the unquestioned source of the Eighth Amendment wording— suggests that the prohibition against “cruel and unusual punishments” was included therein out of aversion to severe punishments not legally authorized and not within the jurisdiction of the courts to impose. To the extent that the term “unusual” had any importance in the English version, it was apparently intended as a reference to illegal punishments.2

*377From every indication, the Framers of the Eighth Amendment intended to give the phrase a meaning far different from that of its English precursor. The records of the debates in several of the state conventions called to ratify the 1789 draft Constitution submitted prior to the addition of the Bill of Rights show that the Framers’ exclusive concern was the absence of any ban on tortures.3 The later inclusion of the “cruel and unusual punishments” clause was in response to these objections. There was no discussion of the interrelationship of the terms “cruel” and “unusual,” and there is nothing in the debates supporting the inference that the Founding Fathers would have been receptive to torturous or excessively cruel punishments even if usual in character or authorized by law.

The cases decided under the Eighth Amendment are consistent with the tone of the ratifying debates. In Wilkerson v. Utah, 99 U. S. 130 (1879), this Court held that execution by shooting was not a prohibited mode of carrying out a sentence of death. Speaking to the mean*378ing of the Cruel and Unusual Punishments Clause, the Court stated,

“[I]t is safe to affirm that punishments of torture . . . and all others in the same line of unnecessary cruelty, are forbidden by that amendment to the Constitution.” Id., at 136.

The Court made no reference to the role of the term “unusual” in the constitutional guarantee.

In the case of In re Kemmler, 136 U. S. 436 (1890), the Court held the Eighth Amendment inapplicable to the States and added the following dictum:

“So that, if the punishment prescribed for an of-fence against the laws of the State were manifestly cruel and unusual, as burning at the stake, crucifixion, breaking on the wheel, or the like, it would be the duty of the courts to adjudge such penalties to be within the . . . [prohibition of the New York constitution]. And we think this equally true of the Eighth Amendment, in its application to Congress.

“. . . Punishments are cruel when they involve torture or a lingering death; but the punishment of death is not cruel, within the meaning of that word as used in the Constitution. It implies there something inhuman and barbarous, something more than the mere extinguishment of life.” Id., at 446-447.

This language again reveals an exclusive concern with extreme cruelty. The Court made passing reference to the finding of the New York courts that electrocution was an “unusual” punishment, but it saw no need to discuss the significance of that term as used in the Eighth Amendment.

Opinions in subsequent cases also speak of extreme cruelty as though that were the sum and substance of the constitutional prohibition. See O’Neil v. Vermont, 144 U. S. 323, 339-340 (1892) (Field, J., dissenting); Weems *379v. United States, 217 U. S. 349, 372-373 (1910); Louisiana ex rel. Francis v. Resweber, 329 U. S. 459, 464 (1947). As summarized by Mr. Chief Justice Warren in the plurality opinion in Trop v. Dulles, 356 U. S. 86, 100 n. 32 (1958):

“Whether the word ‘unusual’ has any qualitative meaning different from ‘cruel’ is not clear. On the few occasions this Court has had to consider the meaning of the phrase, precise distinctions between cruelty and unusualness do not seem to have been drawn. See Weems v. United States, supra; O’Neil v. Vermont, supra; Wilkerson v. Utah, supra. These cases indicate that the Court simply examines the particular punishment involved in light of the basic prohibition against inhuman treatment, without regard to any subtleties of meaning that might be latent in the word ‘unusual.’ ”

I do not suggest that the presence of the word “unusual” in the Eighth Amendment is merely vestigial, having no relevance to the constitutionality of any punishment that might be devised. But where, as here, we consider a punishment well known to history, and clearly authorized by legislative enactment, it disregards the history of the Eighth Amendment and all the judicial comment that has followed to rely on the term “unusual” as affecting the outcome of these cases. Instead, I view these cases as turning on the single question whether capital punishment is “cruel” in the constitutional sense. The term “unusual” cannot be read as limiting the ban on “cruel” punishments or as somehow expanding the meaning of the term “cruel.” For this reason I am unpersuaded by the facile argument that since capital punishment has always been cruel in the everyday sense of the word, and has become unusual due to decreased use, it is, therefore, now “cruel and unusual.”

*380II

Counsel for petitioners properly concede that capital punishment was not impermissibly cruel at the time of the adoption of the Eighth Amendment. Not only do the records of the debates indicate that the Founding Fathers were limited in their concern to the prevention of torture, but it is also clear from the language of the Constitution itself that there was no thought whatever of the elimination of capital punishment. The opening sentence of the Fifth Amendment is a guarantee that the death penalty not be imposed “unless on a presentment or indictment of a Grand Jury.” The Double Jeopardy Clause of the Fifth Amendment is a prohibition against being “twice put in jeopardy of life” for the same offense. Similarly, the Due Process Clause commands “due process of law” before an accused can be “deprived of life, liberty, or property.” Thus, the explicit language of the Constitution affirmatively acknowledges the legal power to impose capital punishment; it does not expressly or by implication acknowledge the legal power to impose any of the various punishments that have been banned as cruel since 1791. Since the Eighth Amendment was adopted on the same day in 1791 as the Fifth Amendment, it hardly needs more to establish that the death penalty was not “cruel” in the constitutional sense at that time.

In the 181 years since the enactment of the Eighth Amendment, not a single decision of this Court has cast the slightest shadow of a doubt on the constitutionality of capital punishment. In rejecting Eighth Amendment attacks on particular modes of execution, the Court has more than once implicitly denied that capital punishment is impermissibly “cruel” in the constitutional sense. Wilkerson v. Utah, 99 U. S. 130 (1879); Louisiana ex rel. Francis v. Resweber, 329 U. S., at 464. In *381re Kemmler, 136 U. S. 436 (1890) (dictum). It is only 14 years since Mr. Chief Justice Warren, speaking for four members of the Court, stated without equivocation:

“Whatever the arguments may be against capital punishment, both on moral grounds and in terms of accomplishing the purposes of punishment — -and they are forceful — the death penalty has been employed throughout our history, and, in a day when it is still widely accepted, it cannot be said to violate the constitutional concept of cruelty.” Trop v. Dulles, 356 U. S., at 99.

It is only one year since Mr. Justice Black made his feelings clear on the constitutional issue:

“The Eighth Amendment forbids 'cruel and unusual punishments.’ In my view, these words cannot be read to outlaw capital punishment because that penalty was in common use and authorized by law here and in the countries from which our ancestors came at the time the Amendment was adopted. It is inconceivable to me that the framers intended to end capital punishment by the Amendment.” Mc-Gautha v. California, 402 U. S. 183, 226 (1971) (separate opinion).

By limiting its grants of certiorari, the Court has refused even to hear argument on the Eighth Amendment claim on two occasions in the last four years. Witherspoon v. Illinois, cert. granted, 389 U. S. 1035, rev’d, 391 U. S. 510 (1968); McGautha v. California, cert. granted, 398 U. S. 936 (1970), aff’d, 402 U. S. 183 (1971). In these cases the Court confined its attention to the procedural aspects of capital trials, it being implicit that the punishment itself could be constitutionally imposed. Nonetheless, the Court has now been asked to hold that a punishment clearly permissible under the Constitution at the time of its adoption and accepted as such by every *382member of the Court until today, is suddenly so cruel as to be incompatible with the Eighth Amendment.

Before recognizing such an instant evolution in the law, it seems fair to ask what factors have changed that capital punishment should now be “cruel” in the constitutional sense as it has not been in the past. It is apparent that there has been no change of constitutional significance in the nature of the punishment itself. Twentieth century modes of execution surely involve no greater physical suffering than the means employed at the time of the Eighth Amendment’s adoption. And although a man awaiting execution must inevitably experience extraordinary mental anguish4 no one suggests that this anguish is materially different from that experienced by condemned men in 1791, even though protracted appellate review processes have greatly increased the waiting time on “death row.” To be sure, the ordeal of the condemned man may be thought cruel in the sense that all suffering is thought cruel. But if the Constitution proscribed every punishment producing severe emotional stress, then capital punishment would clearly have been impermissible in 1791.

However, the inquiry cannot end here. For reasons unrelated to any change in intrinsic cruelty, the Eighth Amendment prohibition cannot fairly be limited to those punishments thought excessively cruel and barbarous at the time of the adoption of the Eighth Amendment. A punishment is inordinately cruel, in the sense we must deal with it in these cases, chiefly as perceived by the society so characterizing it. The standard of extreme cruelty is not merely descriptive, but necessarily embodies a moral judgment. The standard itself remains the same, but its applicability must change as the basic mores of society change. This notion is not *383new to Eighth Amendment adjudication. In Weems v. United States, 217 U. S. 349 (1910), the Court referred with apparent approval to the opinion of the commentators that “[t]he clause of the Constitution . . . may be therefore progressive, and is not fastened to the obsolete but may acquire meaning as public opinion becomes enlightened by a humane justice.” 217 U. S., at 378. Mr. Chief Justice Warren, writing the plurality opinion in Trop v. Dulles, supra, stated, “The Amendment must fidraw its meaning from the evolving standards of decency that mark the progress of a maturing society.” 356 U. S., at 101. Nevertheless, the Court up to now has never actually held that a punishment has become impermis-sibly cruel due to a shift in the weight of accepted social values; nor has the Court suggested judicially manageable criteria for measuring such a shift in moral consensus.

The Court’s quiescence in this area can be attributed to the fact that in a democratic society legislatures, not courts, are constituted to respond to the will and consequently the moral values of the people. For this reason, early commentators suggested that the “cruel and unusual punishments” clause was an unnecessary constitutional provision.5 As acknowledged in the principal brief for petitioners, “both in constitutional contemplation and in fact, it is the legislature, not the Court, which responds to public opinion and immediately reflects the society’s standards of decency.”6 *384Accordingly, punishments such as branding and the cutting off of ears, which were commonplace at the time of the adoption of the Constitution, passed from the penal scene without judicial intervention because they became basically offensive to- the people and the legislatures responded to this sentiment.

Beyond any doubt, if we were today called upon to review such punishments, we would find them excessively cruel because we could say with complete assurance that contemporary society universally rejects such bizarre penalties. However, this speculation on the Court's probable reaction to such punishments is not of itself significant. The critical fact is that this Court has never had to hold that a mode of punishment authorized by a domestic legislature was so cruel as to be fundamentally at odds with our basic notions of decency. Cf. Weems v. United States, supra. Judicial findings of impermissible cruelty have been limited, for the most part, to offensive punishments devised without specific authority by prison officials, not by legislatures. See, e. g., Jackson v. Bishop, 404 F. 2d 571 (CA8 1968); Wright v. McMann, 387 F. 2d 519 (CA2 1967). The paucity of judicial decisions invalidating legislatively prescribed punishments is powerful evidence that in this country legislatures have in fact been responsive — albeit belatedly at times — to changes in social attitudes and moral values.

I do not suggest that the validity of legislatively authorized punishments presents no justiciable issue under the Eighth Amendment, but, rather, that the primacy of the legislative role narrowly confines the scope of judicial inquiry. Whether or not provable, and whether or not true at all times, in a democracy the legislative judgment is presumed to embody the basic standards of decency prevailing in the society. This presumption can only be negated by unambiguous and compelling evidence of legislative default.

*385III

There are no obvious indications that capital punishment offends the conscience of society to such a degree that our traditional deference to the legislative judgment must be abandoned. It is not a punishment such as burning at the stake that everyone would ineffably find to be repugnant to all civilized standards. Nor is it a punishment so roundly condemned that only a few aberrant legislatures have retained it on the statute books. Capital punishment is authorized by statute in 40 States, the District of Columbia, and in the federal courts for the commission of certain crimes.7 On four occasions in the last 11 years Congress has added to the list of federal crimes punishable by death.8 In looking for reliable indicia of contemporary attitude, none more trustworthy has been advanced.

One conceivable source of evidence that legislatures have abdicated their essentially barometric role with respect to community values would be public opinion polls, of which there have been many in the past decade addressed to the question of capital punishment. Without assessing the reliability of such polls, or intimating that any judicial reliance could ever be placed on them, *386it need only be noted that the reported results have shown nothing approximating the universal condemnation of capital punishment that might lead us to suspect that the legislatures in general have lost touch with current social values.9

Counsel for petitioners rely on a different body of empirical evidence. They argue, in effect, that the number of cases in which the death penalty is imposed, as compared with the number of cases in which it is statutorily available, reflects a general revulsion toward the penalty that would lead to its repeal if only it were more generally and widely enforced. It cannot be gainsaid that by the choice of juries — and sometimes judges10— the death penalty is imposed in far fewer than half the cases in which it is available.11 To go further and char*387acterize the rate of imposition as “freakishly rare,” as petitioners insist, is unwarranted hyperbole. And regardless of its characterization, the rate of imposition does not impel the conclusion that capital punishment is now regarded as intolerably cruel or uncivilized.

It is argued that in those capital cases where juries have recommended mercy, they have given expression to civilized values and effectively renounced the legislative authorization for capital punishment. At the same time it is argued that where juries have made the awesome decision to send men to their deaths, they have acted arbitrarily and without sensitivity to prevailing standards of decency. This explanation for the infrequency of imposition of capital punishment is unsupported by known facts, and is inconsistent in principle with everything this Court has ever said about the functioning of juries in capital eases.

In McGautha v. California, supra, decided only one year ago, the Court held that there was no mandate in the Due Process Clause of the Fourteenth Amendment that juries be given instructions as to when the death penalty should be imposed. After reviewing the autonomy that juries have traditionally exercised in capital cases and noting the practical difficulties of framing manageable instructions, this Court concluded that judicially articulated standards were not needed to insure a responsible decision as to penalty. Nothing in McGautha licenses capital juries to act arbitrarily or assumes that they have so acted in the past. On the contrary, the assumption underlying the McGautha ruling is that juries “will act with *388due regard for the consequences of their decision.” 402 U. S., at 208.

The responsibility of juries deciding capital cases in our system of justice was nowhere better described than in Witherspoon v. Illinois, supra:

“[A] jury that must choose between life imprisonment and capital punishment can do little more— and must do nothing less — than express the con science of the community on the ultimate question of life or death.”

“And one of the most important functions any jury can perform in making such a selection is to maintain a link between contemporary community values and the penal system — a link without which the determination of punishment could hardly reflect ‘the evolving standards of decency that mark the progress of a maturing society' ” 391 U. S., at 519 and n. 15 (emphasis added).

The selectivity of juries in imposing the punishment of death is properly viewed as a refinement on, rather than a repudiation of, the statutory authorization for that penalty. Legislatures prescribe the categories of crimes for which the death penalty should be available, and, acting as “the conscience of the community,” juries are entrusted to determine in individual cases that the ultimate punishment is warranted. Juries are undoubtedly influenced in this judgment by myriad factors. The motive or lack of motive of the perpetrator, the degree of injury or suffering of the victim or victims, and the degree of brutality in the commission of the crime would seem to be prominent among these factors. Given the general awareness that death is no longer a routine punishment for the crimes for which it is made available, it is hardly surprising that juries have been increasingly meticulous in their imposition of the penalty. But to *389assume from the mere fact of relative infrequency that only a random assortment of pariahs are sentenced to death, is to cast grave doubt on the basic integrity of our jury system.

It would, of course, be unrealistic to assume that juries have been perfectly consistent in choosing the cases where the death penalty is to be imposed, for no human institution performs with perfect consistency. There are doubtless prisoners on death row who would not be there had they been tried before a different jury or in a different State. In this sense their fate has been controlled by a fortuitous circumstance. However, this element of fortuity does not stand as an indictment either of the general functioning of juries in capital cases or of the integrity of jury decisions in individual cases. There is no empirical basis for concluding that juries have generally failed to discharge in good faith the responsibility described in Witherspoon — that of choosing between life and death in individual cases according to the dictates of community values.12

*390The rate of imposition of death sentences falls far short of providing the requisite unambiguous evidence that the legislatures of 40 States and the Congress have turned their backs on current or evolving standards of decency in continuing to make the death penalty available. For, if selective imposition evidences a rejection of capital punishment in those cases where it is not imposed, it surely evidences a correlative affirmation of the penalty in those cases where it is imposed. Absent some clear indication that the continued imposition of the death penalty on a selective basis is violative of prevailing standards of civilized conduct, the Eighth Amendment cannot be said to interdict its use.

*391In two of these cases we have been asked to rule on the narrower question whether capital punishment offends the Eighth Amendment when imposed as the punishment for the crime of forcible rape.13 It is true that the death penalty is authorized for rape in fewer States than it is for murder,14 and that even in those States it is applied more sparingly for rape than for murder.15 But for the reasons aptly brought out in the opinion of Mr. Justice Powell, post, at 456-461, I do not believe these differences can be elevated to the level of an Eighth Amendment distinction. This blunt constitutional command cannot be sharpened to carve neat distinctions corresponding to the categories of crimes defined by the legislatures.

IV

Capital punishment has also been attacked as viola-tive of the Eighth Amendment on the ground that it is not needed to achieve legitimate penal aims and is thus “unnecessarily cruel.” As a pure policy matter, this approach has much to recommend it, but it seeks to give a dimension to the Eighth Amendment that it was' never intended to have and promotes a line of inquiry that this Court has never before pursued.

The Eighth Amendment, as I have noted, was included in the Bill of Rights to guard against the use of torturous and inhuman punishments, not those of limited efficacy. One of the few to speak out against the adop*392tion of the Eighth Amendment asserted that it is often necessary to use cruel punishments to deter crimes.16 But among those favoring the Amendment, no sentiment was expressed that a punishment of extreme cruelty could ever be justified by expediency. The dominant theme of the Eighth Amendment debates was that the ends of the criminal laws cannot justify the use of measures of extreme cruelty to achieve them. Cf. Rochin v. California, 342 U. S. 165, 172-173 (1952).

The apparent seed of the “unnecessary cruelty” argument is the following language, quoted earlier, found in Wilkerson v. Utah, supra:

“Difficulty would attend the effort to define with exactness the extent of the constitutional provision which provides that cruel and unusual punishments shall not be inflicted; but it is safe to affirm that punishments of torture . . . and all others in the same line of unnecessary cruelty, are forbidden by that amendment to the Constitution.” 99 U. S., at 135-136 (emphasis added).

To lift the italicized phrase from the context of the Wilkerson opinion and now view it as a mandate for assessing the value of punishments in achieving the aims of penology is a gross distortion; nowhere are such aims even mentioned in the Wilkerson opinion. The only fair reading of this phrase is that punishments similar to torture in their extreme cruelty are prohibited by the Eighth Amendment. In Louisiana ex rel. Francis v. Resweber, 329 U. S., at 463, 464, the Court made reference to the Eighth Amendment’s prohibition against the infliction of “unnecessary pain” in carrying out an execution. The context makes abundantly clear that the Court was disapproving the wanton infliction of phys*393ical pain, and once again not advising pragmatic analysis of punishments approved by legislatures.17

Apart from these isolated uses of the word “unnecessary/' nothing in the cases suggests that it is for the courts to make a determination of the efficacy of punishments. The decision in Weems v. United States, supra, is not to the contrary. In Weems the Court held that for the crime of falsifying public documents, the punishment imposed under the Philippine Code of 15 years' imprisonment at hard labor under shackles, followed by perpetual surveillance, loss of voting rights, loss of the right to hold public office, and loss of right to change domicile freely, was violative of the Eighth Amendment. The case is generally regarded as holding that a punishment may be excessively cruel within the meaning of the Eighth Amendment because it is grossly out of proportion to the severity of the crime;18 some view the decision of the Court primarily as *394a reaction to the mode of the punishment itself.19 Under any characterization of the holding, it is readily apparent that the decision grew out of the Court’s overwhelming abhorrence of the imposition of the particular penalty for the particular crime; it was making an essentially moral judgment, not a dispassionate assessment of the need for the penalty. The Court specifically disclaimed “the right to assert a judgment against that of the legislature of the expediency of the laws . . . 217 U. S., at 378. Thus, apart from the fact that the Court in Weems concerned itself with the crime committed as well as the punishment imposed, the case marks no departure from the largely unarticulable standard of extreme cruelty. However intractable that standard may be, that is what the Eighth Amendment is all about. The constitutional provision is not addressed to social utility and does not command that enlightened principles of penology always be followed.

By pursuing the necessity approach, it becomes even more apparent that it involves matters outside the purview of the Eighth Amendment. Two of the several aims of punishment are generally associated with capital punishment — retribution and deterrence. It is argued that retribution can be discounted because that, after all, is what the Eighth Amendment seeks to eliminate. There is no authority suggesting that the Eighth Amendment was intended to purge the law of its retributive elements, and the Court has consistently assumed that retribution is a legitimate dimension of the punishment of crimes. See Williams v. New York, 337 U. S. 241, 248 (1949); United States v. Lovett, 328 U. S. 303, 324 (1946) (Frankfurter, J., concurring). Furthermore, responsible legal thinkers of widely varying *395persuasions have debated the sociological and philosophical aspects of the retribution question for generations, neither side being able to convince the other.20 It would be reading a great deal into the Eighth Amendment to hold that the punishments authorized by-legislatures cannot constitutionally reflect a retributive purpose.

The less esoteric but no less controversial question is whether the death penalty acts as a superior deterrent. Those favoring abolition find no evidence that it does.21 Those favoring retention start from the intuitive notion that capital punishment should act as the most effective deterrent and note that there is no convincing evidence that it does not.22 Escape from this empirical stalemate is sought by placing the burden of proof on the States and concluding that they have failed to demonstrate that capital punishment is a more effective deterrent than life imprisonment. Numerous justifications have been advanced for shifting the burden, and they *396are not without their rhetorical appeal. However, these arguments are not descended from established constitutional principles, but are born of the urge to bypass an unresolved factual question.23 Comparative deterrence is not a matter that lends itself to precise measurement; to shift the burden to the States is to provide an illusory solution to an enormously complex problem. If it were proper to put the States to the test of demonstrating the deterrent value of capital punishment, we could just as well ask them to prove the need for life imprisonment or any other punishment. Yet I know of no convincing evidence that life imprisonment is a more effective deterrent than 20 years' imprisonment, or even that a $10 parking ticket is a more effective deterrent than a $5 parking ticket. In fact, there are some who go so far as to challenge the notion that any punishments deter crime.24 If the States are unable to adduce convincing proof rebutting such assertions, does it then follow that all punishments are suspect as being “cruel and unusual” within the meaning of the Constitution? On the contrary, I submit that the questions raised by the necessity approach are beyond the pale of judicial inquiry under the Eighth Amendment.

y

Today the Court has not ruled that capital punishment is per se violative of the Eighth Amendment; nor has it ruled that the punishment is barred for any particular class or classes of crimes. The substantially similar concurring opinions of Mr. Justice Stewart and Mr. Justice White, which are necessary to support the judgment setting aside petitioners’ sentences, stop *397short of reaching the ultimate question. The actual scope of the Court’s ruling, which I take to be embodied in these concurring opinions, is not entirely clear. This much, however, seems apparent: if the legislatures are to continue to authorize capital punishment for some crimes, juries and judges can no longer be permitted to make the sentencing determination in the same manner they have in the past.25 This approach — not urged in oral arguments or briefs — misconceives the nature of the constitutional command against “cruel and unusual punishments,” disregards controlling case law, and demands a rigidity in capital cases which, if possible of achievement, cannot be regarded as a welcome change. Indeed the contrary seems to be the case.

As I have earlier stated, the Eighth Amendment forbids the imposition of punishments that are so cruel and inhumane as to violate society’s standards of civilized conduct. The Amendment does not prohibit all punishments the States are unable to prove necessary to deter or control crime. The Amendment is not concerned with the process by which a State determines that a particular punishment is to be imposed in a particular case. And the Amendment most assuredly does not speak to the power of legislatures to confer sentencing discretion on juries, rather than to fix all sentences by statute.

The critical factor in the concurring opinions of both Mr. Justice Stewart and Mr. Justice White is the infrequency with which the penalty is imposed. This factor is taken not as evidence of society’s abhorrence *398of capital punishment — the inference that petitioners would have the Court draw — but as the earmark of a deteriorated system of sentencing. It is concluded that petitioners’ sentences must be set aside, not because the punishment is impermissibly cruel, but because juries and judges have failed to exercise their sentencing discretion in acceptable fashion.

To be sure, there is a recitation cast in Eighth Amendment terms: petitioners’ sentences are “cruel” because they exceed that which the legislatures have deemed necessary for all cases;26 petitioners’ sentences are “unusual” because they exceed that which is imposed in most cases.27 This application of the words of the Eighth Amendment suggests that capital punishment can be made to satisfy Eighth Amendment values if its rate of imposition is somehow multiplied; it seemingly follows that the flexible sentencing system created by the legislatures, and carried out by juries and judges, has yielded more mercy than the Eighth Amendment can stand. The implications of this approach are mildly ironical. For example, by this measure of the Eighth Amendment, the elimination of death-qualified juries in Witherspoon v. Illinois, 391 U. S. 510 (1968), can only be seen in retrospect as a setback to “the evolving standards of decency that mark the progress of a maturing society.” Trop v. Dulles, 356 U. S., at 101.

This novel formulation of Eighth Amendment principles — albeit necessary to satisfy the terms of our limited grant of certiorari — does not lie at the heart of these concurring opinions. The decisive grievance of the opinions — not translated into Eighth Amendment terms— is that the present system of discretionary sentencing *399in capital cases has failed to produce evenhanded justice; the problem is not that too few have been sentenced to die, but that the selection process has followed no rational pattern.28 This claim of arbitrariness is not only lacking in empirical support29 but also it manifestly fails to establish that the death penalty is a “cruel and unusual” punishment. The Eighth Amendment was included in the Bill of Rights to assure that certain types of punishments would never be imposed, not to channelize the sentencing process. The approach of these concurring opinions has no antecedent in the Eighth Amendment cases. It is essentially and exclusively a procedural due process argument.

This ground of decision is plainly foreclosed as well as misplaced. Only one year ago, in McGautha v. California, the Court upheld the prevailing system of sentencing in capital cases. The Court concluded:

“In light of history, experience, and the present limitations of human knowledge, we find it quite impossible to say that committing to the untrammeled discretion of the jury the power to pronounce life or death in capital cases is offensive to anything in the Constitution.” 402 U. S., at 207.

In reaching this decision, the Court had the benefit of extensive briefing, full oral argument, and six months of careful deliberations. The Court’s labors are documented by 130 pages of opinions in the United States Reports. All of the arguments and factual contentions accepted *400in the concurring opinions today were considered and rejected by the Court one year ago. McGautha was an exceedingly difficult case, and reasonable men could fairly disagree as to the result. But the Court entered its judgment, and if stare decisis means anything, that decision should be regarded as a controlling pronouncement of law.

Although the Court’s decision in McGautha was technically confined to the dictates of the Due Process Clause of the Fourteenth Amendment, rather than the Eighth Amendment as made applicable to the States through the Due Process Clause of the Fourteenth Amendment, it would be disingenuous to suggest that today’s ruling has done anything less than overrule McGautha in the guise of an Eighth Amendment adjudication. It may be thought appropriate to subordinate principles of stare decisis where the subject is as sensitive as capital punishment and the stakes are so high, but these external considerations were no less weighty last year. This pattern of decisionmaking will do little to inspire confidence in the stability of the law.

While I would not undertake to make a definitive statement as to the parameters of the Court’s ruling, it is clear that if state legislatures and the Congress wish to maintain the availability of capital punishment, significant statutory changes will have to be made. Since the two pivotal concurring opinions turn on the assumption that the punishment of death is now meted out in a random and unpredictable manner, legislative bodies may seek to bring their laws into compliance with the Court’s ruling by providing standards for juries and judges to follow in determining the sentence in capital cases or by more narrowly defining the crimes for which the penalty is to be imposed.30 If such standards can be devised or *401the crimes more meticulously defined, the result cannot be detrimental. However, Mr. Justice Harlan’s opinion for the Court in McGautha convincingly demonstrates that all past efforts “to identify before the fact” the cases in which the penalty is to be imposed have been “uniformly unsuccessful.” 402 U. S., at 197. One problem is that “the factors which determine whether the sentence of death is the appropriate penalty in particular cases are too complex to be compressed within the limits of a simple formula . . . .” Report of Royal Commission on Capital Punishment, 1949-1953, Cmd. 8932, ¶ 498, p. 174 (1953). As the Court stated in McGautha, “[t]he infinite variety of cases and facets to each case would make general standards either meaningless 'boilerplate’ or a statement of the obvious that no jury would need.” 402 U. S., at 208. But even assuming that suitable guidelines can be established, there is no assurance that sentencing patterns will change so long as juries are possessed of the power to determine the sentence or to bring in a verdict of guilt on a charge carrying a lesser sentence; juries have not been inhibited in the exercise of these powers in the past. Thus, unless the Court in McGautha misjudged the experience of history, there is little reason to believe that sentencing standards in any form will substantially alter the discretionary character of the prevailing system of sentencing in capital cases. That system may fall short of perfection, but it is yet to be shown that a different system would produce more satisfactory results.

Real change could clearly be brought about if legislatures provided mandatory death sentences in such a way as to deny juries the opportunity to bring in a verdict on a lesser charge; under such a system, the death sentence could only be avoided by a verdict of acquittal. If this is the only alternative that the legislatures can safely pursue under today’s ruling, I would have preferred that the Court opt for total abolition.

*402It seems remarkable to me that with our basic trust in lay jurors as the keystone in our system of criminal justice, it should now be suggested that we take the most sensitive and important of all decisions away from them. I could more easily be persuaded that mandatory sentences of death, without the intervening and ameliorating impact of lay jurors, are so arbitrary and doctrinaire that they violate the Constitution. The very infrequency of death penalties imposed by jurors attests their cautious and discriminating reservation of that penalty for the most extreme cases. I had thought that nothing was clearer in history, as we noted in McGautha one year ago, than the American abhorrence of “the common-law rule imposing a mandatory death sentence on all convicted murderers.” 402 U. S., at 198. As the concurring opinion of Me. Justice Marshall shows, ante, at 339, the 19th century movement away from mandatory death sentences marked an enlightened introduction of flexibility into the sentencing process. It recognized that individual culpability is not always measured by the category of the crime committed. This change in sentencing practice was greeted by the Court as a humanizing development. See Winston v. United States, 172 U. S. 303 (1899); cf. Calton v. Utah, 130 U. S. 83 (1889). See also Andres v. United States, 333 U. S. 740, 753 (1948) (Frankfurter, J., concurring). I do not see how this history can be ignored and how it can be suggested that the Eighth Amendment demands the elimination of the most sensitive feature of the sentencing system.

As a general matter, the evolution of penal concepts in this country has not been marked by great progress, nor have the results up to now been crowned with significant success. If anywhere in the whole spectrum of criminal justice fresh ideas deserve sober analysis, the sentencing and porrectional area ranks high on the list. But it has been widely accepted that mandatory sentences for *403crimes do not best serve the ends of the criminal justice system. Now, after the long process of drawing away from the blind imposition of uniform sentences for every person convicted of a particular offense, we are confronted with an argument perhaps implying that only the legislatures may determine that a sentence of death is appropriate, without the intervening evaluation of jurors or judges. This approach threatens to turn back the progress of penal reform, which has moved until recently at too slow a rate to absorb significant setbacks.

VI

Since there is no majority of the Court on the ultimate issue presented in these cases, the future of capital punishment in this country has been left in an uncertain limbo. Rather than providing a final and unambiguous answer on the basic constitutional question, the collective impact of the majority’s ruling is to demand an undetermined measure of change from the various state legislatures and the Congress. While I cannot endorse the process of decisionmaking that has yielded today’s result and the restraints that that result imposes on legislative action, I am not altogether displeased that legislative bodies have been given the opportunity, and indeed unavoidable responsibility, to make a thorough re-evaluation of the entire subject of capital punishment. If today’s opinions demonstrate nothing else, they starkly show that this is an area where legislatures can act far more effectively than courts.

The legislatures are free to eliminate capital punishment for specific crimes or to carve out limited exceptions to a general abolition of the penalty, without adherence to the conceptual strictures of the Eighth Amendment. The legislatures can and should make an assessment of the deterrent influence of capital punishment, both generally and as affecting the commission of specific types of *404crimes. If legislatures come to doubt the efficacy of capital punishment, they can abolish it, either completely or on a selective basis. If new evidence persuades them that they have acted unwisely, they can reverse their field and reinstate the penalty to the extent it is thought warranted. An Eighth Amendment ruling by judges cannot be made with such flexibility or discriminating precision.

The world-wide trend toward limiting the use of capital punishment, a phenomenon to which we have been urged to give great weight, hardly points the way to a judicial solution in this country under a written Constitution. Rather, the change has generally come about through legislative action, often on a trial basis and with the retention of the penalty for certain limited classes of crimes.31 Virtually nowhere has change been wrought by so crude a tool as the Eighth Amendment. The complete and unconditional abolition of capital punishment in this country by judicial fiat would have undermined the careful progress of the legislative trend and foreclosed further inquiry on many as yet unanswered questions in this area.

Quite apart from the limitations of the Eighth Amendment itself, the preference for legislative action is justified by the inability of the courts to participate in the *405debate at the level where the controversy is focused. The case against capital punishment is not the product of legal dialectic, but rests primarily on factual claims, the truth of which cannot be tested by conventional judicial processes. The five opinions in support of the judgments differ in many respects, but they share a willingness to make sweeping factual assertions, unsupported by empirical data, concerning the manner of imposition and effectiveness of capital punishment in this country. Legislatures will have the opportunity to make a more penetrating study of these claims with the familiar and effective tools available to them as they are not to us.

The highest judicial duty is to recognize the limits on judicial power and to permit the democratic processes to deal with matters falling outside of those limits. The “hydraulic pressure [s]” 32 that Holmes spoke of as being generated by cases of great import have propelled the Court to go beyond the limits of judicial power, while fortunately leaving some room for legislative judgment.

Me. Justice Blackmun,

dissenting.

I join the respective opinions of The Chief Justice, Me. Justice Powell, and Me. Justice Rehnquist, and add only the following, somewhat personal, comments.

1. Cases such as these provide for me an excruciating agony of the spirit. I yield to no one in the depth of my distaste, antipathy, and, indeed, abhorrence, for the death penalty, with all its aspects of physical distress and fear and of moral judgment exercised by finite minds. That distaste is buttressed by a belief that capital punishment serves no useful purpose that can be demonstrated. For me, it violates childhood’s training and life’s experiences, and is not compatible *406with the philosophical convictions I have been able to develop. It is antagonistic to any sense of “reverence for life.” Were I a legislator, I would vote against the death penalty for the policy reasons argued by counsel for the respective petitioners and expressed and adopted in the several opinions filed by the Justices who vote to reverse these judgments.

2. Having lived for many years in a State that does not have the death penalty,1 that effectively abolished it in 1911,2 and that carried out its last execution on February 13, 1906,3 capital punishment had never been a part of life for me. In my State, it just did not exist. So far as I can determine, the State, purely from a statistical deterrence point of view, was neither the worse nor the better for its abolition, for, as the concurring opinions observe, the statistics prove little, if anything. But the State and its citizens accepted the fact that the death penalty was not to be in the arsenal of possible punishments for any crime.

3. I, perhaps alone among the present members of the Court, am on judicial record as to this. As a member of the United States Court of Appeals, I first struggled silently with the issue of capital punishment in Feguer v. United States, 302 F. 2d 214 (CA8 1962), cert. denied, 371 U. S. 872 (1962). The defendant in that case may have been one of the last to be executed under federal auspices. I struggled again with the issue, and once more refrained from comment, in my writing for an en banc court in Pope v. United States, 372 F. 2d 710 (CA8 1967), vacated (upon acknowledgment by the Solicitor General of error revealed by the subsequently decided United States v. Jackson, 390 U. S. 570 (1968)) and remanded, 392 U. S. 651 (1968). Finally, in Max*407well v. Bishop, 398 F. 2d 138 (CA8 1968), vacated and remanded, sua sponte, by the Court on grounds not raised below, 398 U. S. 262 (1970), I revealed, solitarily and not for the panel, my distress and concern. 398 F. 2d, at 153-154.4 And in Jackson v. Bishop, 404 F. 2d 571 (CA8 1968), I had no hesitancy in writing a panel opinion that held the use of the strap by trusties upon fellow Arkansas prisoners to be a violation of the Eighth Amendment. That, however, was in-prison punishment imposed by inmate-foremen.

4. The several concurring opinions acknowledge, as they must, that until today capital punishment was accepted and assumed as not unconstitutional per se under the Eighth Amendment or the Fourteenth Amendment. This is either the flat or the implicit holding of a unanimous Court in Wilkerson v. Utah, 99 U. S. 130, 134-135, in 1879; of a unanimous Court in In re Kemmler, 136 U. S. 436, 447, in 1890; of the Court in Weems v. United States, 217 U. S. 349, in 1910; of all those members of the Court, a majority, who addressed the issue in Louisiana ex rel. Francis v. Resweber, 329 U. S. 459, 463-464, 471-472, in 1947; of Mr. Chief Justice Warren, speaking for himself and three others (Justices Black, Doug*408las, and Whittaker) in Trop v. Dulles, 356 U. S. 86, 99, in 1958;5 in the denial of certiorari in Rudolph v. Alabama, 375 U. S. 889, in 1963 (where, however, Justices Douglas, Brennan, and Goldberg would have heard argument with respect to the imposition of the ultimate penalty on a convicted rapist who had “neither taken nor endangered human life”); and of Mr. Justice Black in McGautha v. California, 402 U. S. 183, 226, decided only last Term on May 3, 1971.6

Suddenly, however, the course of decision is now the opposite way, with the Court evidently persuaded that somehow the passage of time has taken us to a place of greater maturity and outlook. The argument, plausible and high-sounding as it may be, is not persuasive, for it is only one year since McGautha, only eight and one-half years since Rudolph, 14 years since Trop, and 25 years since Francis, and we have been presented with nothing that demonstrates a significant movement of any kind in these brief periods. The Court has just decided that it is time to strike down the death penalty. There would have been as much reason to do this *409when any of the cited cases were decided. But the Court refrained from that action on each of those occasions.

The Court has recognized, and I certainly subscribe to the proposition, that the Cruel and Unusual Punishments Clause “may acquire meaning as public opinion becomes enlightened by a humane justice.” Weems v. United States, 217 U. S., at 378. And Mr. Chief Justice Warren, for a plurality of the Court, referred to “the evolving standards of decency that mark the progress of a maturing society.” Trop v. Dulles, 356 U. S., at 101. Mr. Jefferson expressed the same thought well.7

*410My problem, however, as I have indicated, is the suddenness of the Court’s perception of progress in the human attitude since decisions of only a short while ago.

5. To reverse the judgments in these cases is, of course, the easy choice. It is easier to strike the balance in favor of life and against death. It is comforting to relax in the thoughts — perhaps the rationalizations— that this is the compassionate decision for a maturing society; that this is the moral and the “right” thing to do; that thereby we convince ourselves that we are moving down the road toward human decency; that we value life even though that life has taken another or others or has grievously scarred another or others and their families; and that we are less barbaric than we were in 1879, or in 1890, or in 1910, or in 1947, or in 1958, or in 1963, or a year ago, in 1971, when Wilkerson, Kemmler, Weems, Francis, Trop, Rudolph, and Mc-Gautha were respectively decided.

This, for me, is good argument, and it makes some sense. But it is good argument and it makes sense only in a legislative and executive way and not as a judicial expedient. As I have said above, were I a legislator, I would do all I could to sponsor and to vote for legislation abolishing the death penalty. And were I the chief executive of a sovereign State, I would be sorely tempted to exercise executive clemency as Governor Rockefeller of Arkansas did recently just before he departed from office. There — on the Legislative Branch of the State or Federal Government, and secondarily, on the Executive Branch — is where the authority and responsibility for this kind of action lies. The authority should not be taken over by the judiciary in the modern guise of an Eighth Amendment issue.

I do not sit on these cases, however, as a legislator, responsive, at least in part, to the will of constituents. *411Our task here, as must so frequently be emphasized and re-emphasized, is to pass upon the constitutionality of legislation that has been enacted and that is challenged. This is the sole task for judges. We should not allow our personal preferences as to the wisdom of legislative and congressional action, or our distaste for such action, to guide our judicial decision in cases such as these. The temptations to cross that policy line are very great. In fact, as today’s decision reveals, they are almost irresistible.

6. The Court, in my view, is somewhat propelled toward its result by the interim decision of the California Supreme Court, with one justice dissenting, that the death penalty is violative of that State’s constitution. People v. Anderson, 6 Cal. 3d 628, 493 P. 2d 880 (Feb. 18, 1972). So far as I am aware, that was the first time the death penalty in its entirety has been nullified by judicial decision. Cf. Ralph v. Warden, 438 F. 2d 786, 793 (CA4 1970), cert. denied, post, p. 942. California’s moral problem was a profound one, for more prisoners were on death row there than in any other State. California, of course, has the right to construe its constitution as it will. Its construction, however, is hardly a precedent for federal adjudication.

7. I trust the Court fully appreciates what it is doing when it decides these cases the way it does today. Not only are the capital punishment laws of 39 States and the District of Columbia struck down, but also all those provisions of the federal statutory structure that permit the death penalty apparently are voided. No longer is capital punishment possible, I suspect, for, among other crimes, treason, 18 U. S. C. § 2381; or assassination of the President, the Vice President, or those who stand elected to those positions, 18 U. S. C. § 1751; or assassination of a Member or member-elect of Congress, 18 U. S. C. § 351; or espionage, 18 U. S. C. § 794; *412or rape within the special maritime jurisdiction, 18 U. S. C. § 2031; or aircraft or motor vehicle destruction where death occurs, 18 U. S. C. § 34; or explosives offenses where death results, 18 U. S. C. §§ 844 (d) and (f); or train wrecking, 18 U. S. C. § 1992; or aircraft piracy, 49 U. S. C. § 1472 (i). Also in jeopardy, perhaps, are the death penalty provisions in various Articles of the Uniform Code of Military Justice. 10 U. S. C. §§ 885, 890, 894, 899, 901, 904, 906, 913, 918, and 920. All these seem now to be discarded without a passing reference to the reasons, or the circumstances, that prompted their enactment, some very recent, and their retention in the face of efforts to repeal them.

8. It is of passing interest to note a few voting facts with respect to recent federal death penalty legislation:

A. The aircraft piracy statute, 49 U. S. C. § 1472 (i), was enacted September 5, 1961. The Senate vote on August 10 was 92-0. It was announced that Senators Chavez, Fulbright, Neuberger, and Symington were absent but that, if present, all four would vote yea. It was also announced, on the other side of the aisle, that Senator Butler was ill and that Senators Beall, Carlson, and Morton were absent or detained, but that those four, if present, would vote in the affirmative. These announcements, therefore, indicate that the true vote was 100-0. 107 Cong. Rec. 15440. The House passed the bill without recorded vote. 107 Cong. Rec. 16849.

B. The presidential assassination statute, 18 U. S. C. § 1751, was approved August 28, 1965, without recorded votes. Ill Cong. Rec. 14103, 18026, and 20239.

C. The Omnibus Crime Control Act of 1970 was approved January 2, 1971. Title IV thereof added the congressional assassination statute that is now 18 U. S. C. § 351. The recorded House vote on October 7, 1970, was 341-26, with 63 not voting and 62 of those paired. 116 Cong. Rec. 35363-35364. The Senate vote on October 8 *413was 59-0, with 41 not voting, but with 21 of these announced as favoring the bill. 116 Cong. Rec. 35743. Final votes after conference were not recorded. 116 Cong. Rec. 42150, 42199.

It is impossible for me to believe that the many lawyer-members of the House and Senate — including, I might add, outstanding leaders and prominent candidates for higher office — were callously unaware and insensitive of constitutional overtones in legislation of this type. The answer, of course, is that in 1961, in 1965, and in 1970 these elected representatives of the' people — far more conscious of the temper of the times, of the maturing of society, and of the contemporary demands for man’s dignity, than are we who sit cloistered on this Court— took it as settled that the death penalty then, as it always had been, was not in itself unconstitutional. Some of those Members of Congress, I suspect, will be surprised at this Court’s giant stride today.

9. If the reservations expressed by my Brother Stewart (which, as I read his opinion, my Brother White shares) were to command support, namely, that capital punishment may not be unconstitutional so long as it be man-datorily imposed, the result, I fear, will be that statutes struck down today will be re-enacted by state legislatures to prescribe the death penalty for specified crimes without any alternative for the imposition of a lesser punishment in the discretion of the judge or jury, as the case may be. This approach, it seems to me, encourages legislation that is regressive and of an antique mold, for it eliminates the element of mercy in the imposition of punishment. I thought we had passed beyond that point in our criminology long ago.

10. It is not without interest, also, to note that, although the several concurring opinions acknowledge the heinous and atrocious character of the offenses committed by the petitioners, none of those opinions makes *414reference to the misery the petitioners’ crimes occasioned to the victims, to the families of the victims, and to the communities where the offenses took place. The arguments for the respective petitioners, particularly the oral arguments, were similarly and curiously devoid of reference to the victims. There is risk, of course, in a comment such as this, for it opens one to the charge of emphasizing the retributive. But see Williams v. New York, 337 U. S. 241, 248 (1949). Nevertheless, these cases are here because offenses to innocent victims were perpetrated. This fact, and the terror that occasioned it, and the fear that stalks the streets of many of our cities today perhaps deserve not to be entirely overlooked. Let us hope that, with the Court’s decision, the terror imposed will be forgotten by those up'on whom it was visited, and that our society will reap the hoped-for benefits of magnanimity.

Although personally I may rejgice at the Court’s result, I find it difficult to accept or to justify as a matter of history, of law, or of constitutional pronouncement. I fear the Court has overstepped. It has sought and has achieved an end.

Mr. Justice Powell,

with whom The Chief Justice, Mr. Justice Blackmun, and Mr. Justice Rehnquist join, dissenting.

The Court granted certiorari in these cases to consider whether the death penalty is any longer a permissible form of punishment. 403 U. S. 952 (1971). It is the judgment of five Justices that the death penalty, as customarily prescribed and implemented in this country today, offends the constitutional prohibition against cruel and unusual punishments. The reasons for that judgment are stated in five separate opinions, expressing as many separate rationales. In my view, none of these opinions provides a constitutionally adequate foundation for the Court’s decision.

*415Mr. Justice Douglas concludes that capital punishment is incompatible with notions of "equal protection” that he finds to be “implicit” in the Eighth Amendment. Ante, at 257. Mr. Justice Brennan bases his judgment primarily on the thesis that the penalty “does not comport with human dignity.” Ante, at 270. Mr. Justice Stewart concludes that the penalty is applied in a “wanton” and “freakish” manner. Ante, at 310. For Mr. Justice White it is the “infrequency” with which the penalty is imposed that renders its use unconstitutional. Ante, at 313. Mr. Justice Marshall finds that capital punishment is an impermissible form of punishment because it is “morally unacceptable” and “excessive.” Ante, at 360, 358.

Although the central theme of petitioners’ presentations in these cases is that the imposition of the death penalty is per se unconstitutional, only two of today’s opinions explicitly conclude that so sweeping a determination is mandated by the Constitution. Both Mr. Justice Brennan and Mr. Justice Marshall call for the abolition of all existing state and federal capital punishment statutes. They intimate as well that no capital statute could be devised in the future that might comport with the Eighth Amendment. While the practical consequences of the other three opinions are less certain, they at least do not purport to render impermissible every possible statutory scheme for the use of capital punishment that legislatures might hereafter devise.1 Insofar as these latter opinions fail, at least ex*416plicitly, to go as far as petitioners’ contentions would carry them, their reservations are attributable to a willingness to accept only a portion of petitioners’ thesis. For the reasons cogently set out in the Chief Justice’s dissenting opinion (ante, at 396-403), and for reasons stated elsewhere in this opinion, I find my Brothers’ less-than-absolute-abolition judgments unpersuasive. Because those judgments are, for me, not dispositive, I shall focus primarily on the broader ground upon which the petitions in these cases are premised. The foundations of my disagreement with that broader thesis are equally applicable to each of the concurring opinions. I will, therefore, not endeavor to treat each one separately. Nor will I attempt to predict what forms of capital statutes, if any, may avoid condemnation in the future under the variety of views expressed by the collective majority today. That difficult task, not performed in any of the controlling opinions, must go unanswered until other cases presenting these more limited inquiries arise.

Whatever uncertainties may hereafter surface, several of the consequences of today’s decision are unmistakably clear. The decision is plainly one of the greatest im*417portance. The Court's judgment removes the death sentences previously imposed on some 600 persons awaiting punishment in state and federal prisons throughout the country. At least for the present, it also bars the States and the Federal Government from seeking sentences of death for defendants awaiting trial on charges for which capital punishment was heretofore a potential alternative. The happy event for these countable few constitutes, however, only the most visible consequence of this decision. Less measurable, but certainly of no less significance, is the shattering effect this collection of views has on the root principles of stare decisis, federalism, judicial restraint and — most importantly — separation of powers.

The Court rejects as not decisive the clearest evidence that the Framers of the Constitution and the authors of the Fourteenth Amendment believed that those documents posed no barrier to the death penalty. The Court also brushes aside an unbroken line of precedent reaffirming the heretofore virtually unquestioned constitutionality of capital punishment. Because of the pervasiveness of the constitutional ruling sought by petitioners, and accepted in varying degrees by five members of the Court, today’s departure from established precedent invalidates a staggering number of state and federal laws. The capital punishment laws of no less than 39 States2 and the District of Columbia are nullified. In addition, numerous provisions of the Criminal Code of the United States and of the Uniform Code of Mili*418tary Justice also are voided. The Court’s judgment not only wipes out laws presently in existence, but denies to Congress and to the legislatures of the 50 States the power to adopt new policies contrary to the policy selected by the Court. Indeed, it is the view of two of my Brothers that the people of each State must be denied the prerogative to amend their constitutions to provide for capital punishment even selectively for the most heinous crime.

In terms of the constitutional role of this Court, the impact of the majority’s ruling is all the greater because the decision encroaches upon an area squarely within the historic prerogative of the legislative branch — both state and federal — to protect the citizenry through the designation of penalties for prohibitable conduct. It is the very sort of judgment that the legislative branch is competent to make and for which the judiciary is ill-equipped. Throughout our history, Justices of this Court have emphasized the gravity of decisions invalidating legislative judgments, admonishing the nine men who sit on this bench of the duty of self-restraint, especially when called upon to apply the expansive due process and cruel and unusual punishment rubrics. I can recall no case in which, in the name of deciding constitutional questions, this Court has subordinated national and local democratic processes to such an extent. Before turning to address the thesis of petitioners’ case against capital punishment — a thesis that has proved, at least in large measure, persuasive to a majority of this Court— I first will set out the principles that counsel against the Court’s sweeping decision.

I

The Constitution itself poses the first obstacle to petitioners’ argument that capital punishment is per se unconstitutional. The relevant provisions are the Fifth, *419Eighth, and Fourteenth Amendments. The first of these provides in part:

“No person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury . . . ; nor shall any person be subject for the same offence to be twice put in jeopardy of life or limb; . . . nor be deprived of life, liberty, or property, without due process of law . . . .”

Thus, the Federal Government’s power was restricted in order to guarantee those charged with crimes that the prosecution would have only a single opportunity to seek imposition of the death penalty and that the death penalty could not be exacted without due process and a grand jury indictment. The Fourteenth Amendment, adopted about 77 years after the Bill of Rights, imposed the due process limitation of the Fifth Amendment upon the States’ power to authorize capital punishment.

The Eighth Amendment, adopted at the same time as the Fifth, proscribes “cruel and unusual” punishments. In an effort to discern its meaning, much has been written about its history in the opinions of this Court and elsewhere.3 That history need not be restated here since, whatever punishments the Framers of the Constitution may have intended to prohibit under the “cruel and unusual” language, there cannot be the slightest doubt that they intended no absolute bar on the Government’s authority to impose the death penalty. McGautha v. *420California, 402 U. S. 183, 226 (1971) (separate opinion of Black, J.). As much is made clear by the three references to capital punishment in the Fifth Amendment. Indeed, the same body that proposed the Eighth Amendment also provided, in the first Crimes Act of 1790, for the death penalty for a number of offenses. 1 Stat. 112.

Of course, the specific prohibitions within the Bill of Rights are limitations on the exercise of power; they are not an affirmative grant of power to the Government. I, therefore, do not read the several references to capital punishment as foreclosing this Court from considering whether the death penalty in a particular case offends the Eighth and Fourteenth Amendments. Nor are “cruel and unusual punishments” and “due process of law” static concepts whose meaning and scope were sealed at the time of their writing. They were designed to be dynamic and to gain meaning through application to specific circumstances, many of which were not contemplated by their authors. While flexibility in the application of these broad concepts is one of the hallmarks of our system of government, the Court is not free to read into the Constitution a meaning that is plainly at variance with its language. Both the language of the Fifth and Fourteenth Amendments and the history of the Eighth Amendment confirm beyond doubt that the death penalty was considered to be a constitutionally permissible punishment. It is, however, within the historic process of constitutional adjudication to challenge the imposition of the death penalty in some barbaric manner or as a penalty wholly disproportionate to a particular criminal act. And in making such a judgment in a case before it, a court may consider contemporary standards to the extent they are relevant. While this weighing of a punishment against the Eighth Amendment standard on a case-by-case basis is consonant with history and precedent, it is not what *421petitioners demand in these cases. They seek nothing less than the total abolition of capital punishment by judicial fiat.

II

Petitioners assert that the constitutional issue is an open one uncontrolled by prior decisions of this Court. They view the several cases decided under the Eighth Amendment as assuming the constitutionality of the death penalty without focusing squarely upon the issue. I do not believe that the case law can be so easily cast aside. The Court on numerous occasions has both assumed and asserted the constitutionality of capital punishment. In several cases that assumption provided a necessary foundation for the decision, as the issue was whether a particular means of carrying out a capital sentence would be allowed to stand. Each of those decisions necessarily was premised on the assumption that some method of exacting the penalty was permissible.

The issue in the first capital case in which the Eighth Amendment was invoked, Wilkerson v. Utah, 99 U. S. 130 (1879), was whether carrying out a death sentence by public shooting was cruel and unusual punishment. A unanimous Court upheld that form of execution, noting first that the punishment itself, as distinguished from the mode of its infliction, was “not pretended by the counsel of the prisoner” (id., at 137) to be cruel and unusual. The Court went on to hold that:

“Cruel and unusual punishments are forbidden by the Constitution, but the authorities . . . are quite sufficient to show that the punishment of shooting as a mode of executing the death penalty for the crime of murder in the first degree is not included in that category . . . .” Id., at 134-135.

Eleven years later, in In re Kemmler, 136 U. S. 436 (1890), the Court again faced a question involving the *422method of carrying out a capital sentence. On review of a denial of habeas corpus relief by the Supreme Court of New York, this Court was called on to decide whether electrocution, which only very recently had been adopted by the New York Legislature as a means of execution, was impermissibly cruel and unusual in violation of the Fourteenth Amendment.4 Chief Justice Fuller, speaking for the entire Court, ruled in favor of the State. Electrocution had been selected by the legislature, after careful investigation, as “the most humane and practical method known to modern science of carrying into effect the sentence of death.” Id., at 444. The Court drew a clear line between the penalty itself and the mode of its execution:

“Punishments are cruel when they involve torture or a lingering death; but the punishment of death *423is not cruel, within the meaning of that word as used in the Constitution. It implies there something inhuman and barbarous, something more than the mere extinguishment of life.” Id., at 447.

More than 50 years later, in Louisiana ex rel. Francis v. Resweber, 329 U. S. 459 (1947), the Court considered a case in which, due to a mechanical malfunction, Louisiana’s initial attempt to electrocute a convicted murderer had failed. Petitioner sought to block a second attempt to execute the sentence on the ground that to do so would constitute cruel and unusual punishment. In the plurality opinion written by Mr. Justice Reed, concurred in by Chief Justice Vinson and Justices Black and Jackson, relief was denied. Again the Court focused on the manner of execution, never questioning the propriety of the death sentence itself.

“The case before us does not call for an examination into any punishments except that of death. . . . The traditional humanity of modern Anglo-American law forbids the infliction of unnecessary pain in the execution of the death sentence. . . .

“. . . The cruelty against which the Constitution protects a convicted man is cruelty inherent in the method of punishment, not the necessary suffering involved in any method employed to extinguish life humanely.” Id., at 463-464.

Mr. Justice Frankfurter, unwilling to dispose of the case under the Eighth Amendment’s specific prohibition, approved the second execution attempt under the Due Process Clause. He concluded that “a State may be found to deny a person due process by treating even one guilty of crime in a manner that violates standards of *424decency more or less universally accepted though not when it treats him by a mode about which opinion is fairly divided.” Id., at 469-470.

The four dissenting Justices, although finding a second attempt at execution to be impermissibly cruel, expressly recognized the validity of capital punishment:

“In determining whether the proposed procedure is unconstitutional, we must measure it against a lawful electrocution. . . . Electrocution, when instantaneous, can be inflicted by a state in conformity with due process of law. . . .

“The all-important consideration is that the execution shall be so instantaneous and substantially painless that the punishment shall be reduced, as nearly as possible, to no more than that of death itself.” Id., at 474 (original emphasis).

Each of these cases involved the affirmance of a death sentence where its validity was attacked as violating the Eighth Amendment. Five opinions were written in these three cases, expressing the views of 23 Justices. While in the narrowest sense it is correct to say that in none was there a frontal attack upon the constitutionality of the death penalty, each opinion went well beyond an un-articulated assumption of validity. The power of the States to impose capital punishment was repeatedly and expressly recognized.

In addition to these cases in which the constitutionality of the death penalty was a necessary foundation for the decision, those who today would have this Court undertake the absolute abolition of the death penalty also must reject the opinions of other cases stipulating or assuming the constitutionality of capital punishment. Trop v. Dulles, 356 U. S. 86, 99, 100 (1958) ; Weems v. United States, 217 U. S. 349, 382, 409 (1910) *425(White, J., joined by Holmes, J., dissenting).5 See also McGautha v. California, 402 U. S., at 226 (separate opinion of Black, J.); Robinson v. California, 370 U. S. 660, 676 (1962) (Douglas, J., concurring).

The plurality opinion in Trop v. Dulles, supra, is of special interest since it is this opinion, in large measure, that provides the foundation for the present attack on the death penalty.6 It is anomalous that the standard urged by petitioners — “evolving standards of decency that mark the progress of a maturing society” (356 U. S., at 101) — should be derived from an opinion that so un-qualifiedly rejects their arguments. Chief Justice Warren, joined by Justices Black, Douglas, and Whittaker, stated flatly:

“At the outset, let us put to one side the death penalty as an index of the constitutional limit on punishment. Whatever the arguments may be against capital punishment, both on moral grounds and in terms of accomplishing the purposes of punishment — and they are forceful — the death penalty has been employed throughout our history, and, in a day when it is still widely accepted, it cannot be said to violate the constitutional concept of cruelty.” Id., at 99.

The issue in Trop was whether forfeiture of citizenship was a cruel and unusual punishment when imposed on *426a wartime deserter who had gone “over the hill” for less than a day and had willingly surrendered. In examining the consequences of the relatively novel punishment of denationalization,7 Chief Justice Warren drew a line between “traditional” and “unusual” penalties:

“While the State has the power to punish, the [Eighth] Amendment stands to assure that this power be exercised within the limits of civilized standards. Fines, imprisonment and even execution may be imposed depending upon the enormity of the crime, but any technique outside the bounds of these traditional penalties is constitutionally suspect.” Id., at 100.

The plurality’s repeated disclaimers of any attack on capital punishment itself must be viewed as more than offhand dicta since those views were written in direct response to the strong language in Mr. Justice Frankfurter’s dissent arguing that denationalization could not be a disproportionate penalty for a concededly capital offense.8

The most recent precedents of this Court — Witherspoon v. Illinois, 391 U. S. 510 (1968), and McGautha v. California, supra — are also premised to a significant degree on the constitutionality of the death penalty. While the scope of review in both cases was limited to questions involving the procedures for selecting juries *427and regulating their deliberations in capital cases,9 those opinions were “singularly academic exercise [s]” 10 if the members of this Court were prepared at those times to find in the Constitution the complete prohibition of the death penalty. This is especially true of Mr. Justice Harlan’s opinion for the Court in Mc-Gautha, in which, after a full review of the history of capital punishment, he concluded that “we find it quite impossible to say that committing to the untrammeled discretion of the jury the power to pronounce life or death in capital cases is offensive to anything in the Constitution.” Id., at 207.11

*428Perhaps enough has been said to demonstrate the unswerving position that this Court has taken in opinions spanning the last hundred years. On virtually every occasion that any opinion has touched on the question of' the constitutionality of the death penalty, it has been asserted affirmatively, or tacitly assumed, that the Constitution does not prohibit the penalty. No Justice of the Court, until today, has dissented from this consistent reading of the Constitution. The petitioners in these cases now before the Court cannot fairly avoid the weight of this substantial body of precedent merely by asserting that there is no prior decision precisely in point. Stare decisis, if it is a doctrine founded on principle, surely applies where there exists a long line of cases endorsing or necessarily assuming the validity of a particular matter of constitutional interpretation. Green v. United States, 356 U. S. 165, 189-193 (1958) (Frankfurter, J., concurring). While these oft-repeated expressions of unchallenged belief in the constitutionality of capital punishment may not justify a summary disposition of the constitutional question before us, they are views expressed and joined in over the years by no less than 29 Justices of this Court and therefore merit the greatest respect.12 Those who now resolve to set those views aside indeed have a heavy burden.

Ill

Petitioners seek to avoid the authority of the foregoing cases, and the weight of express recognition in the Constitution itself, by reasoning which will not withstand analysis. The thesis of petitioners’ case derives from several opinions in which members of this Court *429have recognized the dynamic nature of the prohibition against cruel and unusual punishments. The final meaning of those words was not set in 1791. Rather, to use the words of Chief Justice Warren speaking for a plurality of the Court in Trop v. Dulles, 356 U. S., at 100-101:

“[T]he words of the Amendment are not precise, and . . . their scope is not static. The Amendment must draw its meaning from the evolving standards of decency that mark the progress of a maturing society.”

But this was not new doctrine. It was the approach to the Eighth Amendment taken by Mr. Justice McKenna in his opinion for the Court in Weems v. United States, 217 U. S. 349 (1910). Writing for four Justices sitting as the majority of the six-man Court deciding the case, he concluded that the clause must be “progressive”; it is not “fastened to the obsolete but may acquire meaning as public opinion becomes enlightened by a humane justice.” Id., at 378. The same test was offered by Mr. Justice Frankfurter in his separate concurrence in Louisiana ex rel. Francis v. Resweber, 329 U. S., at 469. While he rejected the notion that the Fourteenth Amendment made the Eighth Amendment fully applicable to the States, he nonetheless found as a matter of due process that the States were prohibited from “treating even one guilty of crime in a manner that violates standards of decency more or less universally accepted.”

Whether one views the question as one of due process or of cruel and unusual punishment, as I do for convenience in this case, the issue is essentially the same.13 The fundamental premise upon which either standard is based is that notions of what constitutes cruel and unusual punishment or due process do evolve. *430Neither the Congress nor any state legislature would today tolerate pillorying, branding, or cropping or nailing of the ears — punishments that were in existence during our colonial era.14 Should, however, any such punishment be prescribed, the courts would certainly enjoin its execution. See Jackson v. Bishop, 404 F. 2d 571 (CA8 1968). Likewise, no court would approve any method of implementation of the death sentence found to involve unnecessary cruelty in light of presently available alternatives. Similarly, there may well be a process of evolving attitude with respect to the application of the death sentence for particular crimes.15 See McGautha v. California, 402 U. S., at 242 (Douglas, J., dissenting).

But we are not asked to consider the permissibility of any of the several methods employed in carrying out the death sentence. Nor are we asked, at least as part of the core submission in these cases, to determine whether the penalty might be a grossly excessive punishment for some specific criminal conduct. Either inquiry would call for a discriminating evaluation of particular means, or of the relationship between particular conduct and its punishment. Petitioners’ principal argument goes far beyond the traditional process of case-by-case inclusion and exclusion. Instead the argument insists on an unprecedented constitutional rule of absolute prohibition of capital punishment for any crime, regardless of its depravity and impact on society. In calling for a precipitate and final judicial end to this form of penalty as offensive to evolving standards of decency, petitioners would have this Court abandon the traditional and more refined approach consistently followed in its prior Eighth Amendment precedents. What they are saying, in effect, is that the evolutionary *431process has come suddenly to an end; that the ultimate wisdom as to the appropriateness of capital punishment under all circumstances, and for all future generations, has somehow been revealed.

The prior opinions of this Court point with great clarity to reasons why those of us who sit on this Court at a particular time should act with restraint before assuming, contrary to a century of precedent, that we now know the answer for all time to come. First, where as here, the language of the applicable provision provides great leeway and where the underlying social policies are felt to be of vital importance, the temptation to read personal preference into the Constitution is understandably great. It is too easy to propound our subjective standards of wise policy under the rubric of more or less universally held standards of decency. See Trop v. Dulles, 356 U. S., at 103 (Warren, C. J.), 119-120 (Frankfurter, J., dissenting); Louisiana ex rel. Francis v. Resweber, 329 U. S., at 470-471 (Frankfurter, J., concurring) ; Weems v. United States, 217 U. S., at 378-379 (McKenna, J.).

The second consideration dictating judicial self-restraint arises from a proper recognition of the respective roles of the legislative and judicial branches. The designation of punishments for crimes is a matter peculiarly within the sphere of the state and federal legislative bodies. See, e. g., In re Kemmler, 136 U. S., at 447; Trop v. Dulles, 356 U. S., at 103. When asked to encroach on the legislative prerogative we are well counseled to proceed with the utmost reticence. The review of legislative choices, in the performance of our duty to enforce the Constitution, has been characterized most appropriately by Mr. Justice Holmes as “the gravest and most delicate duty that this Court is called on to perform.” Blodgett v. Holden, 275 U. S. 142, 147-148 (1927) (separate opinion).

*432How much graver is that duty when we are not asked to pass on the constitutionality of a single penalty under the facts of a single case but instead are urged to overturn the legislative judgments of 40 state legislatures as well as those of Congress. In so doing is the majority able to claim, as did the Court in Weems, that it appreciates “to the fullest the wide range of power that the legislature possesses to adapt its penal laws to conditions as they may exist and punish the crimes of men according to their forms and frequency”? 217 U. S., at 379. I think not. No more eloquent statement of the essential separation of powers limitation on our prerogative can be found than the admonition of Mr. Justice Frankfurter, dissenting in Trop. His articulation of the traditional view takes on added significance where the Court undertakes to nullify the legislative judgments of the Congress and four-fifths of the States.

“What is always basic when the power of Congress to enact legislation is challenged is the appropriate approach to judicial review of congressional legislation .... When the power of Congress to pass a statute is challenged, the function of this Court is to determine whether legislative action lies clearly outside the constitutional grant of power to which it has been, or may fairly be, referred. In making this determination, the Court sits in judgment on the action of a co-ordinate branch of the Government while keeping unto itself — as it must under our constitutional system — the final determination of its own power to act. . . .

“Rigorous observance of the difference between limits of power and wise exercise of power — between questions of authority and questions of prudence — requires the most alert appreciation of this decisive but subtle relationship of two concepts that too easily coalesce. No less does it require a *433disciplined will to adhere to the difference. It is not easy to stand aloof and allow want of wisdom to prevail, to disregard one’s own strongly held view of what is wise in the conduct of affairs. But it is not the business of this Court to pronounce policy. It must observe a fastidious regard for limitations on its own power, and this precludes the Court’s giving effect to its own notions of what is wise or politic. That self-restraint is of the essence in the observance of the judicial oath, for the Constitution has not authorized the judges to sit in judgment on the wisdom of what Congress and the Executive Branch do.” 356 U. S., at 119-120.

See also Mr. Justice White’s dissenting opinion in Weems v. United States, 217 U. S., at 382.

IV

Although determining the range of available punishments for • a particular crime is a legislative function, the very presence of the Cruel and Unusual Punishments Clause within the Bill of Rights requires, in the context of a specific case, that courts decide whether particular acts of the Congress offend that Amendment. The Due Process Clause of the Fourteenth Amendment imposes on the judiciary a similar obligation to scrutinize state legislation. But the proper exercise of that constitutional obligation in the cases before us today must be founded on a full recognition of the several considerations set forth above — the affirmative references to capital punishment in the Constitution, the prevailing precedents of this Court, the limitations on the exercise of our power imposed by tested principles of judicial self-restraint, and the duty to avoid encroachment on the powers conferred upon state and federal legislatures. In the face of these considerations, only the most con-*434elusive of objective demonstrations could warrant this Court in holding capital punishment per se unconstitutional. The burden of seeking so sweeping a decision against such formidable obstacles is almost insuperable. Viewed from this perspective, as I believe it must be, the case against the death penalty falls far short.

Petitioners’ contentions are premised, as indicated above, on the long-accepted view that concepts embodied in the Eighth and Fourteenth Amendments evolve. They present, with skill and persistence, a list of “objective indicators” which are said to demonstrate that prevailing standards of human decency have progressed to the final point of requiring the Court to hold, for all cases and for all time, that capital punishment is unconstitutional.

Briefly summarized, these proffered indicia of contemporary standards of decency include the following: (i) a worldwide trend toward the disuse of the death penalty;16 (ii) the reflection in the scholarly literature of a progressive rejection of capital punishment founded essentially on moral opposition to such treatment;17 (iii) the decreasing numbers of executions over the last 40 years and especially over the last decade;18 (iv) the *435small number of death sentences rendered in relation to the number of cases in which they might have been imposed; 19 and (v) the indication of public abhorrence of *436the penalty reflected in the circumstance that executions are no -longer public affairs.20 The foregoing is an incomplete summary but it touches the major bases of petitioners’ presentation. Although they are not appropriate for consideration as objective evidence, petitioners strongly urge two additional propositions. They contend, first, that the penalty survives public condemnation only through the infrequency, arbitrariness, and discriminatory nature of its application, and, second, that there no longer exists any legitimate justification for the utilization of the ultimate penalty. These contentions, which have proved persuasive to several of the Justices constituting the majority, deserve separate consideration and will be considered in the ensuing sections. Before turning to those arguments, I first address the argument based on “objective” factors.

Any attempt to discern contemporary standards of decency through the review of objective factors must take into account several overriding considerations which petitioners choose to discount or ignore. In a democracy *437the first indicator of the public’s attitude must always be found in the legislative judgments of the people’s chosen representatives. Mr. Justice Marshall’s opinion today catalogues the salient statistics. Forty States,21 the District of Columbia, and the Federal Government still authorize the death penalty for a wide variety of crimes. That number has remained relatively static since the end of World War I. Ante, at 339-341. That does not mean, however, that capital punishment has become a forgotten issue in the legislative arena. As recently as January, 1971, Congress approved the death penalty for congressional assassination. 18 U. S. C. § 351. In 1965 Congress added the death penalty for presidential and vice presidential assassinations. 18 U. S. C. § 1751. Additionally, the aircraft piracy statute passed in 1961 also carries the death penalty. 49 U. S. C. § 1472 (i). Mr. Justice Blackmun’s dissenting opinion catalogues the impressive ease with which each of these statutes was approved. Ante, at 412-413. On the converse side, a bill proposing the abolition of capital punishment for all federal crimes was introduced in 1967 but failed to reach the Senate floor.22

At the state level, New York, among other States, has recently undertaken reconsideration of its capital crimes. A law passed in 1965 restricted the use of capital punishment to the crimes of murder of a police officer and murder by a person serving a sentence of life imprisonment. N. Y. Penal Code § 125.30 (1967).

I pause here to state that I am at a loss to under*438stand how those urging this Court to pursue a course of absolute abolition as a matter of constitutional judgment can draw any support from the New York experience. As is also the case with respect to recent legislative activity in Canada23 and Great Britain,24 New York’s decision to restrict the availability of the death penalty is a product of refined and discriminating legislative judgment, reflecting, not the total rejection of capital punishment as inherently cruel, but a desire to limit it to those circumstances in which legislative judgment deems retention to be in the public interest. No such legislative flexibility is permitted by the contrary course petitioners urge this Court to follow.25

In addition to the New York experience, a number of other States have undertaken reconsideration of capital punishment in recent years. In four States the penalty has been put to a vote of the people through public referenda — a means likely to supply objective evidence of community standards. In Oregon a referendum seeking abolition of capital punishment failed in 1958 but was subsequently approved in 1964.26 Two years later the penalty was approved in Colorado by a wide margin.27 *439In Massachusetts in 1968, in an advisory referendum, the voters there likewise recommended retention of the penalty. In 1970, approximately 64% of the voters in Illinois approved the penalty.28 In addition, the National Commission on Reform of Federal Criminal Laws reports that legislative committees in Massachusetts, Pennsylvania, and Maryland recommended abolition, while committees in New Jersey and Florida recommended retention.29 The legislative views of other States have been summarized by Professor Hugo Bedau in his compilation of sources on capital punishment entitled The Death Penalty in America:

“What our legislative representatives think in the two score states which still have the death penalty may be inferred from the fate of the bills to repeal or modify the death penalty filed during recent years in the legislatures of more than half of these states. In about a dozen instances, the bills emerged from committee for a vote. But in none except Delaware did they become law. In those states where these bills were brought to the floor of the legislatures, the vote in most instances wasn’t even close.” 30

This recent history of activity with respect to legislation concerning the death penalty abundantly refutes the abolitionist position.

The second and even more direct source of information *440reflecting the public’s attitude toward capital punishment is the jury. In Witherspoon v. Illinois, 391 U. S. 510 (1968), Mr. Justice Stewart, joined by Justices Brennan and Marshall, characterized the jury’s historic function in the sentencing process in the following terms:

“[T]he jury is given broad discretion to decide whether or not death is ‘the proper penalty’ in a given case, and a juror’s general views about capital punishment play an inevitable role in any such decision.

“A man who opposes the death penalty, no less than one who favors it, can make the discretionary judgment entrusted to him by the State and can thus obey the oath he takes as a juror. . . . Guided by neither rule nor standard, ... a jury that must choose between life imprisonment and capital punishment can do little more — and must do nothing less — than express the conscience of the community on the ultimate question of life or death.”

“[0]ne of the most important functions any jury can perform in making such a selection is to maintain a link between contemporary community values and the penal system — a link without which the determination of punishment could hardly reflect ‘the evolving standards of decency that mark the progress of a maturing society.’ Trop v. Dulles, . . .”31

Any attempt to discern, therefore, where the prevailing standards of decency lie must take careful account of *441the jury’s response to the question of capital punishment. During the 1960’s juries returned in excess of a thousand death sentences, a rate of approximately two per week. Whether it is true that death sentences were returned in less than 10% of the cases as petitioners estimate or whether some higher percentage is more accurate,32 these totals simply do not support petitioners’ assertion at oral argument that “the death penalty is virtually unanimously repudiated and condemned by the conscience of contemporary society.” 33 It is also worthy of note that the annual rate of death sentences has remained relatively constant over the last 10 years and that the figure for 1970 — 127 sentences — is the highest annual total since 1961.34 It is true that the sentencing rate might be expected to rise, rather than remain constant, when the number of violent crimes increases as it has in this country.35 And it may be conceded that the constancy in these statistics indicates the unwillingness of juries to demand the ultimate penalty in many cases where it might be imposed. But these considerations fall short of indicating that juries are imposing the death penalty with such rarity as to justify this Court in reading into this circumstance a public rejection of capital punishment.36

*442One must conclude, contrary to petitioners’ submission, that the indicators most likely to reflect the public’s view — legislative bodies, state referenda and the juries which have the actual responsibility — do not support the contention that evolving standards of decency require total abolition of capital punishment.37 Indeed, *443the weight of the evidence indicates that the public generally has not accepted either the morality or the social merit of the views so passionately advocated by the articulate spokesmen for abolition. But however one may assess the amorphous ebb and flow of public opinion generally on this volatile issue, this type of inquiry lies at the periphery — not the core — of the judicial process in constitutional cases. The assessment of popular opinion is essentially a legislative, not a judicial, function.

V

Petitioners seek to salvage their thesis by arguing that the infrequency and discriminatory nature of the actual resort to the ultimate penalty tend to diffuse public opposition. We are told that the penalty is imposed exclusively on uninfluential minorities — “the poor and powerless, personally ugly and socially unacceptable.” 38 It is urged that this pattern of application assures that large segments of the public will be either uninformed or unconcerned and will have no reason to measure the punishment against prevailing moral standards.

Implicitly, this argument concedes the unsoundness of petitioners’ contention, examined above under Part IV, that objective evidence shows a present and widespread community rejection of the death penalty. It is now said, *444in effect, not that capital punishment presently offends our citizenry, but that the public would be offended if the penalty were enforced in a nondiscriminatory manner against a significant percentage of those charged with capital crimes, and if the public were thereby made aware of the moral issues surrounding capital punishment. Rather than merely registering the objective indicators on a judicial balance, we are asked ultimately to rest a far-reaching constitutional determination on a prediction regarding the subjective judgments of the mass of our people under hypothetical assumptions that may or may not be realistic.

Apart from the impermissibility of basing a constitutional judgment of this magnitude on such speculative assumptions, the argument suffers from other defects. If, as petitioners urge, we are to engage in speculation, it is not at all certain that the public would experience deep-felt revulsion if the States were to execute as many sentenced capital offenders this year as they executed in the mid-1930’s.39 It seems more likely that public reaction, rather than being characterized by undifferentiated rejection, would depend upon the facts and circumstances surrounding each particular case.

Members of this Court know, from the petitions and appeals that come before us regularly, that brutish and revolting murders continue to occur with disquieting frequency. Indeed, murders are so commonplace *445in our society that only the most sensational receive significant and sustained publicity. It could hardly be suggested that in any of these highly publicized murder cases — the several senseless assassinations or the too numerous shocking multiple murders that have stained this country’s recent history — the public has exhibited any signs of “revulsion” at the thought of executing the convicted murderers. The public outcry, as we all know, has been quite to the contrary. Furthermore, there is little reason to suspect that the public’s reaction would differ significantly in response to other less publicized murders. It is certainly arguable that many such murders, because of their senselessness or barbarousness, would evoke a public demand for the death penalty rather than a public rejection of that alternative. Nor is there any rational basis for arguing that the public reaction to any of these crimes would be muted if the murderer were “rich and powerful.” The demand for the ultimate sanction might well be greater, as a wealthy killer is hardly a sympathetic figure. While there might be specific cases in which capital punishment would be regarded as excessive and shocking to the conscience of the "community, it can hardly be argued that the public’s dissatisfaction with the penalty in particular cases would translate into a demand for absolute abolition.

In pursuing the foregoing speculation, I do not suggest that it is relevant to the appropriate disposition of these cases. The purpose of the digression is to indicate that judicial decisions cannot be founded on such speculations and assumptions, however appealing they may seem.

But the discrimination argument does not rest alone on a projection of the assumed effect on public opinion of more frequent executions. Much also is made of the undeniable fact that the death penalty has a greater impact on the lower economic strata of society, which *446include a relatively higher percentage of persons of minority racial and ethnic group backgrounds. The argument drawn from this fact is two-pronged. In part it is merely an extension of the speculative approach pursued by petitioners, %. e., that public revulsion is suppressed in callous apathy because the penalty does not affect persons from the white middle class which constitutes the majority in this country. This aspect, however, adds little to the infrequency rationalization for public apathy which I have found unpersuasive.

As Mr. Justice Marshall’s opinion today demonstrates, the argument does have a more troubling aspect. It is his contention that if the average citizen were aware of the disproportionate burden of capital punishment borne by the “poor, the ignorant, and the underprivileged,” he would find the penalty “shocking to his conscience and sense of justice” and would not stand for its further use. Ante, at 365-366, 369. This argument, like the apathy rationale, calls for further speculation on the part of the Court. It also illuminates the quicksands upon which we are asked to base this decision. Indeed, the two contentions seem to require contradictory assumptions regarding the public’s moral attitude toward capital punishment. The apathy argument is predicated on the assumption that the penalty is used against the less influential elements of society, that the public is fully aware of this, and that it tolerates use of capital punishment only because of a callous indifference to the offenders who are sentenced. Mr. Justice Marshall’s argument, on the other hand, rests on the contrary assumption that the public does not know against whom the penalty is enforced and that if the public were educated to this fact it would find the punishment intolerable. Ante, at 369. Neither assumption can claim to be an entirely accurate portrayal of public attitude; for some acceptance of capital punishment might be a conse*447quence of hardened apathy based on the knowledge of infrequent and uneven application, while for others acceptance may grow only out of ignorance. More significantly, however, neither supposition acknowledges what, for me, is a more basic flaw.

Certainly the claim is justified that this criminal sanction falls more heavily on the relatively impoverished and underprivileged elements of society. The “have-nots” in every society always have been subject to greater pressure to commit crimes and to fewer constraints than their more affluent fellow citizens. This is, indeed, a tragic byproduct of social and economic deprivation, but it is not an argument of constitutional proportions under the Eighth or Fourteenth Amendment. The same discriminatory impact argument could be made with equal force and logic with respect to those sentenced to prison terms. The Due Process Clause admits of no distinction between the deprivation of “life” and the deprivation of “liberty.” If discriminatory impact renders capital punishment cruel and unusual, it likewise renders invalid most of the prescribed penalties for crimes of violence. The root causes of the higher incidence of criminal penalties on “minorities and the poor” will not be cured by abolishing the system of penalties. Nor, indeed, could any society have a viable system of criminal justice if sanctions were abolished or ameliorated because most of those who commit crimes happen to be underprivileged. The basic problem results not from the penalties imposed for criminal conduct but from social and economic factors that have plagued humanity since the beginning of recorded history, frustrating all efforts to create in any country at any time the perfect society in which there are no “poor,” no “minorities” and no “underprivileged.” 40 *448The causes underlying this problem are unrelated to the constitutional issue before the Court.

Finally, yet another theory for abolishing the death penalty — reflected in varying degrees in each of the concurring opinions today — is predicated on the discriminatory impact argument. Quite apart from measuring the public’s acceptance or rejection of the death penalty under the “standards of decency” rationale, Mr. Justice Douglas finds the punishment cruel and unusual because it is “arbitrarily” invoked. He finds that “the basic theme of equal protection is implicit” in the Eighth Amendment, and that the Amendment is violated when jury sentencing may be characterized as arbitrary or discriminatory. Ante, at 249. While Mr. Justice Stewart does not purport to rely on notions of equal protection, he also rests primarily on what he views to. be a history of arbitrariness. Ante, at 309-310.41 Whatever may be the facts with respect to jury sentencing, this argument calls for a reconsideration of the “standards” aspects of the Court’s decision in McGautha v. California, 402 U. S. 183 (1971). Although that is the unmistakable thrust of these opinions today, I see no reason to reassess the standards question considered so carefully in Mr. Justice Harlan’s opinion for the Court *449last Term. Having so recently reaffirmed our historic dedication to entrusting the sentencing function to the jury’s “untrammeled discretion” (id., at 207), it is difficult to see how the Court can now hold the entire process constitutionally defective under the Eighth Amendment. For all of these reasons I find little merit in the various discrimination arguments, at least in the several lights in which they have been cast in these cases.

Although not presented by any of the petitioners today, a different argument, premised on the Equal Protection Clause, might well be made. If a Negro defendant, for instance, could demonstrate that members of his race were being singled out for more severe punishment than others charged with the same offense, a constitutional violation might be established. This was the contention made in Maxwell v. Bishop, 398 F. 2d 138 (CA8 1968), vacated and remanded on other grounds, 398 U. S. 262 (1970), in which the Eighth Circuit was asked to issue a writ of habeas corpus setting aside a death sentence imposed on a Negro defendant convicted of rape. In that case substantial statistical evidence was introduced tending to show a pronounced disproportion in the number of Negroes receiving death sentences for rape in parts of Arkansas and elsewhere in the South. That evidence was not excluded but was found to be insufficient to show discrimination in sentencing in Maxwell’s trial. Me. Justice Blackmun, then sitting on the Court of Appeals for the Eighth Circuit, concluded:

“The petitioner’s argument is an interesting one and we are not disposed to say that it could not have some validity and weight in certain situations. Like the trial court, however ... we feel that the argument does not have validity and pertinent application to Maxwell’s case.

*450“We are not yet ready to condemn and upset the result reached in every case of a Negro rape defendant in the State of Arkansas on the basis of broad theories of social and statistical injustice. . . .

“We do not say that there is no ground for suspicion that the death penalty for rape may have been discriminatorily applied over the decades in that large area of states whose statutes provide for it. There are recognizable indicators of this. But . . . improper state practice of the past does not automatically invalidate a procedure of the present. . . .” Id., at 146-148.

I agree that discriminatory application of the death penalty in the past, admittedly indefensible, is no justification for holding today that capital punishment is invalid in all cases in which sentences were handed out to members of the class discriminated against. But Maxwell does point the way to a means of raising the equal protection challenge that is more consonant with precedent and the Constitution’s mandates than the several courses pursued by today’s concurring opinions.

A final comment on the racial discrimination problem seems appropriate. The possibility of racial bias in the trial and sentencing process has diminished in recent years. The segregation of our society in decades past, which contributed substantially to the severity of punishment for interracial crimes, is now no longer prevalent in this country. Likewise, the day is past when juries do not represent the minority group elements of the community. The assurance of fair trials for all citizens is greater today than at any previous time in our history. Because standards of criminal justice have “evolved” in a manner favorable to the accused, discriminatory imposition of capital punishment is far less likely today than in the past.

*451VI

Petitioner in Branch v. Texas, No. 69-5031, and to a lesser extent the petitioners in the other cases before us today, urge that capital punishment is cruel and unusual because it no longer serves any rational legislative interests. Before turning to consider whether any of the traditional aims of punishment justify the death penalty, I should make clear the context in which I approach this aspect of the cases.

First, I find no support — in the language of the Constitution, in its history, or in the cases arising under it — for the view that this Court may invalidate a category of penalties because we deem less severe penalties adequate to serve the ends of penology. While the cases affirm our authority to prohibit punishments that are cruelly inhumane (e. g., Wilkerson v. Utah, 99 U. S., at 135-136; In re Kemmler, 136 U. S., at 447), and punishments that are cruelly excessive in that they are disproportionate to particular crimes (see Part VII, infra), the precedents of this Court afford no basis for striking- down a particular form of punishment because we may be persuaded that means less stringent would be equally efficacious.

Secondly, if we were free to question the justifications for the use of capital punishment, a heavy burden would rest on those who attack the legislatures’ judgments to prove the lack of rational justifications. This Court has long held that legislative decisions in this area, which lie within the special competency of that branch, are entitled to a presumption of validity. See, e. g., Trop v. Dulles, 356 U. S., at 103; Louisiana ex rel. Francis v. Resweber, 329 U. S., at 470 (Frankfurter, J., concurring) ; Weems v. United States, 217 U. S., at 378-379; In re Kemmler, 136 U. S., at 449.

*452I come now to consider, subject to the reservations above expressed, the two justifications most often cited for the retention of capital punishment. The concept of retribution — though popular for centuries — is now criticized as unworthy of a civilized people. Yet this Court has acknowledged the existence of a retributive element in criminal sanctions and has never heretofore found it impermissible. In Williams v. New York, 337 U. S. 241 (1949), Mr. Justice Black stated that,

“Retribution is no longer the dominant objective of the criminal law. Reformation and rehabilitation of offenders have become important goals of criminal jurisprudence.” Id., at 248.

It is clear, however, that the Court did not reject retribution altogether. The record in that case indicated that one of the reasons why the trial judge imposed the death penalty was his sense of revulsion at the “shocking details of the crime.” Id., at 244. Although his motivation was clearly retributive, the Court upheld the trial judge’s sentence.42 Similarly, Mr. Justice Marshall noted in his plurality opinion in Powell v. Texas, 392 U. S. 514, 530 (1968), that this Court “has never held that anything in the Constitution requires that penal sanctions be designed solely to achieve therapeutic or rehabilitative effects.”43

*453While retribution alone may seem an unworthy justification in a moral sense, its utility in a system of criminal justice requiring public support has long been recognized. Lord Justice Denning, now Master of the Rolls of the Court of Appeal in England, testified on this subject before the British Royal Commission on Capital Punishment:

“Many are inclined to test the efficacy of punishment solely by its value as a deterrent: but this is too narrow a view. Punishment is the way in which society expresses its denunciation of wrong doing: and, in order to maintain respect for law, it is essential that the punishment inflicted for grave crimes should adequately reflect the revulsion felt by the great majority of citizens for them. It is a mistake to consider the objects of punishment as being deterrent or reformative or preventive and nothing else. If this were so, we should not send to prison a man who was guilty of motor manslaughter, but only disqualify him from driving; but would public opinion be content with this? The truth is that some crimes are so outrageous that society insists on adequate punishment, because the wrong-doer deserves it, irrespective of whether it is a deterrent or not.” 44

The view expressed by Lord Denning was cited approvingly in the Royal Commission’s Report, recognizing “a *454strong and widespread demand for retribution.” 45 Mr. Justice Stewart makes much the same point in his opinion today when he concludes that expression of man’s retributive instincts in the sentencing process “serves an important purpose in promoting the stability of a society governed by law.” Ante, at 308. The view, moreover, is not without respectable support in the jurisprudential literature in this country,46 despite a substantial body of opinion to the contrary.47 And it is conceded on all sides that, not infrequently, cases arise that are so shocking or offensive that the public demands the ultimate penalty for the transgressor.

Deterrence is a more appealing justification, although opinions again differ widely. Indeed, the deterrence issue lies at the heart of much of the debate between the abolitionists and retentionists.48 Statistical studies, based primarily on trends in States that have abolished the penalty, tend to support the view that the death penalty has not been proved to be a superior deterrent.49 Some dispute the validity of this conclusion,50 pointing *455out that the studies do not show that the death penalty has no deterrent effect on any categories of crimes. On the basis of the literature and studies currently available, I find myself in agreement with the conclusions drawn by the Royal Commission following its exhaustive study of this issue:

“The general conclusion which we reach, after careful review of all the evidence we have been able to obtain as to the deterrent effect of capital punishment, may be stated as follows. Prima facie the penalty of death is likely to have a stronger effect as a deterrent to normal human beings than any other form of punishment, and there is some evidence (though no convincing statistical evidence) that this is in fact so. But this effect does not operate universally or uniformly, and there are many offenders on whom it is limited and may often be negligible. It is accordingly important to view this question in a just perspective and not base a penal policy in relation to murder on exaggerated estimates of the uniquely deterrent force of the death penalty.” 51

Only recently this Court was called on to consider the deterrence argument in relation to punishment by fines i for public drunkenness. Powell v. Texas, 392 U. S. 514 (1968). The Court was unwilling to strike down the Texas statute on grounds that it lacked a rational foundation. What Mb. Justice Marshall said there would seem to have equal applicability in this case:

“The long-standing and still raging debate over the validity of the deterrence justification for penal sanctions has not reached any sufficiently clear conclusions to permit it to be said that such sanctions are ineffective in any particular context or for any *456particular group of people who are able to appreciate the consequences of their acts. . . Id., at 531.

As I noted at the outset of this section, legislative judgments as to the efficacy of particular punishments are presumptively rational and may not be struck down under the Eighth Amendment because this Court may think that some alternative sanction would be more appropriate. Even if such judgments were within the judicial prerogative, petitioners have failed to show that there exist no justifications for the legislative enactments challenged in these cases.52 While the evidence and arguments advanced by petitioners might have proved profoundly persuasive if addressed to a legislative body, they do not approach the showing traditionally required before a court declares that the legislature has acted irrationally.

VII

In two of the cases before us today juries imposed sentences of death after convictions for rape.53 In these cases we are urged to hold that even if capital punishment is permissible for some crimes, it is a cruel and unusual punishment for this crime. Petitioners in these cases rely on the Court's opinions holding that the Eighth Amendment, in addition to prohibiting punishments *457deemed barbarous and inhumane, also condemns punishments that are greatly disproportionate to the crime charged. This reading of the Amendment was first expressed by Mr. Justice Field in his dissenting opinion in O’Neil v. Vermont, 144 U. S. 323, 337 (1892), a case in which a defendant charged with a large number of violations of Vermont’s liquor laws received a fine in excess of $6,600, or a 54-year jail sentence if the fine was not paid. The majority refused to consider the question on the ground that the Eighth Amendment did not apply to the States. The dissent, after carefully examining the history of that Amendment and the Fourteenth, concluded that its prohibition was binding on Vermont and that it was directed against “all punishments which by their excessive length or severity are greatly dispro-portioned to the offences charged.” Id., at 339-340.54

The Court, in Weems v. United States, 217 U. S. 349 (1910), adopted Mr. Justice Field’s view. The defendant, in Weems, charged with falsifying Government documents, had been sentenced to serve 15 years in cadena temporal, a punishment which included carrying chains at the wrists and ankles and the perpetual loss of the right to vote and hold office. Finding the sentence grossly excessive in length and condition of imprisonment, the Court struck it down. This notion of disproportionality — that particular sentences may be cruelly excessive for particular crimes — has been cited with approval in more recent decisions of this Court. See Robinson v. California, 370 U. S., at 667; Trop v. Dulles, 356 U. S., at 100; see also Howard v. Fleming, 191 U. S. 126, 135-136 (1903).

These cases, while providing a rationale for gauging the constitutionality of capital sentences imposed for rape, *458also indicate the existence of necessary limitations on the judicial function. The use of limiting terms in the various expressions of this test found in the opinions — grossly excessive, greatly disproportionate — emphasizes that the Court’s power to strike down punishments as excessive must be exercised with the greatest circumspection. As I have noted earlier, nothing in the history of the Cruel and Unusual Punishments Clause indicates that it may properly be utilized by the judiciary to strike down punishments — authorized by legislatures and imposed by juries — in any but the extraordinary case. This Court is not empowered to sit as a court of sentencing review, implementing the personal views of its members on the proper role of penology. To do so is to usurp a function committed to the Legislative Branch and beyond the power and competency of this Court.

Operating within these narrow limits, I find it quite impossible to declare the death sentence grossly excessive for all rapes. Rape is widely recognized as among the most serious of violent crimes, as witnessed by the very fact that it is punishable by death in 16 States and by life imprisonment in most other States.55 The several reasons why rape stands so high on the list of serious crimes are well known: It is widely viewed as the most atrocious of intrusions upon the privacy and dignity of the victim; never is the crime committed accidentally; rarely can it be said to be unpremeditated; *459often the victim suffers serious physical injury; the psychological impact can often be as great as the physical consequences; in a real sense, the threat of both types of injury is always present.56 For these reasons, and for the reasons arguing against abolition of the death penalty altogether, the excessiveness rationale provides no basis for rejection of the penalty for rape in all cases.

The argument that the death penalty for rape lacks rational justification because less severe punishments might be viewed as accomplishing the proper goals of penology is as inapposite here as it was in considering per se abolition. See Part VI supra. The state of knowledge with respect to the deterrent value of the sentence for this crime is inconclusive.57 Moreover, what has been said about the concept of retribution applies with equal force where the crime is rape. There are many cases in which the sordid, heinous nature of a particular crime, demeaning, humiliating, and often physically or psychologically traumatic, will call for public condemnation. In a period in our country’s history when the frequency of this crime is increasing alarmingly,58 it is indeed a grave event for the Court to take from the States whatever deterrent and retributive weight the death penalty retains.

Other less sweeping applications of the dispropor-tionality concept have been suggested. Recently the Fourth Circuit struck down a death sentence in Ralph v. Warden, 438 F. 2d 786 (1970), holding that the death penalty was an appropriate punishment for rape *460only where life is “endangered.” Chief Judge Hayns-worth, who joined in the panel’s opinion, wrote separately in denying the State of Maryland’s petition for rehearing in order to make clear the basis for his joinder. He stated that, for him, the appropriate test was not whether life was endangered, but whether the victim in fact suffered “grievous physical or psychological harm.” Id., at 794. See Rudolph v. Alabama, 375 U. S. 889 (1963) (dissent from the denial of certiorari).

It seems to me that both of these tests depart from established principles and also raise serious practical problems. How are those cases in which the victim’s life is endangered to be distinguished from those in which no danger is found? The threat of serious injury is implicit in the definition of rape; the victim is either forced into submission by physical violence or by the threat of violence. Certainly that test would provide little comfort for either of the rape defendants in the cases presently before us. Both criminal acts were accomplished only after a violent struggle. Petitioner Jackson held a scissors blade against his victim’s neck. Petitioner Branch had less difficulty subduing his 65-year-old victim. Both assailants threatened to kill their victims. See Mr. Justice Douglas’ opinion, ante, at 252-253. The alternate test, limiting the penalty to cases in which the victim suffers physical or emotional harm, might present even greater problems of application. While most physical effects may be seen and objectively measured, the emotional impact may be impossible to gauge at any particular point in time. The extent and duration of psychological trauma may not be known or ascertainable prior to the date of trial.

While I reject each of these attempts to establish specific categories of cases in which the death penalty may be deemed excessive, I view them as groping *461toward what is for me the appropriate application of the Eighth Amendment. While in my Anew the dispro-portionality test may not be used either to strike down the death penalty for rape altogether or to install the Court as a tribunal for sentencing review, that test may find its application in the peculiar circumstances of specific cases. Its utilization should be limited to the rare case in which the death penalty is rendered for a crime technically falling within the legislatively defined class but factually falling outside the likely legislative intent in creating the category. Specific rape cases (and specific homicides as well) can be imagined in which the conduct of the accused would render the ultimate penalty a grossly excessive punishment. Although this case-by-case approach may seem painfully slow and inadequate to those who wish the Court to assume an activist legislative role in reforming criminal punishments, it is the approach dictated both by our prior opinions and by a due recognition of the limitations of judicial power. This approach, rather than the majority’s more pervasive and less refined judgment, marks for me the appropriate course under the Eighth Amendment.

VIII

I now return to the overriding question in these cases: whether this Court, acting in conformity with the Constitution, can justify its judgment to abolish capital punishment as heretofore known in this country. It is important to keep in focus the enormity of the step undertaken by the Court today. Not only does it invalidate hundreds of state and federal laws, it deprives those jurisdictions of the power to legislate with respect to capital punishment in the future, except in a manner consistent with the cloudily outlined views of those Justices who do not purport to undertake total abolition. *462Nothing short of an amendment to the United States Constitution can reverse the Court’s judgments. Meanwhile, all flexibility is foreclosed. The normal democratic process, as well as the opportunities for the several States to respond to the will of their people expressed through ballot referenda (as in Massachusetts, Illinois, and Colorado),59 is now shut off.

The sobering disadvantage of constitutional adjudication of this magnitude is the universality and permanence of the judgment. The enduring merit of legislative action is its responsiveness to the democratic process, and to revision and change: mistaken judgments may be corrected and refinements perfected. In England 60 and Canada61 critical choices were made after studies canvassing all competing views, and in those countries revisions may be made in light of experience.62

As recently as 1967 a presidential commission did consider, as part of an overall study of crime in this country, whether the death penalty should be abolished. *463The commission’s unanimous recommendation was as follows:

“The question whether capital punishment is an appropriate sanction is a policy decision to be made by each State. Where it is retained, the types of offenses for which it is available should be strictly limited, and the law should be enforced in an evenhanded and nondiscriminatory manner, with procedures for review of death sentences that are fair and expeditious. When a State finds that it cannot administer the penalty in such a manner, or that the death penalty is being imposed but not carried into effect, the penalty should be abandoned.” 63

The thrust of the Commission’s recommendation, as presently relevant, is that this question “is a policy-decision to be made by each State.” There is no hint that this decision could or should be made by the judicial branch.

The National Commission on Reform of Federal Criminal Laws also considered the capital punishment issue. The introductory commentary of its final report states that “a sharp division [existed] within the Commission on the subject of capital punishment,” although a *464majority favored its abolition.64 Again, consideration of the question was directed to the propriety of retention or abolition as a legislative matter. There was no suggestion that the difference of opinion existing among commission members, and generally across the country, could or should be resolved in one stroke by a decision of this Court.65 Similar activity was, before today, evident at the state level with re-evaluation having been undertaken by special legislative committees in some States and by public ballot in others.66

With deference and respect for the views of the Justices who differ, it seems to me that all these studies — both in this country and elsewhere — suggest that, as a matter of policy and precedent, this is a classic case for the exercise of our oft-announced allegiance to judicial restraint. I know of no case in which greater gravity and delicacy have attached to the duty that this Court is called on to perform whenever legislation — state or federal — is challenged on constitutional grounds.67 It seems to me that the sweeping judicial action undertaken today reflects a *465basic lack of faith and confidence in the democratic process. Many may regret, as I do, the failure of some legislative bodies to address the capital punishment issue with greater frankness or effectiveness. Many might decry their failure either to abolish the penalty entirely or selectively, or to establish standards for its enforcement. But impatience with the slowness, and even the unresponsiveness, of legislatures is no justification for judicial'intrusion upon their historic powers. Rarely has there been a more appropriate opportunity for this Court to heed the philosophy of Mr. Justice Oliver Wendell Holmes. As Mr. Justice Frankfurter reminded the Court in Trop:

“[T]he whole of [Mr. Justice Holmes’] work during his thirty years of service on this Court should be a constant reminder that the power to invalidate legislation must not be exercised as if, either in constitutional theory or in the art of government, it stood as the sole bulwark against unwisdom or excesses of the moment.” 356 U. S., at 128.

Me. Justice Rehnquist,

with whom The Chief Justice, Mr. Justice Blackmun, and Me. Justice Powell join, dissenting.

The Court’s judgments today strike down a penalty that our Nation’s legislators have thought necessary since our country was founded. My Brothers Douglas, Brennan, and Marshall would at one fell swoop invalidate laws enacted by Congress and 40 of the 50 state legislatures, and would consign to the limbo of unconstitutionality under a single rubric penalties for offenses as varied and unique as murder, piracy, mutiny, highjacking, and desertion in the face of the enemy. My Brothers Stewart and White, asserting reliance on a more limited rationale — the reluctance of judges and juries actually to impose the death penalty in the majority of capital *466cases — join in the judgments in these cases. Whatever its precise rationale, today’s holding necessarily brings into sharp relief the fundamental question of the role of judicial review in a democratic society. How can government by the elected representatives of the people co-exist with the power of the federal judiciary, whose members are constitutionally insulated from responsiveness to the popular will, to declare invalid laws duly enacted by the popular branches of government?

The answer, of course, is found in Hamilton’s Federalist Paper No. 78 and in Chief Justice Marshall’s classic opinion in Marbury v. Madison, 1 Cranch 137 (1803). An oft-told story since then, it bears summarization once more. Sovereignty resides ultimately in the people as a whole and, by adopting through their States a written Constitution for the Nation and subsequently adding amendments to that instrument, they have both granted certain powers to the National Government, and denied other powers to the National and the State Governments. Courts are exercising no more than the judicial function conferred upon them by Art. Ill of the Constitution when they assess, in a case before them, whether or not a particular legislative enactment is within the authority granted by the Constitution to the enacting body, and whether it runs afoul of some limitation placed by the Constitution on the authority of that body. For the theory is that the people themselves have spoken in the Constitution, and therefore its commands are superior to the commands of the legislature, which is merely an agent of the people.

The Founding Fathers thus wisely sought to have the best of both worlds, the undeniable benefits of both democratic self-government and individual rights protected against possible excesses of that form of govern merit

The courts in cases properly before them have been entrusted under the Constitution with the last word, short of constitutional amendment, as to whether a law passed *467by the legislature conforms to the Constitution. But just because courts in general, and this Court in particular, do have the last word, the admonition of Mr. Justice Stone dissenting in United States v. Butler must be constantly borne in mind:

“[W]hile unconstitutional exercise of power by the executive and legislative branches of the government is subject to judicial restraint, the only check upon our own exercise of power is our own sense of self-restraint." 297 U. S. 1, 78-79 (1936).

Rigorous attention to the limits of this Court’s authority is likewise enjoined because of the natural desire that beguiles judges along with other human beings into imposing their own views of goodness, truth, and justice upon others. Judges differ only in that they have the power, if not the authority, to enforce their desires. This is doubtless why nearly two centuries of judicial precedent from this Court counsel the sparing use of that power. The most expansive reading of the leading constitutional cases does not remotely suggest that this Court has been granted a roving commission, either by the Founding Fathers or by the framers of the Fourteenth Amendment, to strike down laws that are based upon notions of policy or morality suddenly found unacceptable by a majority of this Court. The Framers of the Constitution would doubtless have agreed with the great English political philosopher John Stuart Mill when he observed:

“The disposition of mankind, whether as rulers or as fellow-citizens, to impose their own opinions and inclinations as a rule of conduct on others, is so energetically supported by some of the best and by some of the worst feelings incident to human nature, that it is hardly ever kept under restraint by anything but want of power.” On Liberty 28 (1885).

*468A separate reason for deference to the legislative judgment is the consequence of human error on the part of the judiciary with respect to the constitutional issue before it. Human error there is bound to be, judges being men and women, and men and women being what they are. But an error in mistakenly sustaining the constitutionality of a particular enactment, while wrongfully depriving the individual of a right secured to him by the Constitution, nonetheless does so by simply letting stand a duly enacted law of a democratically chosen legislative body. The error resulting from a mistaken upholding of an individual’s constitutional claim against the validity of a legislative enactment is a good deal more serious. For the result in such a case is not to leave standing a law duly enacted by a representative assembly, but to impose upon the Nation the judicial fiat of a majority of a court of judges whose connection with the popular will is remote at best.

The task of judging constitutional cases imposed by Art. Ill cannot for this reason be avoided, but it must surely be approached with the deepest humility and genuine deference to legislative judgment. Today’s decision to invalidate capital punishment is, I respectfully submit, significantly lacking in those attributes. For the reasons well stated in the opinions of The Chief Justice, Mr. Justice Blackmun, and Mr. Justice Powell, I conclude that this decision holding unconstitutional capital punishment is not an act of judgment, but rather an act of will. It completely ignores the strictures of Mr. Justice Holmes, writing more than 40 years ago in Baldwin v. Missouri:

“I have not yet adequately expressed the more than anxiety that I feel at the ever increasing scope given to the Fourteenth Amendment in cutting down what I believe to be the constitutional rights of the States. As the decisions now stand, I see hardly *469any limit but the sky to the invalidating of those rights if they happen to strike a majority of this Court as for any reason undesirable. I cannot believe that the Amendment was intended to give us carte blanche to embody our economic or moral beliefs in its prohibitions. Yet I can think of no narrower reason that seems to me to justify the present and the earlier decisions to which I have referred. Of course the words 'due process of law/ if taken in their literal meaning, have no application to this case; and while it is too late to deny that they have been given a much more extended and artificial signification, still we ought to remember the great caution shown by the Constitution in limiting the power of the States, and should be slow to construe the clause in the Fourteenth Amendment as committing to the Court, with no guide but the Court’s own discretion, the validity of whatever laws the States may pass.” 281 U. S. 586, 595 (1930) (dissenting opinion).

More than 20 years ago, Justice Jackson made a similar observation with respect to this Court’s restriction of the States in the enforcement of their own criminal laws:

“The use of the due process clause to disable the States in protection of society from crime is quite as dangerous and delicate a use of federal judicial power as to use it to disable them from social or economic experimentation.” Ashcraft v. Tennessee, 322 U. S. 143, 174 (1944) (dissenting opinion).

If there can be said to be one dominant theme in the Constitution, perhaps more fully articulated in the Federalist Papers than in the instrument itself, it is the notion of checks and balances. The Framers were well aware of the natural desire of office holders as well as others to seek to expand the scope and authority of their *470particular office at the expense of others. They sought to provide against success in such efforts by erecting adequate checks and balances in the form of grants of authority to each branch of the government in order to counteract and prevent usurpation on the part of the others.

This philosophy of the Framers is best described by one of the ablest and greatest of their number, James Madison, in Federalist No. 51:

“In framing a government which is to be administered by men over men, the great difficulty lies in this: You must first enable the government to controul the governed; and in the next place, oblige it to controul itself.”

Madison’s observation applies to the Judicial Branch with at least as much force as ti> the Legislative and Executive Branches. While overreaching by the Legist lative and Executive Branches may result in the sacrifice of individual protections that the Constitution was designed to secure against action of the State, judicial overreaching may result in sacrifice of the equally important right of the people ti> govern themselves. The Due Process and Equal Protection Clauses of the Fourteenth Amendment were “never intended to destroy the States’ power to govern themselves.” Black, J., in Oregon v. Mitchell, 400 U. S. 112, 126 (1970).

The very nature of judicial review, as pointed out by Justice Stone in his dissent in the Butler case, makes the courts the least subject to Madisonian check in the'event that they shall, for the best of motives, expand judicial authority beyond the limits contemplated by the Framers. It is for this reason that judicial self-restraint is surely an implied, if not an expressed, condition of the grant of authority of judicial review. The Court’s holding in these cases has been reached, I believe, in complete disregard of that implied condition.

Judgment of the Court, and opinion of

Mr. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens, announced by Mr. Justice Powell.

The issue presented by this case is whether the imposition of the sentence of death for the crime of murder under the law of Florida violates the Eighth and Fourteenth Amendments.

I

The petitioner, Charles William Proffitt, was tried, found guilty, and sentenced to death for the first-degree *245murder of Joel Medgebow. The circumstances surrounding the murder were testified to by the decedent’s wife, who was present at the time it was committed. On July 10, 1973, Mrs. Medgebow awakened around 5 a. m. in the bedroom of her apartment to find her husband sitting up in bed, moaning. He was holding what she took to be a ruler.1 Just then a third person jumped up, hit her several times with his fist, knocked her to the. floor, and ran out of the house. It soon appeared that Medgebow had been fatally stabbed with a butcher knife. Mrs. Medgebow was not able to identify the attacker, although she was able to give a description of him.2

The petitioner’s wife testified that on the night before the murder the petitioner had gone to work dressed in a white shirt and gray pants, and that he had returned at about 5:15 a. m. dressed in the same clothing but without shoes. She said that after a short conversation the petitioner had packed his clothes and departed. A young woman boarder, who overheard parts of the petitioner’s conversation with his wife, testified that the petitioner had told his wife that he had stabbed and killed a man with a butcher knife while he was burglarizing a place, and that he had beaten a woman. One of the petitioner’s coworkers testified that they had been drinking together until 3:30 or 3:45 on the morning of the murder and that the petitioner had then driven him home. He said that the petitioner at this time was wearing gray pants and a white shirt.

The jury found the defendant guilty as charged. Sub*246sequently, as provided by Florida law, a separate hearing was held to determine whether the petitioner should be sentenced to death or to life imprisonment. Under the state law that decision turned on whether certain statutory aggravating circumstances surrounding the crime outweighed any statutory mitigating circumstances found to exist.3 At that hearing it was shown that the petitioner had one prior conviction, a 1967 charge of breaking and entering. The State also introduced the testimony of the physician (Dr, Crumbley) at the jail where the petitioner had been held pending trial. He testified that the petitioner had come to him as a physician, and told him that he was concerned that he would harm other people in the future, that he had had an uncontrollable desire to kill that had already resulted in his killing one man, that this desire was building up again, and that he wanted psychiatric help so he would not kill again. Dr. Crumbley also testified that, in his opinion, the petitioner was dangerous and would be a danger to his fellow inmates if imprisoned, but that his condition could be treated successfully.

The jury returned an advisory verdict recommending the sentence of death. The trial judge ordered an independent psychiatric evaluation of the petitioner, the results of which indicated that the petitioner was not, then or at the time of the murder, mentally impaired. The judge then sentenced the petitioner to death. In his written findings supporting the sentence, the judge found as aggravating circumstances that (1) the murder was premeditated and occurred in the course of a felony (burglary); (2) the petitioner has the propensity to commit murder; (3) the murder was especially heinous, atrocious, and cruel; and (4) the petitioner knowingly, through his intentional act, created a great risk of serious *247bodily harm and death to many persons. The judge also found specifically that none of the statutory mitigating circumstances existed. The Supreme Court of Florida affirmed. 315 So. 2d 461 (1975). We granted certiorari, 423 U. S. 1082 (1976), to consider whether the imposition of the death sentence in this case constitutes cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments.

II

The petitioner argues that the imposition of the death penalty under any circumstances is cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments. We reject this argument for the reasons stated today in Gregg v. Georgia, ante, at 168-187.

III

A

In response to Furman v. Georgia, 408 U. S. 238 (1972), the Florida Legislature adopted new statutes that authorize the imposition of the death penalty on those convicted of first-degree murder. Fla. Stat. Ann. § 782.04 (1) (Supp. 1976-1977).4 At the same time Florida *248adopted a new capital-sentencing procedure, patterned in large part on the Model Penal Code. See § 921.141 (Supp. 1976-1977).5 Under the new statute, if a defendant is found guilty of a capital offense, a separate evi-dentiary hearing is held before the trial judge and jury to determine his sentence. Evidence may be presented on any matter the judge deems relevant to sentencing and must include matters relating to certain legislatively specified aggravating and mitigating circumstances. Both the prosecution and the defense may present argument on whether the death penalty shall be imposed.

At the conclusion of the hearing the jury is directed to consider “[wjhether sufficient mitigating circumstances exist . . . which outweigh the aggravating circumstances found to exist; and . . . [b]ased on these considerations, whether the defendant should be sentenced to life [imprisonment] or death.” §§ 921.141 (2) (b) and (c) (Supp. 1976-1977).6 The jury’s verdict is determined by *249majority vote. It is only advisory; the actual sentence is determined by the trial judge. The Florida Supreme Court has stated, however, that “[i]n order to sustain a sentence of death following a jury recommendation of life, the facts suggesting a sentence of death should be so clear and convincing that virtually no reasonable person could differ.” Tedder v. State, 322 So. 2d 908, 910 (1975). Accord, Thompson v. State, 328 So. 2d 1, 5 *250(1976). Cf. Spinkellink v. State, 313 So. 2d 666, 671 (1976).7

The trial judge is also directed to weigh the statutory aggravating and mitigating circumstances when he determines the sentence to be imposed on a defendant. The statute requires that if the trial court imposes a sentence of death, “it shall set forth in writing its findings upon which the sentence of death is based as to the facts: (a) [t]hat sufficient [statutory] aggravating circumstances exist . . . and (b) [t]hat there are insufficient [statutory] mitigating circumstances ... to outweigh the aggravating circumstances.” § 921.141 (3) (Supp. 1976-1977).8

The statute provides for automatic review by the Supreme Court of Florida of all cases in which a death sentence has been imposed. § 921.141 (4) (Supp. 1976-1977). The law differs from that of Georgia in that it does *251not require the court to conduct any specific form of review. Since, however, the trial judge must justify the imposition of a death sentence with written findings, meaningful appellate review of each such sentence is made possible, and the Supreme Court of Florida, like its Georgia counterpart, considers its function to be to “[guarantee] that the [aggravating and mitigating] reasons present in one case will reach a similar result to that reached under similar circumstances in another case. . . . If a defendant is sentenced to die, this Court can review that case in light of the other decisions and determine whether or not the punishment is too great.” State v. Dixon, 283 So. 2d 1, 10 (1973).

On their face these procedures, like those used in Georgia, appear to meet the constitutional deficiencies identified in Furman. The sentencing authority in Florida, the trial judge, is directed to weigh eight aggravating factors against seven mitigating factors to determine whether the death penalty shall be imposed. This determination requires the trial judge to focus on the circumstances of the crime and the character of the individual defendant. He must, inter alia, consider whether the defendant has a prior criminal record, whether the defendant acted under duress or under the influence of extreme mental or emotional disturbance, whether the defendant’s role in the crime was that of a minor accomplice, and whether the defendant’s youth argues in favor of a more lenient sentence than might otherwise be imposed. The trial judge must also determine whether the crime was committed in the course of one of several enumerated felonies, whether it was committed for pecuniary gain, whether it was committed to assist in an escape from custody or to prevent a lawful arrest, and whether the crime was especially heinous, atrocious, or cruel. To answer these questions, which are not un*252like those considered by a Georgia sentencing jury, see Gregg v. Georgia, ante, at 197, the sentencing judge must focus on the individual circumstances of each homicide and each defendant.

The basic difference between the Florida system and the Georgia system is that in Florida the sentence is determined by the trial judge rather than by the jury.9 This Court has pointed out that jury sentencing in a capital case can perform an important societal function, Witherspoon v. Illinois, 391 U. S. 510, 519 n. 15 (1968), but it has never suggested that jury sentencing is constitutionally required. And it would appear that judicial sentencing should lead, if anything, to even greater consistency in the imposition at the trial court level of capital punishment, since a trial judge is more experienced in sentencing than a jury, and therefore is better able to impose sentences similar to those imposed in analogous cases.10

The Florida capital-sentencing procedures thus seek to *253assure that the death penalty will not be imposed in an arbitrary or capricious manner. Moreover, to the extent that any risk to the contrary exists, it is minimized by Florida's appellate review system, under which the evidence of the aggravating and mitigating circumstances is reviewed and reweighed by the Supreme Court of Florida “to determine independently whether the imposition of the ultimate penalty is warranted.” Songer v. State, 322 So. 2d 481, 484 (1975). See also Sullivan v. State, 303 So. 2d 632, 637 (1974). The Supreme Court of Florida, like that of Georgia, has not hesitated to vacate a death sentence when it has determined that the sentence should not have been imposed. Indeed, it has vacated 8 of the 21 death sentences that it has reviewed to date. See Taylor v. State, 294 So. 2d 648 (1974); Lamadline v. State, 303 So. 2d 17 (1974); Slater v. State, 316 So. 2d 539 (1975); Swan v. State, 322 So. 2d 485 (1975); Tedder v. State, 322 So. 2d 908 (1975); Halliwell v. State, 323 So. 2d 557 (1975); Thompson v. State, 328 So. 2d 1 (1976); Messer v. State, 330 So. 2d 137 (1976).

Under Florida's capital-sentencing procedures, in sum, trial judges are given specific and detailed guidance to assist them in deciding whether to impose a death penalty or imprisonment for life. Moreover, their decisions are reviewed to ensure that they are consistent with other sentences imposed in similar circumstances. Thus, in Florida, as in Georgia, it is no longer true that there is “ 'no meaningful basis for distinguishing the few cases in which [the death penalty] is imposed from the many cases in which it is not.’ ” Gregg v. Georgia, ante, at 188, quoting Furman v. Georgia, 408 U. S., at 313 (White, J., concurring). On its face the Florida system thus satisfies the constitutional deficiencies identified in Furman.

*254B

As in Gregg, the petitioner contends, however, that, while perhaps facially acceptable, the new sentencing procedures in actual effect are merely cosmetic, and that arbitrariness and caprice still pervade the system under which Florida imposes the death penalty.

(1)

The petitioner first argues that arbitrariness is inherent in the Florida criminal justice system because it allows discretion to be exercised at each stage of a criminal proceeding — the prosecutor’s decision whether to charge a capital offense in the first place, his decision whether to accept a plea to a lesser offense, the jury’s consideration of lesser included offenses, and, after conviction and unsuccessful appeal, the Executive’s decision whether to commute a death sentence. As we noted in Gregg, this argument is based on a fundamental misinterpretation of Furman, and we reject it for the reasons expressed in Gregg. See ante, at 199.

(2)

The petitioner next argues that the new Florida sentencing procedures in reality do not eliminate the arbitrary infliction of death that was condemned in Furman. Basically he contends that the statutory aggravating and mitigating circumstances are vague and overbroad,11 and that the statute gives no guidance as to how the mitigating and aggravating circumstances should be weighed in any specific case.

*255(a)

Initially the petitioner asserts that the enumerated aggravating and mitigating circumstances are so vague and so broad that virtually “any capital defendant becomes a candidate for the death penalty . . . In particular, the petitioner attacks the eighth and third statutory aggravating circumstances, which authorize the death penalty to be imposed if the crime is “especially heinous, atrocious, or cruel,” or if “[t]he defendant knowingly created a great risk of death to many persons.” §§ 921.141 (5)(h), (c) (Supp. 1976-1977). These provisions must be considered as they have been construed by the Supreme Court of Florida.

That court has recognized that while it is arguable “that all killings are atrocious, . . . [s] till, we believe that the Legislature intended something 'especially’ heinous, atrocious or cruel when it authorized the death penalty for first degree murder.” Tedder v. State, 322 So. 2d, at 910. As a consequence, the court has indicated that the eighth statutory provision is directed only at “the conscienceless or pitiless crime which is unnecessarily torturous to the victim.” State v. Dixon, 283 So. 2d, at 9. See also Alford v. State, 307 So. 2d 433, 445 (1975); Halliwell v. State, supra, at 561.12 We *256cannot say that the provision, as so construed, provides inadequate guidance to those charged with the duty of recommending or imposing sentences in capital cases. See Gregg v. Georgia, ante, at 200-203.

In the only case, except for the instant case, in which the third aggravating factor — “[t]he defendant knowingly created a great risk of death to many persons”— was found, Alvord v. State, 322 So. 2d 533 (1975), the State Supreme Court held that the defendant created a great risk of death because he “obviously murdered two of the victims in order .to avoid a surviving witness to the [first] murder.” Id., at 540.13 As construed by the Supreme Court of Florida these provisions are not impermissibly vague.14

*257(b)

The petitioner next attacks the imprecision of the mitigating circumstances. He argues that whether a defendant acted “under the influence of extreme mental or emotional disturbance,” whether a defendant’s capacity “to conform his conduct to the requirements of law was substantially impaired,” or whether a defendant’s participation as an accomplice in a capital felony was “relatively minor,” are questions beyond the capacity of a jury or judge to determine. See §§ 921.141 (6)(b), (f), (d) (Supp. 1976-1977).

He also argues that neither a jury nor a judge is capable of deciding how to weigh a defendant’s age or determining whether he had a “significant history of prior criminal activity.” See §§ 921.141 (6) (g)-, (a) (Supp. 1976-1977). In a similar vein the petitioner argues that it is not possible to make a rational determination whether there are “sufficient” aggravating circumstances that are not outweighed by the mitigating circumstances, since the state law assigns no specific weight to any of the various circumstances to be considered. See § 921.141 (Supp. 1976-1977).

While these questions and decisions may be hard, they require no more line drawing than is commonly required of a factfinder in a lawsuit. For example, juries have traditionally evaluated the validity of defenses such as insanity or reduced capacity, both of which involve the same considerations as some of the above-mentioned *258mitigating circumstances. While the various factors to be considered by the sentencing authorities do not have numerical weights assigned to them, the requirements of Furman are satisfied when the sentencing authority’s discretion is guided and channeled by requiring examination of specific factors that argue in favor of or against imposition of the death penalty, thus eliminating total arbitrariness and capriciousness in its imposition.

The directions given to judge and jury by the Florida statute are sufficiently clear and precise to enable the various aggravating circumstances to be weighed against the mitigating ones. As a result, the trial court’s sentencing discretion is guided and channeled by a system that focuses on the circumstances of each individual homicide and individual defendant in deciding whether the death penalty is to be imposed.

(c)

Finally, the Florida statute has a provision designed to assure that the death penalty will not be imposed on a capriciously selected group of convicted defendants. The Supreme Court of Florida reviews each death sentence to ensure that similar results are reached in similar cases.15

Nonetheless the petitioner attacks the Florida appellate review process because the role of the Supreme Court of Florida in reviewing death sentences is necessarily subjective and unpredictable. While it may be true that that court has not chosen to formulate a rigid objective test as its standard of review for all cases, it does not follow that the appellate review process is ineffective or arbitrary. In fact, it is apparent that the Florida court has undertaken responsibly to perform its function of death sentence review with a maximum of *259rationality and consistency. For example, it has several times compared the circumstances of a case under review with those of previous cases in which it has assessed the imposition of death sentences. See, e. g., Alford v. State, 307 So. 2d, at 445; Alvord v. State, 322 So. 2d, at 540-541. By following this procedure the Florida court has in effect adopted the type of proportionality review mandated by the Georgia statute. Cf. Gregg v. Georgia, ante, at 204-206. And any suggestion that the Florida court engages in only cursory or rubber-stamp review of death penalty cases is totally controverted by the fact that it has vacated over one-third of the death sentences that have come before it. See supra, at 253.16

IV

Florida, like Georgia, has responded to Furman by enacting legislation that passes constitutional muster. That legislation provides that after a person is convicted of first-degree murder, there shall be an informed, focused, guided, and objective inquiry into the question whether he should be sentenced to death. If a death sentence is imposed, the sentencing authority articulates in writing the statutory reasons that led to its decision. Those reasons, and the evidence supporting them, are conscientiously reviewed by a court which, because of *260its statewide jurisdiction, can assure consistency, fairness, and rationality in the evenhanded operation of the state law. As in Georgia, this system serves to assure that sentences of death will not be “wantonly” or “freakishly” imposed. See Furman v. Georgia, 408 U. S., at 310 (Stewart, J., concurring). Accordingly, the judgment before us is affirmed.

It is so ordered.

[For dissenting opinion of Mr. Justice Brennan, see ante, p. 227.]

[For dissenting opinion of Mr. Justice Marshall, see ante, p. 231.]

Mr. Justice White,

with whom The Chief Justice and Mr. Justice Rehnquist join,

concurring in the judgment.

There is no need to repeat the statement of the facts of this case and of the statutory procedure under which the death penalty was imposed, both of which are described in detail in the opinion of Mr. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens. I agree with them, see Parts III-B (2) (a) and (b), ante, at 255-258, that although the statutory aggravating and mitigating circumstances are not susceptible of mechanical application, they are by no means so vague and over-broad as to leave the discretion of the sentencing authority unfettered. Under Florida law, the sentencing judge is required to impose the death penalty on all first-degree murderers as to whom the statutory aggravating factors outweigh the mitigating factors. There is good reason to anticipate, then, that as to certain categories of murderers, the penalty will not be imposed freakishly or rarely but will be imposed with regularity; and consequently it cannot be said that the death penalty in *261Florida as to those categories has ceased “to be a credible deterrent or measurably to contribute to any other end of punishment in the criminal justice system.” Furman v. Georgia, 408 U. S. 238, 311 (1972) (White, J., concurring) . Accordingly, the Florida statutory scheme for imposing the death penalty does Pot run afoul of this Court’s holding in Furman v. Georgia.

For the reasons set forth in my opinion concurring in the judgment in Gregg v. Georgia, ante, at 224-225, and my dissenting opinion in Roberts v. Louisiana, post, at 348-350, this conclusion is not undercut by the possibility that some murderers may escape the death penalty solely through exercise of prosecutorial discretion or executive clemency. For the reasons set forth in my dissenting opinion in Roberts v. Louisiana, post, at 350-356, I also reject petitioner’s argument that under the Eighth Amendment the death penalty may never be imposed under any circumstances.

I concur in the judgment of affirmance.

Mr. Justice Blackmun,

concurring in the judgment.

I concur in the judgment. See Furman v. Georgia, 408 U. S. 238, 405-414 (1972) (Blackmun, J., dissenting), and id., at 375, 414, and 465.

Judgment of the Court, and opinion of

Mr. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens, announced by Mr. Justice Stevens.

The issue in this case is whether the imposition of the sentence of death for the crime of murder under the law of Texas violates the Eighth and Fourteenth Amendments to the Constitution.

I

The petitioner in this case, Jerry Lane Jurek, was charged by indictment with the killing of Wendy Adams *265“by choking and strangling her with his hands, and by drowning her in water by throwing her into a river . . . in the course of committing and attempting to commit kidnapping of and forcible rape upon the said Wendy Adams.” 1

*266The evidence at his trial consisted of incriminating statements made by the petitioner,2 the testimony of several people who saw the petitioner and the deceased during the day she was killed, and certain technical evidence. This evidence established that the petitioner, 22 years old at the time, had been drinking beer in the afternoon. He and two young friends later went driving together in his old pickup truck. The petitioner expressed a desire for sexual relations with some young girls they saw, but one of his companions said the girls were too young. The petitioner then dropped his two friends off at a pool hall. He was next seen talking to Wendy, who was 10 years old, at a public swimming pool where her grandmother had left her to swim. Other witnesses testified that they later observed a man resembling the petitioner driving an old pickup truck through town at a high rate of speed, with a young blond girl standing screaming in the bed of the truck. The last witness who saw them heard the girl crying “help me, *267help me.” The witness tried to follow them, but lost them in traffic. According to the petitioner’s statement, he took the girl to the river, choked her,3 and threw her unconscious body in the river. Her drowned body was found downriver two days later.

At the conclusion of the trial the jury returned a verdict of guilty.

Texas law requires that if a defendant has been convicted of a capital offense, the trial court must conduct a separate sentencing proceeding before the same jury that tried the issue of guilt. Any relevant evidence may be introduced at this proceeding, and both prosecution and defense may present argument for or against the sentence of death. The jury is then presented with two (sometimes three) questions,4 the answers to which determine whether a death sentence will be imposed.

During the punishment phase of the petitioner’s trial, several witnesses for the State testified to the petitioner’s bad reputation in the community. The petitioner’s father countered with testimony that the petitioner had always been steadily employed since he had left school and that he contributed to his family’s support.

The jury then considered the two statutory questions relevant to this case: (1) whether the evidence established beyond a reasonable doubt that the murder of the deceased was committed deliberately and with the reasonable expectation that the death of the deceased or another would result, and (2) whether the evidence established beyond a reasonable doubt that there was *268a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society. The jury unanimously answered “yes” to both questions, and the judge, therefore, in accordance with the statute, sentenced the petitioner to death. The Court of Criminal Appeals of Texas affirmed the judgment. . 522 S. W. 2d 934 (1975).

We granted certiorari, 423 U. S. 1082, to consider whether the imposition of the death penalty in this case violates the Eighth and Fourteenth Amendments of the United States Constitution.

II

The petitioner argues that the imposition of the death penalty under any circumstances is cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments. We reject this argument for the reasons stated today in Gregg v. Georgia, ante, at 168-187.

III

A

After this Court held Texas’ system for imposing capital punishment unconstitutional in Branch v. Texas, decided with Furman v. Georgia, 408 U. S. 238 (1972), the Texas Legislature narrowed the scope of its laws relating to capital punishment. The new Texas Penal Code limits capital homicides to intentional and knowing murders committed in five situations: murder of a peace officer or fireman; murder committed in the course of kidnaping, burglary, robbery, forcible rape, or arson; murder committed for remuneration; murder committed while escaping or attempting to escape from a penal institution; and murder committed by a prison inmate when the victim is a prison employee. See Tex. Penal Code § 19.03 (1974).

*269In addition, Texas adopted a new capital-sentencing procedure. See Tex. Code Crim. Proc., Art. 37.071 (Supp. 1975-1976). That procedure requires the jury to answer three questions in a proceeding that takes place subsequent to the return of a verdict finding a person guilty of one of the above categories of murder. The questions the jury must answer are these:

“(1) whether the conduct of the defendant that caused the death of the deceased was committed deliberately and with the reasonable expectation that the death of the deceased or another would result;

“(2) whether there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society; and

“(3) if raised by the evidence, whether the conduct of the defendant in killing the deceased was unreasonable in response to the provocation, if any, by the deceased.” Art. 37.071 (b) (Supp. 1975-1976).

If the jury finds that the State has proved beyond a reasonable doubt that the answer to each of the three questions is yes, then the death sentence is imposed. If the jury finds that the answer to any question is no, then a sentence of life imprisonment results. Arts. 37.071 (c), (e) (Supp. 1975-1976).5 The law also provides for an expedited review by the Texas Court of Criminal Appeals. See Art. 37.071 (f) (Supp. 1975-1976).

*270The Texas Court of Criminal Appeals has thus far affirmed only two judgments imposing death sentences under its post-Forman law — in this case and in Smith v. State, No. 49,809 (Feb. 18, 1976) (rehearing pending; initially reported in advance sheet for 534 S. W. 2d but subsequently withdrawn from bound volume). In the present case the state appellate court noted that its law "limits the circumstances under which the State may seek the death penalty to a small group of narrowly defined and particularly brutal offenses. This insures that the death penalty will only be imposed for the most serious crimes [and] . . . that [it] will only be imposed for the same type of offenses which occur under the same types of circumstances.” 522 S. W. 2d, at 939.

While Texas has not adopted a list of statutory aggravating circumstances the existence of which can justify the imposition of the death penalty as have Georgia and Florida, its action in narrowing the categories of murders for which a death sentence may ever be imposed serves much the same purpose. See McGautha v. California, 402 U. S. 183, 206 n. 16 (1971); Model Penal Code § 201.6, Comment 3, pp. 71-72 (Tent. Draft No. 9, 1959). In fact, each of the five classes of murders made capital by the Texas statute is encompassed in Georgia and Florida by one or more of their statutory aggravating circumstances. For example, the Texas statute requires the jury at the guilt-determining stage to consider whether the crime was committed in the course of a particular felony, whether it was committed for hire, or whether the defendant was an inmate of a penal institution at the time of its commission. Cf. Gregg v. Georgia, ante, at 165-166, n. 9; Proffitt v. Florida, ante, at 248-249, n. 6. Thus, in essence, the Texas statute requires that the jury find the existence of a statutory aggravating circumstance before the death penalty may be imposed. *271So far as consideration of aggravating circumstances is concerned, therefore, the principal difference between Texas and the other two States is that the death penalty is an available sentencing option — even potentially — for a smaller class of murders in Texas. Otherwise the statutes are similar. Each requires the sentencing authority to focus on the particularized nature of the crime.

But a sentencing system that allowed the jury to consider only aggravating circumstances would almost certainly fall short of providing the individualized sentencing determination that we today have held in Woodson v. North Carolina, post, at 303-305, to be required by the Eighth and Fourteenth Amendments. For such a system would approach the mandatory laws that we today hold unconstitutional in Woodson and Roberts v. Louisiana, post, p. 325.6 A jury must be allowed to consider on the basis of all relevant evidence not only why a death sentence should be imposed, but also why it should not be imposed.

Thus, in order to meet the requirement of the Eighth and Fourteenth Amendments, a capital-sentencing system must allow the sentencing authority to consider mitigating circumstances. In Gregg v. Georgia, we today hold constitutionally valid a capital-sentencing system *272that directs the jury to consider any mitigating factors, and in Proffitt v. Florida we likewise hold constitutional a system that directs the judge and advisory jury to consider certain enumerated mitigating circumstances. The Texas statute does not explicitly speak of mitigating circumstances; it directs only that the jury answer three questions. Thus, the constitutionality of the Texas procedures turns on whether the enumerated questions allow consideration of particularized mitigating factors.

The second Texas statutory question 7 asks the jury to determine “whether there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society” if he were not sentenced to death. The Texas Court of Criminal Appeals has yet to define precisely the meanings of such terms as “criminal acts of violence” or “continuing threat to society.” In the present case, however, it indicated that it will interpret this second question so as to allow a defendant to bring to the jury's attention whatever mitigating circumstances he may be able to show:

“In determining the likelihood that the defendant would be a continuing threat to society, the jury *273could consider whether the defendant had a significant criminal record. It could consider the range and severity of his prior criminal conduct. It could further look to the age of the defendant and whether or not at the time of the commission of the offense he was acting under duress or under the domination of another. It could also consider whether the defendant was under an extreme form of mental or emotional pressure, something less, perhaps, than insanity, but more than the emotions of the average man, however inflamed, could withstand.” 522 S. W. 2d, at 939-940.

In the only other case in which the Texas Court of Criminal Appeals has upheld a death sentence, it focused on the question of whether any mitigating factors were present in the case. See Smith v. State, No. 49,809 (Feb. 18, 1976). In that case the state appellate court examined the sufficiency of the evidence to see if a “yes” answer to question 2 should be sustained. In doing so it examined the defendant’s prior conviction on narcotics charges, his subsequent failure to attempt to rehabilitate himself or obtain employment, the fact that he had not acted under duress or as a result of mental or emotional pressure, his apparent willingness to kill, his lack of remorse after the killing, and the conclusion of a psychiatrist that he had a sociopathic personality and that his patterns of conduct would be the same in the future as they had been in the past.

Thus, Texas law essentially requires that one of five aggravating circumstances be found before a defendant can be found guilty of capital murder, and that in considering whether to impose a death sentence the jury may be asked to consider whatever evidence of mitigating circumstances the defense can bring before it. It thus appears that, as in Georgia and Florida, the Texas *274capital-sentencing procedure guides and focuses the jury’s objective consideration of the particularized circumstances of the individual offense and the individual offender before it can impose a sentence of death.

B

As in the Georgia and Florida cases, however, the petitioner contends that the substantial legislative changes that Texas made in response to this Court’s Furman decision are no more than cosmetic in nature and have in fact not eliminated the arbitrariness and caprice of the system -held in Furman to violate the Eighth and Fourteenth Amendments.8

The petitioner first asserts that arbitrariness still pervades the entire criminal justice system of Texas — from the prosecutor's decision whether to charge a capital offense in the first place and then whether to engage in plea bargaining, through the jury's consideration of lesser included offenses, to the Governor’s ultimate power to commute death sentences. This contention fundamentally misinterprets the Furman decision, and we reject it for the reasons set out in our opinion today in Gregg v. Georgia, ante, at 199.

(2)

Focusing on the second statutory question that Texas requires a jury to answer in considering whether to impose a death sentence, the petitioner argues that it is impossible to predict future behavior and that the question is so vague as to be meaningless. It is, of course, not easy to predict future behavior. The fact that such a determination is difficult, however, does not *275mean that it cannot be made. Indeed, prediction of future criminal conduct is an essential element in many of the decisions rendered throughout our criminal justice system. The decision whether to admit a defendant to bail, for instance, must often turn on a judge’s prediction of the defendant’s future conduct.9 And any sentencing authority must predict a convicted person’s probable future conduct when it engages in the process of determining what punishment to impose.10 For those sentenced to prison, these same predictions must be made by parole authorities.11 The task that a Texas jury *276must perform in answering the statutory question in issue is thus basically no different from the task performed countless times each day throughout the American system of criminal justice. What is essential is that the jury have before it all possible relevant information about the individual defendant whose fate it must determine. Texas law clearly assures that all such evidence will be adduced.

IV

We conclude that Texas’ capital-sentencing procedures, like those of Georgia and Florida, do not violate the Eighth and Fourteenth Amendments. By narrowing its definition of capital murder, Texas has essentially said that there must be at least one statutory aggravating circumstance in a first-degree murder case before a death sentence may even be considered. By authorizing the defense to bring before the jury at the separate sentencing hearing whatever mitigating circumstances relating to the individual defendant can be adduced, Texas has ensured that the sentencing jury will have adequate guidance to enable it to perform its sentencing function. By providing prompt judicial review of the jury’s decision in a court with statewide jurisdiction, Texas has provided a means to promote the evenhanded, rational, and consistent imposition of death sentences under law. Because this system serves to assure that sentences of death will not be “wantonly” or “freakishly” imposed, it does not violate the Constitution. Furman v. Georgia, 408 U. S., at 310 (Stewaut, J., concurring). *277Accordingly, the judgment of the Texas Court of Criminal Appeals is affirmed.

It is so ordered.

[For dissenting opinion of Mr. Justice Brennan, see ante, p. 227.]

[For dissenting opinion of Mr. Justice Marshall, see ante, p. 231,]

Mr. Chief Justice Burger,

concurring in judgment.

I concur in the judgment. See Furman v. Georgia, 408 U. S. 238, 375 (1972) (Burger, C. J., dissenting).

Mr. Justice White,

with whom The Chief Justice and Mr. Justice Rehnquist join,

concurring in the judgment.

Following the invalidation of the Texas capital punishment statute in Branch v. Texas, decided with Furman v. Georgia, 408 U. S. 238 (1972), the Texas Legislature re-enacted the death penalty for five types of murder, including murders committed in the course of certain felonies and required that it be imposed providing that, after returning a guilty verdict in such murder cases and after a sentencing proceeding at which all relevant evidence is admissible, the jury answers two questions in the affirmative — and a third if raised by the evidence:

“(1) whether the conduct of the defendant that caused the death of the deceased was committed deliberately and with the reasonable expectation that the death of the deceased or another would result;

“(2) whether there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society; and

“(3) if raised by the evidence, whether the conduct of the defendant in killing the deceased was unrea*278sonable in response to the provocation, if any, by the deceased.” Tex. Code Crim. Proc., Art. 37.071 (b) (Supp. 1975-1976).

The question in this case is whether the death penalty imposed on Jerry Lane Jurek for the crime of felony murder may be carried out consistently with the Eighth and Fourteenth Amendments.

The opinion of Mr. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens describes, and I shall not repeat, the facts of the crime and proceedings leading to the imposition of the death penalty when the jury unanimously gave its affirmative answers to the relevant questions posed in the judge’s post-verdict instructions. I also agree with that opinion that the judgment of the Texas Court of Criminal Appeals, which affirmed the conviction and judgment, must be affirmed here. 522 S. W. 2d 934 (1975).

For the reasons stated in my dissent in Roberts v. Louisiana, post, at 350-356, I cannot conclude that the Eighth Amendment forbids the death penalty under any and all circumstances. I also cannot agree with petitioner’s other major contention that under the new Texas statute and the State’s criminal justice system in general, the criminal jury and other law enforcement officers exercise such a range of discretion that the death penalty will be imposed so seldom, so arbitrarily, and so freakishly that the new statute suffers from the infirmities which Branch v. Texas found in its predecessor. Under the revised law, the substantive crime of murder is defined; and when a murder occurs in one of the five circumstances set out in the statute, the death penalty must be imposed if the jury also makes the certain additional findings against the defendant. Petitioner claims that the additional questions upon which the death sentence depends are so vague that in essence the *279jury possesses standardless sentencing power; but I agree with Justices Stewart, Powell, and Stevens that the issues posed in the sentencing proceeding have a common-sense core of meaning and that criminal juries should be capable of understanding them. The statute does not extend to juries discretionary power to dispense mercy, and it should not be assumed that juries will disobey or nullify their instructions. As of February of this year, 33 persons, including petitioner, had been sentenced to death under the Texas murder statute. I cannot conclude at this juncture that the death penalty under this system will be imposed so seldom and arbitrarily as to serve no useful penological function and hence fall within reach of the decision announced by five Members of the Court in Furman v. Georgia.

Nor, for the reasons I have set out in Roberts, post, at 348-350, and Gregg, ante, at 224-225, am I convinced that this conclusion should be modified because of the alleged discretion which is exercisable by other major functionaries in the State’s criminal justice system. Furthermore, as Justices Stewart, Powell, and Stevens state and as the Texas Court of Criminal Appeals has noted, the Texas capital punishment statute limits the imposition of the death penalty to a narrowly defined group of the most brutal crimes and aims at limiting its imposition to similar offenses occurring under similar circumstances. 522 S. W. 2d, at 939.

I concur in the judgment of affirmance.

Mr. Justice Blackmun,

concurring in the judgment.

I concur in the judgment. See Furman v. Georgia, 408 U. S. 238, 405-414 (1972) (Blackmun, J., dissenting), and id., at 375, 414, and 465.

Judgment of the Court, and opinion of

Mr. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens, announced by Mr. Justice Stewart.

The question in this case is whether the imposition of a death sentence for the crime of first-degree murder under the law of North Carolina violates the Eighth and Fourteenth Amendments.

I

The petitioners were convicted of first-degree murder as the result of their participation in an armed robbery *283of a convenience food store, in the course of which the cashier was killed and a customer was seriously wounded. There were four participants in the robbery: the petitioners James Tyrone Woodson and Luby Waxton and two others, Leonard Tucker and Johnnie Lee Carroll. At the petitioners’ trial Tucker and Carroll testified for the prosecution after having been permitted to plead guilty to lesser offenses; the petitioners testified in their own defense.

The evidence for the prosecution established that the four men had been discussing a possible robbery for some time. On the fatal day Woodson had been drinking heavily. About 9:30 p. m., Waxton and Tucker came to the trailer where Woodson was staying. When Woodson came out of the trailer, Waxton struck him in the face and threatened to kill him in an effort to make him sober up and come along on the robbery. The three proceeded to Waxton’s trailer where they met Carroll. Waxton armed himself with a nickel-plated derringer, and Tucker handed Woodson a rifle. The four then set out by automobile to rob the store. Upon arriving at their destination Tucker and Waxton went into the store while Carroll and Woodson remained in the car as lookouts. Once inside the store, Tucker purchased a package of cigarettes from the woman cashier. Waxton then also asked for a package of cigarettes, but as the cashier approached him he pulled the derringer out of his hip pocket and fatally shot her at point-blank range. Waxton then took the money tray from the cash register and gave it to Tucker, who carried it out of the store, pushing past an entering customer as he reached the door. After he was outside, Tucker heard a second shot from inside the store, and shortly thereafter Waxton emerged, carrying a handful of paper money. Tucker and Wax-ton got in the car and the four drove away.

*284The petitioners’ testimony agreed in large part with this version of the circumstances of the robbery. It differed diametrically in one important respect: Waxton claimed that he never had a gun, and that Tucker had shot both the cashier and the customer.

During the trial Waxton asked to be allowed to plead guilty to the same lesser offenses to which Tucker had pleaded guilty,1 but the solicitor refused to accept the pleas.2 Woodson, by contrast, maintained throughout the trial that he had been coerced by Waxton, that he waS Therefore innocent, and that he would not consider pleading guilty to any offense.

The petitioners were found guilty on all charges,3 and, as was required by statute, sentenced to death. The Supreme Court of North Carolina affirmed. 287 N. C. 578, 215 S. E. 2d 607 (1975). We granted certiorari, 423 U. S. 1082 (1976), to consider whether the imposition of the death penalties in this case comports 'with *285the Eighth and Fourteenth Amendments to the United States Constitution.

II

The petitioners argue that the imposition of the death penalty under any circumstances is cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments. We reject this argument for the reasons stated today in Gregg v. Georgia, ante, at 168-187.

III

At the time of this Court’s decision in Furman v. Georgia, 408 U. S. 238 (1972), North Carolina law provided that in cases of first-degree murder, the jury in its unbridled discretion could choose whether the convicted defendant should be sentenced to death or to life imprisonment.4 After the Furman decision the Supreme Court of North Carolina in State v. Waddell, 282 N. C. 431; 194 S. E. 2d 19 (1973), held unconstitutional the provision of the death penalty statute that gave the jury the option of returning a verdict of guilty without cap*286ital punishment, but held further that this provision was severable so that the statute survived as a mandatory death penalty law.5

The North Carolina General Assembly in 1974 followed the court’s lead and enacted a new statute that was essentially unchanged from the old one except that it made the death penalty mandatory. The statute now reads as follows:

“Murder in the first and second degree defined; punishment. — A murder which shall be perpetrated by means of poison, lying in wait, imprisonment, starving, torture, or by any other kind of willful, deliberate and premeditated killing, or which shall be committed in the perpetration or attempt to perpetrate any arson, rape, robbery, kidnapping, burglary or other felony, shall be deemed to be murder in the first degree and shall be punished with death. All other kinds of murder shall be deemed murder in the second degree, and shall be punished by imprisonment for a term of not less than two years nor more than life imprisonment in the State’s prison.” N. C. Gen. Stat. §14-17 (Cum. Supp. 1975).

It was under this statute that the petitioners, who committed their crime on June 3, 1974, were tried, convicted, and sentenced to death.

North Carolina, unlike Florida, Georgia, and Texas, has thus responded to the Furman decision by making death the mandatory sentence for all persons convicted *287of first-degree murder.6 In ruling on the constitutionality of the sentences imposed on the petitioners under this North Carolina statute, the Court now addresses for the first time the question whether a death sentence returned pursuant to a law imposing a mandatory death penalty for a broad category of homicidal offenses7 constitutes cruel and unusual punishment within the meaning of the Eighth and Fourteenth Amendments.8 The issue, like that explored in Furman, involves the procedure employed by the State to select persons for the unique and irreversible penalty of death.9

*288A

The Eighth Amendment stands to assure that the State’s power to punish is “exercised within the limits of civilized standards.” Trop v. Dulles, 356 U. S. 86, 100 (1958) (plurality opinion). See id., at 101; Weems v. United States, 217 U. S. 349, 373, 378 (1910); Louisiana ex rel. Francis v. Resweber, 329 U. S. 459, 468-469 (1947) (Frankfurter, J., concurring); 10 Robinson v. California, 370 U. S. 660, 666 (1962); Furman v. Georgia, 408 U. S., at 242 (Douglas, J., concurring); id., at 269-270 (Brennan, J., concurring); id., at 329 (Marshall, J., concurring) ; id., at 382-383 (Burger, C. J., dissenting) ; id., at 409 (Blackmun, J., dissenting); id., at 428-429 (Powell, J., dissenting). Central to the application of the Amendment is a determination of contemporary standards regarding the infliction of punishment. As discussed in Gregg v. Georgia, ante, at 176-182, indicia of societal values identified in prior opinions include history and traditional usage,11 legislative enactments,12 and jury determinations.13

*289In order to provide a frame for assessing the relevancy of these factors in this case we begin by sketching the history of mandatory death penalty statutes in the United States. At the time the Eighth Amendment was adopted in 1791, the States uniformly followed the common-law practice of making death the exclusive and mandatory sentence for certain specified offenses.14 Although the range of capital offenses in the American Colonies was quite limited in comparison to the more than 200 offenses then punishable by death in England,15 the Colonies at the time of the Revolution imposed death sentences on all persons convicted of any of a considerable number of crimes, typically including at a minimum, murder, treason, piracy, arson, rape, robbery, burglary, and sodomy.16 As at common law, all homicides that were not involuntary, provoked, justified, or excused constituted murder and were automatically punished by death.17 Almost from the outset jurors reacted unfavorably to the harshness of mandatory death sentences.18 The States initially responded to this ex*290pression of public dissatisfaction with mandatory statutes by limiting the classes of capital offenses.19

This reform, however, left unresolved the problem posed by the not infrequent refusal of juries to convict murderers rather than subject them to automatic death sentences. In 1794, Pennsylvania attempted to alleviate the undue severity of the law by confining the mandatory death penalty to “murder of the first degree” encompassing all “wilful, deliberate and premeditated” killings. Pa. Laws 1794 c. 1766.20 Other jurisdictions, including Virginia and Ohio, soon enacted similar measures, and within a generation the practice spread to most of the States.21

Despite the broad acceptance of the division of murder into degrees, the reform proved to be an unsatisfactory means of identifying persons appropriately punishable by death. Although its failure was due in part to the amorphous nature of the controlling concepts of will*291fulness, deliberateness, and premeditation,22 a more fundamental weakness of the reform soon became apparent. Juries continued to find the death penalty inappropriate in a significant number of first-degree murder cases and refused to return guilty verdicts for that crime.23

The inadequacy of distinguishing between murderers solely on the basis of legislative criteria narrowing the definition of the capital offense led the States to grant juries sentencing discretion in capital cases. Tennessee in 1838, followed by Alabama in 1841, and Louisiana in 1846, were the first States to abandon mandatory death sentences in favor of discretionary death penalty statutes.24 This flexibility remedied the harshness of mandatory statutes by permitting the jury to respond to mitigating factors by withholding the death penalty. By the turn of the century, 23 States and the Federal Government had made death sentences discretionary for first-degree murder and other capital offenses. During the next two decades 14 additional States replaced their mandatory death penalty statutes. Thus, by the end of World War I, all but eight States, Hawaii, and the District of Columbia either had adopted discretionary death penalty schemes or abolished the death penalty altogether. By 1963, all of these remaining jurisdic*292tions had replaced their automatic death penalty statutes with discretionary jury sentencing.25

The history of mandatory death penalty statutes in *293the United States thus reveals that the practice of sentencing to death all persons convicted of a particular offense has been rejected as unduly harsh and unwork-ably rigid. The two crucial indicators of evolving standards of decency respecting the imposition of punishment in our society — jury determinations and legislative enactments — both point conclusively to the repudiation of automatic death sentences. At least since the Revolution, American jurors have, with some regularity, disregarded their oaths and refused to convict defendants where a death sentence was the automatic consequence of a guilty verdict. As we have seen, the initial movement to reduce the number of capital offenses and to separate murder into degrees was prompted in part by the reaction of jurors as well as by reformers who objected to the imposition of death as the penalty for any crime. Nineteenth century journalists, statesmen, and jurists repeatedly observed that jurors were often deterred from convicting palpably guilty men of first-degree murder under mandatory statutes.26 Thereafter, continuing evidence of jury reluctance to convict persons of capital offenses in mandatory death penalty jurisdictions resulted in legislative authorization of discretionary jury sentencing — by Congress for federal crimes in 1897,27 by North Carolina in 1949,28 and by Congress for the District of Columbia in 1962.29

*294As we have noted today in Gregg v. Georgia, ante, at 179-181, legislative measures adopted by the people’s chosen representatives weigh heavily in ascertaining con*295temporary standards of decency. The consistent course charted by the state legislatures and by Congress since the middle of the past century demonstrates that the aversion of jurors to mandatory death penalty statutes is shared by society at large.30

Still further evidence of the incompatibility of mandatory death penalties with contemporary values is provided by the results of jury sentencing under discretionary statutes. In Witherspoon v. Illinois, 391 U. S. 510 (1968), the Court observed that “one of the most important functions any jury can perform” in exercising its discretion to choose “between life imprisonment and capital punishment” is “to maintain a link between contemporary community values and the penal system.” Id., at 519, and n. 15. Various studies indicate that even in first-degree murder cases juries with sentencing discretion do not impose the death penalty “with any great frequency.” H. Kalven & H. Zeisel, The American Jury 436 (1966).31 The actions of sentencing juries sug*296gest that under contemporary standards of decency death is viewed as an inappropriate punishment for a substantial portion of convicted first-degree murderers.

Although the Court has never ruled on the constitutionality of mandatory death penalty statutes, on several occasions dating back to 1899 it has commented upon our society’s aversion to automatic death sentences. In Winston v. United States, 172 U. S. 303 (1899), the Court noted that the “hardship of punishing with death every crime coming within the definition of murder at common law, and the reluctance of jurors to concur in a capital conviction, have induced American legislatures, in modern times, to allow some cases of murder to be punished by imprisonment, instead of by death.” Id., at 310.32 Fifty years after Winston, the Court underscored the marked transformation in our attitudes toward mandatory sentences: “The belief no longer prevails that every offense in a like legal category calls for an identical *297punishment without regard to the past life and habits of a particular offender. This whole country has traveled far from the period in which the death sentence was an automatic and commonplace result of convictions . . . .” Williams v. New York, 337 U. S. 241, 247 (1949).

More recently, the Court in McGautha v. California, 402 U. S. 183 (1971), detailed the evolution of discretionary imposition of death sentences in this country, prompted by what it termed the American “rebellion against the common-law rule imposing a mandatory death sentence on all convicted murderers.” Id., at 198. See id., at 198-202. Perhaps the one important factor about evolving social values regarding capital punishment upon which the Members of the Furman Court agreed was the accuracy of McGautha’s assessment of our Nation’s rejection of mandatory death sentences. See Furman v. Georgia, 408 U. S., at 245-246 (Douglas, J., concurring); id., at 297-298 (Brennan, J., concurring); id., at 339 (Marshall, J., concurring); id., at 402-403 (Burger, C. J., with whom Blackmun, Powell, and Rehnquist, JJ., joined, dissenting); id., at 413 (Blackmun, J., dissenting). Mr. 'Justice Blackmun, for example, emphasized that legislation requiring an automatic death sentence for specified crimes would be “regressive and of an antique mold” and would mark a return to a “point in our criminology [passed beyond] long ago.” Ibid. The Chief Justice, speaking for the four dissenting Justices in Furman, discussed the question of mandatory death sentences at some length:

“I had thought that nothing was clearer in history, as we noted in McGautha one year ago, than the American abhorrence of 'the common-law rule imposing a mandatory death sentence on all convicted murderers.’ 402 U. S., at 198. As the concurring opinion of Mr. Justice Marshall shows, [408 *298U. S.,] at 339, the 19th century movement away from mandatory death sentences marked an enlightened introduction of flexibility into the sentencing process. It recognized that individual culpability is not always measured by the category of the crime committed. This change in sentencing practice was greeted by the Court as a humanizing development. See Winston, v. United States, 172 U. S. 303 (1899); cf. Calton v. Utah, 130 U. S. 83 (1889). See also Andres v. United States, 333 U. S. 740, 753 (1948) (Frankfurter, J., concurring).” Id., at 402.

Although it seems beyond dispute that, at the time of the Furman decision in 1972, mandatory death penalty statutes had been renounced by American juries and legislatures, there remains the question whether the mandatory statutes adopted by North Carolina and a number of other States following Furman evince a sudden reversal of societal values regarding the imposition of capital punishment. In view of the persistent and unswerving legislative rejection of mandatory death penalty statutes beginning in 1838 and continuing for more than 130 years until Furman,33 it seems evident that the post-Furman enactments reflect attempts by the States to retain the death penalty in a form consistent with the Constitution, rather than a renewed societal acceptance of mandatory death sentencing.34 The fact that some *299States have adopted mandatory measures following Fur-man while others have legislated standards to guide jury discretion appears attributable to diverse readings of this Court’s multi-opinioned decision in that case.35

A brief examination of the background of the current North Carolina statute serves to reaffirm our assessment of its limited utility as an indicator of contemporary values regarding mandatory death sentences. Before 1949, North Carolina imposed a mandatory death sentence on any person convicted of rape or first-degree murder. That year, a study commission created by the state legislature recommended that juries be granted discretion to recommend life sentences in all capital cases:

“We propose that a recommendation of mercy by the jury in capital cases automatically carry with it a life sentence. Only three other states now have the mandatory death penalty and we believe its retention will be definitely harmful. Quite frequently, juries refuse to convict for rape or first degree murder because, from all the circumstances, they do not believe the defendant, although guilty, should suffer death. The result is that verdicts are returned hardly in harmony with evidence. Our *300proposal is already in effect in respect to the crimes of burglary and arson. There is much testimony that it has proved beneficial in such cases. We think the law can now be broadened to include all capital crimes.” Report of the Special Commission For the Improvement of the Administration of Justice, North Carolina, Popular Government 13 (Jan. 1949).

The 1949 session of the General Assembly of North Carolina adopted the proposed modifications of its rape and murder statutes. Although in subsequent years numerous bills were introduced in the legislature to limit further or abolish the death penalty in North Carolina, they were rejected as were two 1969 proposals to return to mandatory death sentences for all capital offenses. See State v. Waddell, 282 N. C., at 441, 194 S. E. 2d, at 26 (opinion of the court); id., at 456-457, 194 S. E. 2d, at 32-33 (Bobbitt, C. J., concurring in part and dissenting in part).

As noted, supra, at 285-286, when the Supreme Court of North Carolina analyzed the constitutionality of the State's death penalty statute following this Court’s decision in Furman, it severed the 1949 proviso authorizing jury sentencing discretion and held that “the remainder of the statute with death as the mandatory punishment . . . remains in full force and effect.” State v. Waddell, supra, at 444-445, 194 S. E. 2d, at 28. The North Carolina General Assembly then followed the course found constitutional in Waddell and enacted a first-degree murder provision identical to the mandatory statute in operation prior to the authorization of jury discretion. The State’s brief in this case relates that the legislature sought to remove “all sentencing discretion [so that] there could be no successful Furman based attack on the North Carolina statute.”

*301It is now well established that the Eighth Amendment draws much of its meaning from “the evolving standards of decency that mark the progress of a maturing society.” Trop v. Dulles, 356 U. S., at 101 (plurality opinion). As the above discussion makes clear, one of the most significant developments in our society’s treatment of capital punishment has been the rejection of the common-law practice of inexorably imposing a death sentence upon every person convicted of a specified offense. North Carolina’s mandatory death penalty statute for first-degree murder departs markedly from contemporary standards respecting the imposition of the punishment of death and thus cannot be applied consistently with the Eighth and Fourteenth Amendments’ requirement that the State’s power to punish “be exercised within the limits of civilized standards.” Id., at 100.36

*302B

A separate deficiency of North Carolina's mandatory death sentence statute is its failure to provide a constitutionally tolerable response to Furman’s, rejection of unbridled jury discretion in the imposition of capital sentences. Central to the limited holding in Furman was the conviction that the vesting of standardless sentencing power in the jury violated the Eighth and Fourteenth Amendments. See Furman v. Georgia, 408 U. S., at 309-310 (Stewart, J., concurring); id., at 313 (White, J., concurring); cf. id., at 253-257 (Douglas, J., concurring). See also id., at 398-399 (Burger, C. J., dissenting). It is argued that North Carolina has remedied the inadequacies of the death penalty statutes held unconstitutional in Furman by withdrawing all sentencing discretion from juries in capital cases. But when one considers the long and consistent American experience with the death penalty in first-degree murder cases, it becomes evident that mandatory statutes enacted in response to Furman have simply papered over the problem of unguided and unchecked jury discretion.

As we have noted in Part III-A, supra, there is general agreement that American juries have persistently refused to convict a significant portion of persons charged with first-degree murder of that offense under mandatory death penalty statutes. The North Carolina study commission, supra, at 299-300, reported that juries in that State “[q]uite frequently” were deterred from rendering guilty verdicts of first-degree murder because of the enormity of the sentence automatically imposed. Moreover, *303as a matter of historic fact, juries operating under discretionary sentencing statutes have consistently returned death sentences in only a minority of first-degree murder cases.37 In view of the historic record, it is only reasonable to assume that many juries under mandatory statutes will continue to consider the grave consequences of a conviction in reaching a verdict. North Carolina’s mandatory death penalty statute provides no standards to guide the jury in its inevitable exercise of the power to determine which first-degree murderers shall live and which shall die. And there is no way under the North Carolina law for the judiciary to check arbitrary and capricious exercise of that power through a review of death sentences.38 Instead of rationalizing the sentencing process, a mandatory scheme may well exacerbate the problem identified in Furman by resting the penalty determination on the particular jury’s willingness to act lawlessly. While a mandatory death penalty statute may reasonably be expected to increase the number of persons sentenced to death, it does not fulfill Furman’s, basic requirement by replacing arbitrary and wanton jury discretion with objective standards to guide, regularize, and make rationally reviewable the process for imposing a sentence of death.

C

A third constitutional shortcoming of the North Carolina statute is its failure to allow the particularized consideration of relevant aspects of the character and record of each convicted defendant before the imposition upon him of a sentence of death. In Furman, members of the Court acknowledged what cannot fairly be denied — that death is a punishment different from all other *304sanctions in kind rather than degree. See 408 U. S., at 286-291 (Brennan, J., concurring); id., at 306 (Stewart, J., concurring). A process that accords no significance to relevant facets of the character and record of the individual offender or the circumstances of the particular offense excludes from consideration in fixing the ultimate punishment of death the possibility of compassionate or mitigating factors stemming from the diverse frailties of humankind. It treats all persons convicted of a designated offense not as uniquely individual human beings, but as members of a faceless, undifferentiated mass to be subjected to the blind infliction of the penalty of death.

This Court has previously recognized that “[f]or the determination of sentences, justice generally requires consideration of more than the particular acts by which the crime was committed and that there be taken into account the circumstances of the offense together with the character and propensities of the offender.” Pennsylvania ex rel. Sullivan v. Ashe, 302 U. S. 51, 55 (1937). Consideration of both the offender and the offense in order to arrive at a just and appropriate sentence has been viewed as a progressive and humanizing development. See Williams v. New York, 337 U. S., at 247-249; Furman v. Georgia, 408 U. S., at 402-403 (Burger, C. J., dissenting). While the prevailing practice of individualizing sentencing determinations generally reflects simply enlightened policy rather than a constitutional imperative, we believe that in capital cases the fundamental respect for humanity underlying the Eighth Amendment, see Trop v. Dulles, 356 U. S., at 100 (plurality opinion), requires consideration of the character and record of the individual offender and the circumstances of the particular offense as a constitutionally indispensable part of the process of inflicting the penalty of death.

*305This conclusion rests squarely on the predicate that the penalty of death is qualitatively different from a sentence of imprisonment, however long. Death, in its finality, differs more from life imprisonment than a 100-year prison term differs from one of only a year or two. Because of that qualitative difference, there is a corresponding difference in the need for reliability in the determination that death is the appropriate punishment in a specific case.39

For the reasons stated, we conclude that the death sentences imposed upon the petitioners under North Carolina's mandatory death sentence statute violated the Eighth and Fourteenth Amendments and therefore must be set aside.40 The judgment of the Supreme Court of North Carolina is reversed insofar as it upheld the death sentences imposed upon the petitioners, and the case is remanded for further proceedings not inconsistent with this opinion.

It is so ordered.

Mr. Justice Brennan,

concurring in the judgment.

For the reasons stated in my dissenting opinion in Gregg v. Georgia, ante, p. 227, I concur in the judgment *306that sets aside the death sentences imposed under the North Carolina death sentence statute as violative of the Eighth and Fourteenth Amendments.

Mr. Justice Marshall,

concurring in the judgment.

For the reasons stated in my dissenting opinion in Gregg v. Georgia, ante, p. 231, I am of the view that the death penalty is a cruel and unusual punishment forbidden by the Eighth and Fourteenth Amendments. I therefore concur in the Court’s judgment.

Mr. Justice White,

with whom The Chief Justice and Mr. Justice Rehnquist join,

dissenting.

Following Furman v. Georgia, 408 U. S. 238 (1972), the North Carolina Supreme Court considered the effect of that case on the North Carolina criminal statutes which imposed the death penalty for first-degree murder and other crimes but which provided that “if at the time of rendering its verdict in open court, the jury shall so recommend, the punishment shall be imprisonment for life in the State’s prison, and the court shall so instruct the jury.” State v. Waddell, 282 N. C. 431, 194 S. E. 2d 19 (1973), determined that Furman v. Georgia invalidated only the proviso giving the jury the power to limit the penalty to life imprisonment and that thenceforward death was the mandatory penalty for the specified capital crimes. Thereafter N. C. Gen. Stat. § 14-17 was amended to eliminate the express dispensing power of the jury and to add kidnaping to the underlying felonies for which death is the specified penalty. As amended in 1974, the section reads as follows:

“A murder which shall be perpetrated by means of poison, lying in wait, imprisonment, starving, torture, or by any other kind of willful, deliberate and premeditated killing, or which shall be committed *307in the perpetration or attempt to perpetrate any arson, rape, robbery, kidnapping, burglary or other felony, shall be deemed to be murder in the first degree and shall be punished with death. All other kinds of murder shall be deemed murder in the second degree, and shall be punished by imprisonment for a term of not less than two years nor more than life imprisonment in the State’s prison.”

It was under this statute that the petitioners in this case were convicted of first-degree murder and the mandatory death sentences imposed.

The facts of record and the proceedings in this case leading to petitioners’ convictions for first-degree murder and their death sentences appear in the opinion of MR. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens. The issues in the case are very similar, if not identical, to those in Roberts v. Louisiana, post, p. 325. For the reasons stated in my dissenting opinion in that case, I reject petitioners’ arguments that the death penalty in any circumstances is a violation of the Eighth Amendment and that the North Carolina statute, although making the imposition of the death penalty mandatory upon proof of guilt and a verdict of first-degree murder, will nevertheless result in the death penalty being imposed so seldom and arbitrarily that it is void under Furman v. Georgia. As is also apparent from my dissenting opinion in Roberts v. Louisiana, I also disagree with the two additional grounds which the plurality sua sponte offers for invalidating the North Carolina statute. I would affirm the judgment of the North Carolina Supreme Court.

Mr. Justice Blackmun,

dissenting.

I dissent for the reasons set forth in my dissent in Furman v. Georgia, 408 U. S. 238, 405-414 (1972), and *308in the other dissenting opinions I joined in that case. Id., at 375, 414, and 465.

MR. Justice Rehnquist,

dissenting.

I

The difficulties which attend the plurality’s explanation for the result it reaches tend at first to obscure difficulties at least as significant which inhere in the unarticulated premises necessarily underlying that explanation. I advert to the latter only briefly, in order to devote the major and following portion of this dissent to those issues which the plurality actually considers.

As an original proposition, it is by no means clear that the prohibition against cruel and unusual punishments embodied in the Eighth Amendment, and made applicable to the States by the Fourteenth Amendment, Robinson v. California, 370 U. S. 660 (1962), was not limited to those punishments deemed cruel and unusual at the time of the adoption of the Bill of Rights. McGautha v. California, 402 U. S. 183, 225 (1971) (opinion of Black, J.). If Weems v. United States, 217 U. S. 349 (1910), dealing not with the Eighth Amendment but with an identical provision contained in the Philippine Constitution, and the plurality opinion in Trop v. Dulles, 356 U. S. 86 (1958), are to be taken as indicating the contrary, they should surely be weighed against statements in cases such as Wilkerson v. Utah, 99 U. S. 130 (1879); In re Kemmler, 136 U. S. 436 (1890); Louisiana ex rel. Francis v. Resweber, 329 U. S. 459, 464 (1947), and the plurality opinion in Trop itself, that the infliction of capital punishment is not in itself violative of the Cruel and Unusual Punishments Clause. Thus for the plurality to begin its analysis with the assumption that it need only demonstrate that “evolving standards of decency” show that contemporary “so-*309eiety” has rejected such provisions is itself a somewhat shaky point of departure. But even if the assumption be conceded, the plurality opinion’s analysis nonetheless founders.

The plurality relies first upon its conclusion that society has turned away from the mandatory imposition of death sentences, and second upon its conclusion that the North Carolina system has “simply papered over” the problem of unbridled jury discretion which two of the separate opinions in Furman v. Georgia, 408 U. S. 238 (1972), identified as the basis for the judgment rendering the death sentences there reviewed unconstitutional. The third “constitutional shortcoming” of the North Carolina statute is said to be “its failure to allow the particularized consideration of relevant aspects of the character and record of each convicted defendant before the imposition upon him of a sentence of death.” Ante, at 303.

I do not believe that any one of these reasons singly, or all of them together, can withstand careful analysis. Contrary to the plurality’s assertions, they would import into the Cruel and Unusual Punishments Clause procedural requirements which find no support in our cases. Their application will result in the invalidation of a death sentence imposed upon a defendant convicted of first-degree murder under the North Carolina system, and the upholding of the same sentence imposed on an identical defendant convicted on identical evidence of first-degree murder under the Florida, Georgia, or Texas systems — a result surely as “freakish” as that condemned in the separate opinions in Furman.

II

The plurality is simply mistaken in its assertion that “[t]he history of mandatory death penalty statutes in the United States thus reveals that the practice of sen*310tencing to death all persons convicted of a particular offense has been rejected as unduly harsh and unwork-ably rigid.” Ante, at 292-293. This conclusion is purportedly based on two historic developments: the first a series of legislative decisions during the 19th century narrowing the class of offenses punishable by death; the second a series of legislative decisions during both the 19th and 20th centuries, through which mandatory imposition of the death penalty largely gave way to jury discretion in deciding whether or not to impose this ultimate sanction. The first development may have some relevance to the plurality’s argument in general but has no bearing at all upon this case. The second development, properly analyzed, has virtually no relevance even to the plurality’s argument.

There can be no question that the legislative and other materials discussed in the plurality’s opinion show a widespread conclusion on the part of state legislatures during the 19th century that the penalty of death was being required for too broad a range of crimes, and that these legislatures proceeded to narrow the range of crimes for which such penalty could be imposed. If this case involved the imposition of the death penalty for an offense such as burglary or sodomy, see ante, at 289, the virtually unanimous trend in the legislatures of the States to exclude such offenders from liability for capital punishment might bear on the plurality’s Eighth Amendment argument. But petitioners were convicted of first-degree murder, and there is not the slightest suggestion in the material relied upon by the plurality that there had been any turning away at all, much less any such unanimous turning away, from the death penalty as a punishment for those guilty of first-degree murder. The legislative narrowing of the spectrum of capital crimes, therefore, while very arguably representing a general societal judgment since the trend was so widespread, simply never *311reached far enough to exclude the sort of aggravated homicide of which petitioners stand convicted.

The second string to the plurality’s analytical bow is that legislative change from mandatory to discretionary imposition of the death sentence likewise evidences societal rejection of mandatory death penalties. The plurality simply does not make out this part of its case, however, in large part because it treats as being of equal dignity with legislative judgments the judgments of particular juries and of individual jurors.

There was undoubted dissatisfaction, from more than one sector of 19th century society, with the operation of mandatory death sentences. One segment of that society was totally opposed to capital punishment, and was apparently willing to accept the substitution of discretionary imposition of that penalty for its mandatory imposition as a halfway house on the road to total abolition. Another segment was equally unhappy with the operation of the mandatory system, but for an entirely different reason. As the plurality recognizes, this second segment of society was unhappy with the operation of the mandatory system, not because of the death sentences imposed under it, but because people obviously guilty of criminal offenses were not being convicted under it. See ante, at 293. Change to a discretionary system was accepted by these persons not because they thought mandatory imposition of the death penalty was cruel and unusual, but because they thought that if jurors were permitted to return a sentence other than death upon the conviction of a capital crime, fewer guilty defendants would be acquitted. See McGautha, 402 U. S., at 199.

So far as the action of juries is concerned, the fact that in some cases juries operating under the mandatory system refused to convict obviously guilty defendants does not reflect any “turning away” from the death penalty, or the mandatory death penalty, supporting the *312proposition that it is “cruel and unusual.” Given the requirement of unanimity with respect to jury verdicts in capital cases, a requirement which prevails today in States which accept a nonunanimous verdict in the case of other crimes, see Johnson v. Louisiana, 406 U. S. 356, 363-364 (1972), it is apparent that a single juror could prevent a jury from returning a verdict of conviction. Occasional refusals to convict, therefore, may just as easily have represented the intransigence of only a small minority of 12 jurors as well as the unanimous judgment of all 12. The fact that the presence of such jurors could prevent conviction in a given case, even though the majority of society, speaking through legislatures, had decreed that it should be imposed, certainly does not indicate that society as a whole rejected mandatory punishment for such offenders; it does not even indicate that those few members of society who serve on juries, as a whole, had doné so.

The introduction of discretionary sentencing likewise creates no inference that contemporary society had rejected the mandatory system as unduly severe. Legislatures enacting discretionary sentencing statutes had no reason to think that there would not be roughly the same number of capital convictions under the new system as under the old. The same subjective juror responses which resulted in juror nullification under the old system were legitimized, but in the absence of those subjective responses to a particular set of facts, a capital sentence could as likely be anticipated under the discretionary system as under the mandatory. And at least some of those who would have been acquitted under the mandatory system would be subjected to at least some punishment under the discretionary system, rather than escaping altogether a penalty for the crime of which they were guilty. That society was unwilling to accept the *313paradox presented to it by the actions of some maverick juries or jurors — the acquittal of palpably guilty defendants — hardly reflects the sort of an “evolving standard of decency” to which the plurality professes obeisance.

Nor do the opinions in Furman which indicate a preference for discretionary sentencing in capital cases suggest in the slightest that a mandatory sentencing procedure would be cruel and unusual. The plurality concedes, as it must, that following Furman 10 States enacted laws providing for mandatory capital punishment. See State Capital Punishment Statutes Enacted Subsequent to Furman v. Georgia, Congressional Research Service Pamphlet 17-22 (June 19, 1974). These enactments the plurality seeks to explain as due to a wrongheaded reading of the holding in Furman. But this explanation simply does not wash. While those States may be presumed to have preferred their prior systems reposing sentencing discretion in juries or judges, they indisputably preferred mandatory capital punishment to no capital punishment at all. Their willingness to enact statutes providing that penalty is utterly inconsistent with the notion that they regarded mandatory capital sentencing as beyond “evolving standards of decency.” The plurality’s glib rejection of these legislative decisions as having little weight on the scale which it finds in the Eighth Amendment seems to me more an instance of its desire to save the people from themselves than a conscientious effort to ascertain the content of any “evolving standard of decency.”

HH I — I

The second constitutional flaw which the plurality finds in North Carolina's mandatory system is that it has simply “papered over” the problem of unchecked *314jury discretion. The plurality states, ante, at 302, that “there is general agreement that American juries have persistently refused to convict a significant portion of persons charged with first-degree murder of that offense under mandatory death penalty statutes.” The plurality also states, ante, at 303, that “as a matter of historic fact, juries operating under discretionary sentencing statutes have consistently returned death sentences in only a minority of first degree murder cases.” The basic factual assumption of the plurality seems to be that for any given number of first-degree murder defendants subject to capital punishment, there will be a certain number of jurors who will be unwilling to impose the death penalty even though they are entirely satisfied that the necessary elements of the substantive offense are made out.

In North Carolina jurors unwilling to impose the death penalty may simply hang a jury or they may so assert themselves that a verdict of not guilty is brought in; in Louisiana they will have a similar effect in causing some juries to bring in a verdict of guilty of a lesser included offense even though all the jurors are satisfied that the elements of the greater offense are made out. Such jurors, of course, are violating their oath, but such violation is not only consistent with the majority’s hypothesis; the majority’s hypothesis is bottomed on its occurrence.

For purposes of argument, I accept the plurality’s hypothesis; but it seems to me impossible to conclude from it that a mandatory death sentence statute such as North Carolina enacted is any less sound constitutionally than are the systems enacted by Georgia, Florida, and Texas which the Court upholds.

In Georgia juries are entitled to return a sentence of life, rather than death, for no reason whatever, simply *315based upon their own subjective notions of what is right and what is wrong. In Florida the judge and jury are required to weigh legislatively enacted aggravating factors against legislatively enacted mitigating factors, and then base their choice between life or death on an estimate of the result of that weighing. Substantial discretion exists here, too, though it is somewhat more canalized than it is in Georgia. Why these types of discretion are regarded by the plurality as constitutionally permissible, while that which may occur in the North Carolina system is not, is not readily apparent. The freakish and arbitrary nature of the death penalty described in the separate concurring opinions of Justices Stewart and White in Furman arose not from the perception that so many capital sentences were being imposed but from the perception that so jew were being imposed. To conclude that the North Carolina system is bad because juror nullification may permit jury discretion while concluding that the Georgia and Florida systems are sound because they require this same discretion, is, as the plurality opinion demonstrates, inexplicable.

The Texas system much more closely approximates the mandatory North Carolina system which is struck down today. The jury is required to answer three statutory questions. If the questions are unanimously answered in the affirmative, the death penalty must be imposed. It is extremely difficult to see how this system can be any less subject to the infirmities caused by juror nullification which the plurality concludes are fatal to North Carolina’s statute. Justices Stewart, Powell, and Stevens apparently think they can sidestep this inconsistency because of their belief that one of the three questions will permit consideration of mitigating factors justifying imposition of a life sentence. It is, however, as those Justices recognize, Jurek v. Texas, ante, at 272-*316273, far from clear that the statute is to be read in such a fashion. In any event, while the imposition of such unlimited consideration of mitigating factors may conform to the plurality’s novel constitutional doctrine that “[a] jury must be allowed to consider on the basis of all relevant evidence not only why a death sentence should be imposed, but also why it should not be imposed,” ante, at 271, the resulting system seems as likely as any to produce the unbridled discretion which was condemned by the separate opinions in Furman.

The plurality seems to believe, see ante, at 303, that provision for appellate review will afford a check upon the instances of juror arbitrariness in a discretionary system. But it is not at all apparent that appellate review of death sentences, through a process of comparing the facts of one case in which a death sentence was imposed with the facts of another in which such a sentence was imposed, will afford any meaningful protection against whatever arbitrariness results from jury discretion. All that such review of death sentences can provide is a comparison of fact situations which must in their nature be highly particularized if not unique, and the only relief which it can afford is to single out the occasional death sentence which in the view of the reviewing court does not conform to the standards established by the legislature.

It is established, of course, that there is no right to appellate review of a criminal sentence. McKane v. Durston, 153 U. S. 684 (1894). That question is not at issue here, since North Carolina, along with the other four States whose systems the petitioners are challenging in these cases, provides appellate review for a death sentence imposed in one of its trial courts.

By definition, of course, there can be no separate appellate review of the factual basis for the sentencing decision *317in a mandatory system. If it is once established in a fairly conducted trial that the defendant has in fact committed the crime in question, the only question as to the sentence which can be raised on appeal is whether a legislative determination that such a crime should be punished by death violates the Cruel and Unusual Punishments Clause of the Eighth Amendment. Here both petitioners were convicted of first-degree murder, and there is no serious question raised by the plurality that death is not a constitutionally permissible penalty for such a crime.

But the plurality sees another role for appellate review in its’description of the reasons why the Georgia, Texas, and Florida systems are upheld, and the North Carolina system struck down. And it is doubtless true that Georgia in particular has made a substantial effort to respond to the concerns expressed in Furman, not an easy task considering the glossolalial manner in which those concerns were expressed. The Georgia Supreme Court has indicated that the Georgia death penalty statute requires it to review death sentences imposed by juries on the basis of rough “proportionality.” It has announced that it will not sustain, at least at the present time, death penalties imposed for armed robbery because that penalty is so seldom imposed by juries for that offense. It has also indicated that it will not sustain death penalties imposed for rape in certain fact situations, because the death penalty has been so seldom imposed on facts similar to those situations.

But while the Georgia response may be an admirable one as a matter of policy, it has imperfections, if a failure to conform completely to the dictates of the separate opinions in Furman be deemed imperfections, which the opinion of Justices Stewart, Powell, and Stevens does not point out. Although there may be some disagree*318ment between that opinion, and the opinion of my Brother White in Gregg v. Georgia, which I have joined, as to whether the proportionality review conducted by the Supreme Court of Georgia is based solely upon capital sentences imposed, or upon all sentences imposed in cases where a capital sentence could have been imposed by law, I shall assume for the purposes of this discussion that the system contemplates the latter. But this is still far from a guarantee of any equality in sentencing, and is likewise no guarantee against juror nullification. Under the Georgia system, the jury is free to recommend life imprisonment, as opposed to death, for no stated reason whatever. The Georgia Supreme Court cannot know, therefore, when it is reviewing jury sentences for life in capital cases, whether the jurors found aggravating circumstances present, but nonetheless decided to recommend mercy, or instead found no aggravating circumstances at all and opted for mercy. So the “proportionality” type of review, while it would perhaps achieve its objective if there were no possible factual lacunae in the jury verdicts, will not achieve its objective because there are necessarily such lacunae.

Identical defects seem inherent in the systems of appellate review provided in Texas and Florida, for neither requires the sentencing authority which concludes that a death penalty is inappropriate to state what mitigating factors were found to be present or whether certain aggravating factors urged by the prosecutor were actually found to be lacking. Without such detailed factual findings Justices Stewart, Powell, and Stevens’ praise of appellate review' as a cure for the constitutional infirmities which they identify seems to me somewhat forced.

Appellate review affords no correction whatever with respect to those fortunate few who are the beneficiaries *319of random discretion exercised by juries, whether under an admittedly discretionary system or under a purportedly mandatory system. It may make corrections at one end of the spectrum, but cannot at the other. It is even less clear that any provision of the Constitution can be read to require such appellate review. If the States wish to undertake such an effort, they are undoubtedly free to do so, but surely it is not required by the United States Constitution.

The plurality’s insistence on “standards” to “guide the jury in its inevitable exercise of the power to determine which . . . murderers shall live and which shall die” is squarely contrary to the Court’s opinion in McGautha v. California, 402 U. S. 183 (1971), written by Mr. Justice Harlan and subscribed to by five other Members of the Court only five years ago. So is the plurality’s latter-day recognition, some four years after the decision of the case, that Furman requires “objective standards to guide, regularize, and make rationally reviewable the process for imposing a sentence of death.” Its abandonment of stare decisis in this repudiation of McGautha is a far lesser mistake than its substitution of a superficial and contrived constitutional doctrine for the genuine wisdom contained in McGautha. There the Court addressed the “standardless discretion” contention in this language:

“In our view, such force as this argument has derives largely from its generality. Those who have come to grips with the hard task of actually attempting to draft means for channeling capital sentencing discretion have confirmed the lesson taught by the history recounted above. To identify before the fact those characterstics of criminal homicides and their perpetrators which call for the death penalty, and to express these characteristics in language *320which can be fairly understood and applied by the sentencing authority, appear to be tasks which are beyond present human ability.

“Thus the British Home Office, which before the recent abolition of capital punishment in that country had the responsibility for selecting the cases from England and Wales which should receive the benefit of the Royal Prerogative of Mercy, observed:

“ 'The difficulty of defining by any statutory provision the types of murder which ought or ought not to be punished by death may be illustrated by reference to the many diverse considerations to which the Home Secretary has regard in deciding whether to recommend clemency. No simple formula can take account of the innumerable degrees of culpability, and no formula which fails to do so can claim to be just or satisfy public opinion.' 1-2 Royal Commission on Capital Punishment, Minutes of Evidence 13 (1949).” 402 U. S., at 20T-205.

“In light of history, experience, and the present limitations of human knowledge, we find it quite impossible to say that committing to the untrammeled discretion of the jury the power to pronounce life or death in capital cases is offensive to anything in the Constitution. The States are entitled to assume that jurors confronted with the truly awesome responsibility of decreeing death for a fellow human will act with due regard for the consequences of their decision and will consider a variety of factors, many of which will have been suggested by the evidence or by the arguments of defense counsel. For a court to attempt to catalog the appropriate factors in this elusive area could inhibit rather than expand the scope of consideration, for no list of cir-*321cumstanees would ever be really complete. The infinite variety of cases and facets to each case would make general standards either meaningless ‘boilerplate’ or a statement of the obvious that no jury would need.” Id., at 207-208 (citation omitted).

It is also worth noting that the plurality opinion repudiates not only the view expressed by the Court in McGautha, but also, as noted in McGautha, the view which had been adhered to by every other American jurisdiction which had considered the question. See id., at 196 n. 8.

IV

The plurality opinion’s insistence, in Part III-C, that if the death penalty is to be imposed there must be “particularized consideration of relevant aspects of the character and record of each convicted defendant” is buttressed by neither case authority nor reason. Its principal claim to distinction is that it contradicts important parts of Part III-A in the same opinion.

Part III-A, which describes what it conceives to have been society’s turning away from the mandatory imposition of the death penalty, purports to express no opinion as to the constitutionality of a mandatory statute for “an extremely narrow category of homicide, such as murder by a prisoner serving a life sentence.” See ante, at 287 n. 7. Yet if “particularized consideration” is to be required in every case under the doctrine expressed in Part III-C, such a reservation in Part III-A is disingenuous at best.

None of the cases half-heartedly cited by the plurality in Part III-C comes within a light-year of establishing the proposition that individualized consideration is a constitutional requisite for the imposition of the death penalty. Pennsylvania ex rel. Sullivan v. Ashe, 302 U. S. 61 (1937), upheld against a claim of violation of *322the Equal Protection Clause a Pennsylvania statute which made the sentence imposed upon a convict breaking out of a penitentiary dependent upon the length of the term which he was serving at the time of the break. In support of its conclusion that Pennsylvania had not denied the convict equal protection, the Court observed:

“The comparative gravity of criminal offenses and whether their consequences are more or less injurious are matters for [the State's] determination. ... It may inflict a deserved penalty merely to vindicate the law or to deter or to reform the offender or for all of these purposes. For the determination of sentences, justice generally requires consideration of more than the particular acts by which the crime was committed and that there be taken into account the circumstances of the offense together with the character and propensities of the offender. His past may be taken to indicate his present purposes and tendencies and significantly to suggest the period of restraint and the kind of discipline that ought to be imposed upon him.” Id., at 55.

These words of Mr. Justice Butler, speaking for the Court in that case, and those of Mr. Justice Black in Williams v. New York, 337 U. S. 241 (1949), the other opinion relied on by the plurality, lend no support whatever to the principle that the Constitution requires individualized consideration. This is not surprising, since even if such a doctrine had respectable support, which it has not, it is unlikely that either Mr. Justice Butler or Mr. Justice Black would have embraced it.

The plurality also relies upon the indisputable proposition that “death is different” for the result which it reaches in Part III-C. But the respects in which death is “different” from other punishment which may be im*323posed upon convicted criminals do not seem to me to establish the proposition that the Constitution requires individualized sentencing.

One of the principal reasons why death is different is because it is irreversible; an executed defendant cannot be brought back to life. This aspect of the difference between death and other penalties would undoubtedly support statutory provisions for especially careful review of the fairness of the trial, the accuracy of the factfinding process, and the fairness of the sentencing procedure where the death penalty is imposed. But none of those aspects of the death sentence is at issue here. Petitioners were found guilty of the crime of first-degree murder in a trial the constitutional validity of which is unquestioned here. And since the punishment of death is conceded by the plurality not to be a cruel and unusual punishment for such a crime, the irreversible aspect of the death penalty has no connection whatever with any requirement for individualized consideration of the sentence.

The second aspect of the death penalty which makes it "different” from other penalties is the fact that it is indeed an ultimate penalty, which ends a human life rather than simply requiring that a living human being be confined for a given period of time in a penal institution. This aspect of the difference may enter into the decision of whether or not it is a “cruel and unusual” penalty for a given offense. But since in this case the offense was first-degree murder, that particular inquiry need proceed no further.

Phe plurality’s insistence on individualized consideration of the sentencing, therefore, does not depend upon any traditional application of the prohibition against cruel and unusual punishment contained in the Eighth Amendment. The punishment here is concededly not *324cruel and unusual, and that determination has traditionally ended judicial inquiry in our cases construing the Cruel and Unusual Punishments Clause. Trop v. Dulles, 356 U. S. 86 (1958); Robinson v. California, 370 U. S. 660 (1962); Louisiana ex rel, Francis v. Resweber, 329 U. S. 459 (1947); Wilkerson v. Utah, 99 U. S. 130 (1879). What the plurality opinion has actually done is to import into the Due Process Clause of the Fourteenth Amendment what it conceives to be desirable procedural guarantees where the punishment of death, con-cededly not cruel and unusual for the crime of which the defendant was convicted, is to be imposed. This is squarely contrary to McGautha, and unsupported by any other decision of this Court.

I agree with the conclusion of the plurality, and with that of Mr. Justice White, that death is not a cruel and unusual punishment for the offense of which these petitioners were convicted. Since no member of the Court suggests that the trial which led to those convictions in any way fell short of the standards mandated by the Constitution, the judgments of conviction should be affirmed. The Fourteenth Amendment, giving the fullest scope to its “majestic generalities,” Fay v. New York, 332 U. S. 261, 282 (1947), is conscripted rather than interpreted when used to permit one but not another system for imposition of the death penalty.

Judgment of the Court, and opinion of

Mr. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens, announced by Mr. Justice Stevens.

The question in this case is whether the imposition of the sentence of death for the crime of first-degree murder under the law of Louisiana violates the Eighth and Fourteenth Amendments.

I

On August 18, 1973, in the early hours of the morning, Richard G. Lowe was found dead in the office of the Lake Charles, La., gas station where he worked. He had been shot four times in the head. Four men — the petitioner, Huey Cormier, Everett Walls, and Calvin Arceneaux — were arrested for complicity in the murder. The petitioner was subsequently indicted by a grand jury on a presentment that he “[d]id unlawfully with the specific intent to kill or to inflict great bodily harm, while engaged in the armed robbery of Richard G. Lowe, commit first degree murder by killing one Richard G. Lowe, in violation of Section One (1) of LSA-R. S. 14:30.”

At the petitioner’s trial, Cormier, Walls, and Arceneaux testified for the prosecution. Their testimony established that just before midnight on August 17, the petitioner discussed with Walls and Cormier the subject of “ripping off that old man at the station,” and that on the early morning of August 18, Arceneaux and the petitioner went to the gas station on the pretext of seeking employment. After Lowe told them that there were no jobs available they surreptitiously made their way into *328the office of the station, where Arceneaux removed a pistol from a desk drawer. The petitioner insisted on taking possession of the pistol. When Lowe returned to the office, the petitioner and Arceneaux assaulted him and then shoved him into a small back room. Shortly thereafter a car drove up. Arceneaux went out and, posing as the station attendant, sold the motorist about three dollars’ worth of gasoline. While still out in front, Arceneaux heard four shots from inside the station. He went back inside and found the petitioner gone and Lowe lying bleeding on the floor. Arceneaux grabbed some empty “money bags” and ran.

The jury found the petitioner guilty as charged. As required by state law, the trial judge sentenced him to death. The Supreme Court of Louisiana affirmed the judgment. 319 So. 2d 317 (1975). We granted certiorari, 423 U. S. 1082 (1976), to consider whether the imposition of the death penalty in this case violates the Eighth and Fourteenth Amendments of the United States Constitution.

II

The Louisiana Legislature in 1973 amended the state statutes relating to murder and the death penalty in apparent response to this Court’s decision in Furman v. Georgia, 408 U. S. 238 (1972). Before these amendments, Louisiana law defined the crime of “murder” as the killing of a human being by an offender with a specific intent to kill or to inflict great bodily harm, or by an offender engaged in the perpetration or attempted perpetration of certain serious felonies, even without an intent to kill.1 The jury was free to return any of four ver-*329diets: guilty, guilty without capital punishment, guilty of manslaughter, or not guilty.2

In the 1973 amendments, the legislature changed this discretionary statute to a wholly mandatory one, requiring that the death penalty be imposed whenever the jury finds the defendant guilty of the newly defined crime of first-degree murder. The revised statute, under which the petitioner was charged, convicted, and sentenced, provides in part that first-degree murder is the killing of a human being when the offender has a specific intent to kill or to inflict great bodily harm and is engaged in the perpetration or attempted perpetration of aggravated kidnaping, aggravated rape, or armed robbery.3 In a *330first-degree murder case, the four responsive verdicts are now guilty, guilty of second-degree murder, guilty of manslaughter, and not guilty. La. Code Crim. Proc. Ann., Art. 814 (A)(1) (Supp. 1975). The jury must be instructed on all these verdicts, whether or not raised by the evidence or requested by the defendant.4

Under the former statute, the jury had the unfettered choice in any case where it found the defendant guilty of murder of returning either a verdict of guilty, which required the imposition of the death penalty, or a verdict of guilty without capital punishment, in which case the punishment was imprisonment at hard labor for life.5 *331Under the new statute the jury is required to determine only whether both conditions existed at the time of the killing; if there was a specific intent to kill or to inflict great bodily harm, and the offender was engaged in an armed robbery, the offense is first-degree murder and the mandatory punishment is death. If only one of these conditions existed, the offense is second-degree murder and the mandatory punishment is imprisonment at hard labor for life. Any qualification or recommendation which a jury might add to its verdict — such as a recommendation of mercy where the verdict is guilty of first-degree murder' — -is without any effect.6

III

The petitioner argues that the imposition of the death penalty under any circumstances is cruel and unusual punishment in violation of the Eighth and Fourteenth Amendments. We reject this argument for the reasons stated today in Gregg v. Georgia, ante, at 168-187.

IV

Louisiana, like North Carolina, has responded to Fur-man by replacing discretionary jury sentencing in capital cases with mandatory death sentences. Under the present Louisiana law, all persons found guilty of first-degree murder, aggravated rape, aggravated kidnaping, or treason are automatically sentenced to death. See La. Rev. Stat. Ann. §§ 14:30, 14:42, 14:44, 14:113 (1974).

There are two major differences between the Louisiana and North Carolina statutes governing first-degree murder cases. First, the crime of first-degree murder in North Carolina includes any willful, deliberate, and *332premeditated homicide and any felony murder, whereas Louisiana limits first-degree murder to five categories of homicide — killing in connection with the commission of certain felonies; killing of a fireman or a peace officer in the performance of his duties; killing for remuneration; killing with the intent to inflict harm on more than one person; and killing by a person with a prior murder conviction or under a current life sentence. Second, Louisiana employs a unique system of responsive verdicts under which the jury in every first-degree murder case must be instructed on the crimes of first-degree murder, second-degree murder, and manslaughter and must be provided with the verdicts of guilty, guilty of second-degree murder, guilty of manslaughter, and not guilty. See La. Code Crim. Proc. Ann., Arts. 809, 814 (Supp. 1975); State v. Cooley, 260 La. 768, 771, 257 So. 2d 400, 401 (1972). By contrast, in North Carolina instructions on lesser included offenses must have a basis in the evidence adduced at trial. See State v. Spivey, 151 N. C. 676, 65 S. E. 995 (1909); cf. State v. Vestal, 283 N. C. 249, 195 S. E. 2d 297 (1973).

That Louisiana has adopted a different and somewhat narrower definition of first-degree murder than North Carolina is not of controlling constitutional significance. The history of mandatory death penalty statutes indicates a firm societal view that limiting the scope of capital murder is an inadequate response to the harshness and inflexibility of a mandatory death sentence statute. See Woodson v. North Carolina, ante, at 289-296. A large group of jurisdictions first responded to- the unacceptable severity of the common-law rule of automatic death sentences for all murder convictions by narrowing the definition of capital homicide. Each of these juris*333dictions found that approach insufficient and subsequently substituted discretionary sentencing for mandatory death sentences. See Woodson v. North Carolina, ante, at 290-292.8

The futility of attempting to solve the problems of mandatory death penalty statutes by narrowing the scope of the capital offense stems from our society’s rejection of the belief that “every offense in a like legal category calls for an identical punishment without regard to the past life and habits of a particular offender.” Williams v. New York, 337 U. S. 241, 247 (1949). See also Pennsylvania ex rel. Sullivan v. Ashe, 302 U. S. 51, 55 (1937). As the dissenting Justices in Furman noted, the 19th century movement away from mandatory death sentences was rooted hi the recognition that “individual culpability is not always measured by the category of crime committed.” 408 U. S., at 402 (Burgee, C. J., joined by Blackmun, Powell, and Rehnquist, JJ., dissenting).

The constitutional vice of mandatory death sentence statutes — lack of focus on the circumstances of the particular offense and the character and propensities of the offender — is not resolved by Louisiana’s limitation of first-degree murder to various categories of killings. The diversity of circumstances presented in cases falling within the single category of killings during the commission of a specified felony, as well as the variety of possible offenders involved in such crimes, underscores the rigidity of Louisiana’s enactment and its similarity to the North Carolina statute. Even the other more narrowly drawn categories of first-degree murder in the Louisiana law afford no meaningful opportunity for consideration of mitigating factors presented by the circum*334stances of the particular crime or by the attributes of the individual offender.9

Louisiana’s mandatory death sentence statute also fails to comply with Furman’s requirement that standardless jury discretion be replaced by procedures that safeguard against the arbitrary and capricious imposition of death sentences. The State claims that it has adopted satisfactory procedures by taking all sentencing authority from juries in capital murder cases. This was accomplished, according to the State, by deleting the jury’s pre-Furman authority to return a verdict of guilty without capital punishment in any murder case. See La. Rev. Stat. Ann. § 14:30 (1974); La. Code Crim. Proc. Ann., Arts. 814, 817 (Supp. 1975).10

Under the current Louisiana system, however, every jury in a first-degree murder case is instructed on the crimes of second-degree murder and manslaughter and permitted to considér those verdicts even if there is not a scintilla of evidence to support the lesser verdicts. See La. Code Crim. Proc. Ann., Arts. 809, 814 (Supp. 1975). And, if a lesser verdict is returned, it is treated as an acquittal of all greater charges. See La. Code Crim. Proc. Ann., Art. 598 (Supp. 1975). This responsive verdict *335procedure not only lacks standards to guide the jury in selecting among first-degree murderers, but it plainly invites the jurors to disregard their oaths and choose a verdict for a lesser offense whenever they feel the death penalty is inappropriate. There is an element of capriciousness in making the jurors' power to avoid the death penalty dependent on their willingness to accept this invitation to disregard the trial judge’s instructions. The Louisiana procedure neither provides standards to channel jury judgments nor permits review to check the arbitrary exercise of the capital jury’s de facto sentencing discretion. See Woodson v. North Carolina, ante, at 302-303.11

The Louisiana statute thus suffers from constitutional deficiencies similar to those identified in the North Carolina statute in Woodson v. North Carolina, ante, p. 280. As in North Carolina, there are no standards provided to guide the jury in the exercise of its power to select, those first-degree murderers who will receive death sentences, and there is no meaningful appellate review of the jury’s *336decision. As in North Carolina, death sentences are mandatory upon conviction for first-degree murder. Louisiana’s mandatory death sentence law employs a procedure that was rejected by that State’s legislature 130 years ago 12 and that subsequently has been renounced by legislatures and juries in every jurisdiction in this Nation. See Woodson v. North Carolina, ante, at 291-296. The Eighth Amendment, which draws much of its meaning from “the evolving standards of decency that mark the progress of a maturing society,” Trop v. Dulles, 356 U. S. 86, 101 (1958) (plurality opinion), simply cannot tolerate the reintroduction of a practice so thoroughly discredited.

Accordingly, we find that the death sentence imposed upon the petitioner under Louisiana’s mandatory death sentence statute violates the Eighth and Fourteenth Amendments and must be set aside. The judgment of the Supreme Court of Louisiana is reversed insofar as it upheld the death sentence imposed upon the petitioner, and the case is remanded for further proceedings not inconsistent with this opinion.

It is so ordered.

Mr. Justice Brennan,

concurring in the judgment.

For the reasons stated in my dissenting opinion in Gregg v. Georgia, ante, p. 227, I concur in the judgment that sets aside the death sentence imposed under the Louisiana death sentence statute as violative of the Eighth and Fourteenth Amendments.

Mr. Justice Marshall,

concurring in the judgment.

For the reasons stated in my dissenting opinion in Gregg v. Georgia, ante, p. 231, I am of the view that the death penalty is a cruel and unusual punishment for*337bidden by the Eighth and Fourteenth Amendments. I therefore concur in the Court’s judgment.

Me. Chief Justice Burger,

dissenting.

I dissent for the reasons set forth in my dissent in Furman v. Georgia, 408 U. S. 238, 375 (1972).

Mr. Justice White,

with whom The Chief Justice, Mr. Justice Blackmun, and Mr. Justice Rehnquist join,

dissenting.

Under the Louisiana statutes in effect prior to 1973, there were three grades of criminal homicide — murder, manslaughter, and negligent homicide. La. Rev. Stat. Ann. § 14:29 (1951). Murder was punishable by death, La. Rev. Stat. Ann. § 14:30 (1951); but a jury finding a defendant guilty of murder was empowered to foreclose the death penalty by returning a verdict of “guilty without capital punishment.” La. Rev. Stat. Ann. § 15:409 (1951). Following Furman v. Georgia, 408 U. S. 238 (1972), which the Louisiana Supreme Court held effectively to have invalidated the Louisiana death penalty,1 the statutes were amended to provide four grades of criminal homicide: first-degree murder, second-degree murder, manslaughter, and negligent homicide. La. Rev. Stat. Ann. § 14:29 (1974 Supp.). First-degree murder was defined as the killing of a human in prescribed situations, including where the offender, with specific intent to kill or to inflict great bodily harm, takes another’s life while per*338petrating or attempting to perpetrate aggravated kidnap-ing, aggravated rape, or armed robbery. La. Rev. Stat. Ann. § 14:30 (1974). The new statute provides that “whoever commits the crime of first degree murder shall be punished by death,” and juries are no longer authorized to return guilty verdicts without capital punishment.2 As had been the case before 1973, the possible *339jury verdicts in first-degree murder cases are also specified by statute. As amended in 1973, these “responsive verdicts,” as to which juries were to be instructed in every first-degree murder case, are: “guilty,” “guilty of second degree murder,” “guilty of manslaughter,” and “not guilty.” La. Code Crim. Proc., Art. 814 (A)(1) (Supp. 1975).

The issue in this case is whether the imposition of the death penalty under this statutory scheme upon a defendant found guilty of first-degree murder is consistent with the Eighth Amendment, which forbids the infliction of “cruel and unusual punishments” and which by virtue of the Fourteenth Amendment is binding upon the States. Robinson v. California, 370 U. S. 660 (1962). I am convinced that it is and dissent from the Court’s judgment.

I

On August 18, 1973, Richard G. Lowe of Lake Charles, La., was found dead in the Texaco service station where *340he worked as an attendant. He had been shot four times in the head with a pistol which was not found on the scene, but which, as it turned out, had been kept by the station manager in a drawer near the cash register. The gun was later recovered from the owner of a bar and was traced to petitioner, who was charged with first-degree murder in an indictment alleging that “with the specific intent to kill or to inflict great bodily harm” and “while engaged in . . . armed robbery,” he had killed Richard G. Lowe.

At the trial Calvin Areeneaux, testifying for the prosecution, stated that he had participated in the robbery and that he had taken the gun from the drawer and given it to petitioner, who had said he wanted it because he had “always wanted to kill a white dude.” The attendant, who had been overpowered, remained inside the station with petitioner while Areeneaux, posing as the station attendant, went outside to tend a customer. According to Areeneaux, Lowe was shot during this interval. Another witness, Everett Walls, testified that he had declined to participate in the robbery but by chance had seen the petitioner at the station with a gun in his hand. According to a third witness, Huey Cormier, who also had refused petitioner's invitation to participate, petitioner had come to Cormier’s house early on August 18 and had said that he “had just shot that old man ... at the filling station.” Record 134-135.

The case went to the jury under instructions advising the jury of the State’s burden of proof and cf the charge in the indictment that petitioner had killed another person with “specific intent to kill or to inflict great bodily harm and done when the accused was engaged in the perpetration of armed robbery.” The elements which the State was required to prove beyond reasonable doubt were explained, including the elements of first-degree *341murder and of armed robbery.3 In accordance with the statute the court also explained the possible verdicts other than first-degree murder: “The law provides that *342in a trial of murder in the first degree, if the jury is not convinced beyond a reasonable doubt that the accused is guilty of the crime of murder in the first degree, but is *343convinced beyond a reasonable doubt that he is guilty of murder in the second degree, it should render a verdict of guilty of murder in the second degree.” The elements of second-degree murder and also of manslaughter were then explained, whereupon the court instructed:

“If you should conclude that the defendant is not guilty of murder in the first degree, but you are convinced beyond a reasonable doubt that he is guilty of murder in the second degree it would be your duty to find that defendant guilty of murder in the second degree.

“If you should conclude that the defendant is not guilty of murder in the first degree or murder in the second degree, but you are convinced beyond a reasonable doubt that he is guilty of manslaughter, it would then be your duty to find the defendant guilty of manslaughter.

“If you should conclude that the defendant is not guilty of murder in the first degree, or murder in the second degree or manslaughter, it would then be your duty to find the defendant not guilty.”

Finally, the court instructed the jury:

“To summarize, you may return any one of the following verdicts:

“1. Guilty as charged.

“2. Guilty of second degree murder.

“3. Guilty of manslaughter.

“4. Not guilty.

“Accordingly, I will now set forth the proper form *344of each verdict that may be rendered, reminding you that only one verdict shall be rendered.

“If you are convinced beyond a reasonable doubt that the defendant is guilty of the offense charged, the form of your verdict should be: 'We, the jury, find the defendant guilty as charged.’

“If you are not convinced beyond a reasonable doubt that the defendant is guilty of murder in the first degree but you are convinced beyond a reasonable doubt that the defendant is guilty of murder in the second degree, the form of your verdict would be: 'We, the jury, find the defendant guilty of second degree murder.’

“If you are not convinced beyond a reasonable doubt that the defendant is guilty of murder in the first degree or murder in the second degree, but you are convinced beyond a reasonable doubt that the defendant is guilty of manslaughter, the form of your verdict would be: 'We, the jury, find the defendant guilty of manslaughter.’

“If you are not convinced that the defendant is guilty of murder in the first degree, murder in the second degree or manslaughter, the form of your verdict would be: 'We, the jury, find the defendant not guilty.’ ”

The jury found the defendant guilty of first-degree murder and the death sentence was imposed. On appeal, the conviction was affirmed, the Louisiana Supreme Court rejecting petitioner’s challenge to the death penalty based on the Eighth Amendment. 319 So. 2d 317 (1975).

II

Petitioner mounts a double attack on the death penalty imposed upon him: First, that the statute under which his sentence was imposed is too little different from *345the provision at issue in Furman v. Georgia to escape the strictures of our decision in that case; second, that death is a cruel and unusual punishment for any crime committed by any defendant under any conditions, an argument presented in Furman and there rejected by four of the six Justices who addressed the issue. I disagree with both submissions.

I cannot conclude that the current Louisiana first-degree murder statute is insufficiently different from the statutes invalidated in Furman’s wake to avoid invalidation under that case. As I have already said, under prior Louisiana law, one of the permissible verdicts that a jury in any capital punishment case was authorized by statute and 'by its instructions to return was “guilty without capital punishment.” Dispensing with the death penalty was expressly placed within the uncontrolled discretion of the jury and in no case involved a breach of its instructions or the controlling statute. A guilty verdict carrying capital punishment required a unanimous verdict; any juror, consistent with his instruction and whatever the evidence might be, was free to vote for a verdict of guilty without capital punishment, thereby, if he persevered, at least foreclosing a capital punishment verdict at that trial.

Under this or similar jury-sentencing arrangements which were in force in Louisiana, Georgia, and most other States that authorized capital punishment, the death penalty came to be imposed less and less frequently, so much so that in Furman v. Georgia the Court concluded that in practice criminal juries, exercising their lawful discretion, were imposing it so seldom and so freakishly and arbitrarily that it was no longer serving the legitimate ends of criminal justice and had come to be cruel and unusual punishment violative of the Eighth Amendment. It was in response to this judgment that Louisiana sought to *346re-enact the death penalty as a constitutionally valid punishment by redefining the crime of first-degree murder and by making' death the mandatory punishment for those found guilty of that crime.

To implement this aim, the present Louisiana law eliminated the “guilty without capital punishment” verdict. Jurors in first-degree murder cases are no longer instructed that they have discretion to withhold capital punishment. Their instructions now are to find the defendant guilty if they believe beyond a reasonable doubt that he committed the crime with which he is charged. A verdict of guilty carries a mandatory death sentence. In the present case, the jury was instructed as to the specific elements constituting the crime of felony murder which the indictment charged. They were also directed that if they believed beyond reasonable doubt that Roberts committed these acts, they were to return a verdict of guilty as charged in the indictment. The jury could not, if it believed the defendant had committed the crime, nevertheless dispense with the death penalty.

The difference between a jury’s having and not having the lawful discretion to spare the life of the defendant is apparent and fundamental. It is undeniable that the unfettered discretion of the jury to save the defendant from death was a major contributing factor in the developments which led us to invalidate the death penalty in Furman v. Georgia. This factor Louisiana has now sought to eliminate by making the death penalty compulsory upon a verdict of guilty in first-degree murder cases. As I see it, we are now in no position to rule that the State’s present law, having eliminated the overt discretionary power of juries, suffers from the same constitutional infirmities which led this Court to invalidate the Georgia death penalty statute in Furman v. Georgia.

*347Even so, petitioner submits that in every capital case the court is required to instruct the jury with respect to lesser included offenses and that the jury therefore has unlimited discretion to foreclose the death penalty by finding the defendant guilty of a lesser included offense for which capital punishment is not authorized. The difficulty with the argument is illustrated by the instructions in this case. The jury was not instructed that it could in its discretion convict of a lesser included offense. The jury's plain instructions, instead, were to return a verdict of guilty of murder as charged if it believed from the evidence that Roberts had committed the specific acts constituting the offense charged and defined by the court. Only if they did not believe Roberts had committed the acts charged in the indictment were the jurors free to consider whether he was guilty of the lesser included offense of second-degree murder, and only if they did not find beyond a reasonable doubt that Roberts was guilty of second-degree murder were they free to consider the offense of manslaughter. As the Supreme Court of Louisiana said in State v. Hill, 297 So. 2d 660, 662 (1974), and repeated in this case, 319 So. 2d, at 322, “the use of these lesser verdicts ... is contingent upon the jury finding insufficient evidence to convict the defendant of first degree murder, with which he is charged.” See also State v. Selman, 300 So. 2d 467, 473 (La. 1974), cert. pending, No. 74-6065.

It is true that the jury in this case, like juries in other capital cases in Louisiana and elsewhere, may violate its instructions and convict of a lesser included offense despite the evidence. But for constitutional purposes I am quite unwilling to equate the raw power of nullification with the unlimited discretion extended jurors under prior Louisiana statutes. In McGautha v. California, 402 U. S. 183 (1971), we rejected the argument that vesting *348standardless sentencing discretion in the jury was unconstitutional under the Due Process Clause. In arriving at that judgment, we noted that the practice of jury sentencing had emerged from the “rebellion against the common-law rule imposing a mandatory death sentence on all convicted murderers,” id., at 198, and from the unsatisfactory experience with attempting to define the various grades of homicide and to specify those for which the death penalty was required. Vesting complete' sentencing power in the jury was the upshot. The difficulties adverted to in McGautha, however, including that of jury nullification, are inadequate to require invalidation of the Louisiana felony-murder rule on the ground that jurors will so often and systematically refuse to follow their instructions that the administration of the death penalty under the current law will not be substantially different from that which obtained under prior statutes.

Nor am I convinced that the Louisiana death penalty for first-degree murder is substantially more vulnerable because the prosecutor is vested with discretion as to the selection and filing of charges, by the practice of plea bargaining or by the power of executive clemency. Petitioner argues that these characteristics of the criminal justice system in Louisiana, combined with the discretion arguably left to the jury as discussed above, insure that the death penalty will be as seldom and arbitrarily applied as it was under the predecessor statutes. The Louisiana statutes, however, define the elements of first-degree murder, and I cannot accept the assertion that state prosecutors will systematically fail to file first-degree murder charges when the evidence warrants it or to seek convictions for first-degree murder on less than adequate evidence. Of course, someone must exercise discretion and judgment as to what charges are to be filed and against whom; but this essential process is *349nothing more than the rational enforcement of the State’s criminal law and the sensible operation of the criminal justice system. The discretion with which Louisiana’s prosecutors are invested and which appears to be no more than normal, furnishes no basis for inferring that capital crimes will be prosecuted so arbitrarily and infrequently that the present death penalty statute is invalid under Furman v. Georgia.

I have much the same reaction to plea bargaining and executive clemency. A prosecutor may seek or accept pleas to lesser offenses where he is not confident of his first-degree murder case, but this is merely the proper exercise of the prosecutor’s discretion as I have already discussed. So too, as illustrated by this case and the North Carolina case, Woodson v. North Carolina, ante, p. 280, some defendants who otherwise would have been tried for first-degree murder, convicted, and sentenced to death are permitted to plead to lesser offenses because they are willing to testify against their codefend-ants. This is a grisly trade, but it is not irrational; for it is aimed at insuring the successful conclusion of a first-degree murder case against one or more other defendants. Whatever else the practice may be, it is neither inexplicable, freakish, nor violative of the Eighth Amendment. Nor has it been condemned by this Court under other provisions of the Constitution. Santobello v. New York, 404 U. S. 257 (1971); North Carolina v. Alford, 400 U. S. 25 (1970); Parker v. North Carolina, 397 U. S. 790 (1970); Brady v. United States, 397 U. S. 742 (1970). See also Chaffin v. Stynchcombe, 412 U. S. 17, 30-31 (1973).

As for executive clemency, I cannot assume that this power, exercised by governors and vested in the President by Art. II, § 2, of the Constitution, will be used in a standardless and arbitrary manner. It is more reason*350able to expect the power to be exercised by the Executive Branch whenever it is concluded that the criminal justice system has unjustly convicted a defendant of first-degree murder and sentenced him to death. The country’s experience with the commutation power does not suggest that it is a senseless lottery, that it operates in an arbitrary or discriminatory manner, or that it will lead to reducing the death penalty to a merely theoretical threat that is imposed only on the luckless few.

I cannot conclude, as do Mr. Justice Stewart, Mr. Justice Powell, and Mr. Justice Stevens (hereinafter the plurality), that under the present Louisiana law, capital punishment will occur so seldom, discriminatorily, or freakishly that it will fail to satisfy the Eighth Amendment as construed and applied in Furman v. Georgia.

Ill

I also cannot agree with the petitioner’s other basic argument that the death penalty, however imposed and for whatever crime, is cruel and unusual punishment. The opposing positions on this issue, as well as the history of the death penalty, were fully canvassed by various Justices in their separate opinions in Furman v. Georgia, and these able and lucid presentations need not be repeated here. It is plain enough that the Constitution drafted by the Framers expressly made room for the death penalty. The Fifth Amendment provides that “no person shall be held to answer for a capital, or otherwise infamous crime, unless on a presentment or indictment of a Grand Jury . . .” and that no person shall be “twice put in jeopardy of life or limb . . . nor be deprived of life . . . without due process of law.”- The Fourteenth Amendment, adopted three-quarters of a century later, likewise enjoined the States from depriving any person of “his life” without due process of law. *351Since the very first Congress, federal law has defined crimes for which the death penalty is authorized. Capital punishment has also been part of the criminal justice system of the great majority of the States ever since the Union was first organized. Until Furman v. Georgia, this Court’s opinions, if they did not squarely uphold the death penalty, consistently assumed its constitutionality. Wilkerson v. Utah, 99 U. S. 130 (1879); In re Kemmler, 136 U. S. 436 (1890); Louisiana ex rel. Francis v. Resweber, 329 U. S. 459 (1947); McGautha v. California, 402 U. S. 183 (1971); Witherspoon v. Illinois, 391 U. S. 510 (1968). In Trop v. Dulles, 356 U. S. 86, 99 (1958), four Members of the Court — Mr. Chief Justice Warren and Justices Black, Douglas, and Whittaker — agreed that “[wjhatever the arguments may be against capital punishment, both on moral grounds and in terms of accomplishing the purposes of punishment — and they are forceful — the death penalty has been employed throughout our history, and, in a day when it is still widely accepted, it cannot be said to violate the constitutional concept of cruelty.”

Until Furman v. Georgia, this was the consistent view of the Court and of every Justice who in a published opinion had addressed the question of the validity of capital punishment under the Eighth Amendment. In Furman, it was concluded by at least two Justices4 that the death penalty had become unacceptable to the great majority of the people of this country and for that reason, alone or combined with other reasons, was invalid *352under the Eighth Amendment, which must be construed and applied to reflect the evolving moral standards of the country. Trop v. Dulles, supra, at 111; Weems v. United States, 217 U. S. 349, 378 (1910). That argument, whether or not accurate at that time, when measured by the manner in which the death penalty was being administered under the then-prevailing statutory schemes, is no longer descriptive of the country’s attitude. Since the judgment in Furman, Congress and 35 state legislatures re-enacted the death penalty for one or more crimes.5 All of these States authorize the death *353penalty for murder of one kind or another. With these profound developments in mind, I cannot say that capital punishment has been rejected by or is offensive to the prevailing attitudes and moral presuppositions in the United States or that it is always an excessively cruel or severe punishment or always a disproportionate punishment for any crime for which it might be imposed.6 These grounds for invalidating the death penalty are foreclosed by recent events, which this Court must accept as demonstrating that capital punishment is acceptable to the contemporary community as just punishment for at least some intentional killings.

It is apparent also that Congress and 35 state legislatures are of the view that capital punishment better serves the ends of criminal justice than would life imprisonment and that it is therefore not excessive in the sense that it serves no legitimate legislative or social ends. Petitioner Roberts, to the contrary, submits that life imprisonment obviously would better serve the end of reformation or rehabilitation and that there is no satisfactory evidence that punishing by death serves more effectively than does life imprisonment the other major ends of imposing serious criminal sanctions: incapacitation *354of the prisoner, the deterrence of others, and moral re-enforcement and retribution. The death penalty is therefore cruel and unusual, it is argued, because it is the purposeless taking of life and the needless imposition of suffering.

The widespread re-enactment of the death penalty, it seems to me, answers any claims that life imprisonment is adequate punishment to satisfy the need for reprobation or retribution. It also seems clear enough that death finally forecloses the possibility that a prisoner will commit further crimes, whereas life imprisonment does not. This leaves the question of general deterrence as the principal battleground: Does the death penalty more effectively deter others from crime than does the threat of life imprisonment?

The debate on this subject started generations ago and is still in progress. Each side has a plethora of fact and opinion in support of its position,7 some of it quite old *355and some of it very new; but neither has yet silenced the other. I need not detail these conflicting materials, most of which are familiar sources. It is quite apparent that the relative efficacy of capital punishment and life imprisonment to deter others from crime remains a matter about which reasonable men and reasonable legislators may easily differ. In this posture of the case, it would be neither a proper or wise exercise of the power of judicial review to refuse to accept the reasonable conclusions of Congress and 35 state legislatures that there are indeed certain circumstances in which the death penalty is the more efficacious deterrent of crime.

It will not do to denigrate these legislative judgments as some form of vestigial savagery or as purely retributive in motivation; for they are solemn judgments, reasonably based, that imposition of the death penalty will save the lives of innocent persons. This concern for life and human values and the sincere efforts of the States to pursue them are matters of the greatest moment with which the judiciary should be most reluctant to interfere. The issue is not whether, had we been legislators, we would have supported or opposed the capital punishment statutes presently before us. The question here under discussion is whether the Eighth Amendment requires us to interfere with the enforcement of these statutes on the grounds that a sentence of life imprisonment for the crimes at issue would as well have served the ends of criminal justice. In my view, the Eighth Amend-*356raent provides no warrant for overturning these convictions on these grounds.

IV

The plurality offers two additional reasons for invalidating the Louisiana statute, neither of which had been raised by the parties and with both of which I disagree.

The plurality holds the Louisiana statute unconstitutional for want of a separate sentencing proceeding in which the sentencing authority may focus on the sentence and consider some or all of the aggravating and mitigating circumstances. In McGautha v. California, 402 U. S. 183 (1971), after having heard the same issues argued twice before in Maxwell v. Bishop, 398 U. S. 262 (1970), we specifically rejected the claims that a defendant’s “constitutional rights were infringed by permitting the jury to impose the death penalty without governing standards” and that “the jury’s imposition of the death sentence in the same proceeding and verdict as determined the issue of guilt was [not] constitutionally permissible.” 402 U. S., at 185. With respect to the necessity of a bifurcated criminal trial, we had reached essentially the same result in Spencer v. Texas, 385 U. S. 554 (1967). In spite of these cases, the plurality holds that the State must provide a procedure under which the sentencer may separately consider the character and record of the individual defendant, along with the circumstances of the particular offense, including any mitigating circumstances that may exist. For myself, I see no reason to reconsider McGautha and would not invalidate the Louisiana statute for its failure to provide what McGautha held it need not provide. I still share the concluding remarks of the Court in McGautha v. California:

“It may well be, as the American Law Institute and the National Commission on Reform of Federal *357Criminal Laws have concluded, that bifurcated trials and criteria for jury sentencing discretion are superior means of dealing with capital cases if the death penalty is to be retained at all. But the Federal Constitution, which marks the limits of our authority in these cases, does not guarantee trial procedures that are the best of all worlds, or that accord with the most enlightened ideas of students of the infant science of criminology, or even those that measure up to the individual predilections of members of this Court. See Spencer v. Texas, 385 U. S. 554 (1967). The Constitution requires no more than that trials be fairly conducted and that guaranteed rights of defendants be scrupulously respected. From a constitutional standpoint we cannot conclude that it is impermissible for a State to consider that the compassionate purposes of jury sentencing in capital cases are better served by having the issues of guilt and punishment determined in a single trial than by focusing the jury’s attention solely on punishment after the issue of guilt has been determined.

“Certainly the facts of these gruesome murders bespeak no miscarriage of justice. The ability of juries, unassisted by standards, to distinguish between those defendants for whom the death penalty is appropriate punishment and those for whom imprisonment is sufficient is indeed illustrated by the discriminating verdict of the jury in McGautha’s case, finding Wilkinson the less culpable of the two defendants and sparing his life.

“The procedures which petitioners challenge are those by which most capital trials in this country are conducted, and by which all were conducted until a few years ago. We have determined that these *358procedures are consistent with the rights to which petitioners were constitutionally entitled, and that their trials were entirely fair. Having reached these conclusions we have performed our task of measuring the States’ process by federal constitutional standards . . . 402 U. S., at 221-222.

Implicit in the plurality’s holding that a separate proceeding must be held at which the sentencer may consider the character and record of the accused is the proposition that States are constitutionally prohibited from considering any crime, no matter how defined, so serious that every person who commits it should be put to death regardless of extraneous factors related to his character. Quite apart from McGautha v. California, supra, I cannot agree. It is axiomatic that the major justification for concluding that a given defendant deserves to be punished is that he committed a crime. Even if the character of the accused must be considered under the Eighth Amendment, surely a State is not constitutionally forbidden to provide that the commission of certain crimes conclusively establishes that the criminal’s character is such that he deserves death. Moreover, quite apart from the character of a criminal, a State should constitutionally be able to conclude that the need to deter some crimes and that the likelihood that the death penalty will succeed in deterring these crimes is such that the death penalty may be made mandatory for all people who commit them. Nothing resembling a reasoned basis for the rejection of these propositions is to be found in the plurality opinion.

The remaining reason offered for invalidating the Louisiana statute is also infirm. It is said that the Eighth Amendment forbids the legislature to require imposition of the death penalty when the elements of the specified crime have been proved to the satisfac*359tion of the jury because historically the concept of the mandatory death sentence has been rejected by the community and departs so far from contemporary standards with respect to the imposition of capital punishment that it must be held unconstitutional.

Although the plurality seemingly makes an unlimited pronouncement, it actually stops short of invalidating any statute making death the required punishment for any crime whatsoever. Apparently there are some crimes for which the plurality in its infinite wisdom will permit the States to require the death sentence to be imposed without the additional procedures which its opinion seems to mandate. There have always been mandatory death penalties for at least some crimes, and the legislatures of at least two States have now again embraced this approach in order to serve what they deem to be their own penological goals.

Furthermore, Justices Stewart, Powell, and Stevens uphold the capital punishment statute of Texas, under which capital punishment is required if the defendant is found guilty of the crime charged and the jury answers two additional questions in the affirmative. Once that occurs, no discretion is left to the jury; death is mandatory. Although Louisiana juries are not required to answer these precise questions, the Texas law is not constitutionally distinguishable from the Louisiana system under which the jury, to convict, must find the elements of the crime, including the essential element of intent to kill or inflict great bodily harm, which, according to the instructions given in this case, must be felonious, “that is, it must be wrong or without any just cause or excuse.”

As the plurality now interprets the Eighth Amendment, the Louisiana and North Carolina statutes are infirm because the jury is deprived of all discretion once it finds the defendant guilty. Yet in the next breath it invali*360dates these statutes because they are said to invite or allow too much discretion: Despite their instructions, when they feel that defendants do not deserve to die, juries will so often and systematically disobey their instructions and find the defendant not guilty or guilty of a noncapital offense that the statute fails to satisfy the standards of Furman v. Georgia. If it is truly the case that Louisiana juries will exercise too much discretion'— and I do not agree that it is — then it seems strange indeed that the statute is also invalidated because it purports to give the jury too little discretion by making the death penalty mandatory. Furthermore, if there is danger of freakish and too infrequent imposition of capital punishment under a mandatory system such as Louisiana’s, there is very little ground for believing that juries will be any more faithful to their instructions under the Georgia and Florida systems where the opportunity is much, much greater for juries to practice their own brand of unbridled discretion.

In any event the plurality overreads the history upon which it so heavily relies. Narrowing the categories of crime for which the death penalty was authorized reflected a growing sentiment that death was an excessive penalty for many crimes, but I am not convinced, as apparently the plurality is, that the decision to vest discretionary sentencing power in the jury was a judgment that mandatory punishments were excessively cruel rather than merely a legislative response to avoid jury nullifications which were occurring with some frequency. That legislatures chose jury sentencing as the least troublesome of two approaches hardly proves legislative rejection of mandatory sentencing. State legislatures may have preferred to vest discretionary sentencing power in a jury rather than to have guilty defendants go scot-free; but I doubt that these events necessarily reflect *361an affirmative legislative preference for discretionary systems or support an inference that legislatures would have chosen them even absent their experience with jury nullification.

Nor does the fact that juries at times refused to convict despite the evidence prove that the mandatory nature of the sentence was the burr under the jury’s saddle rather than that one or more persons on those juries were opposed in principle to the death penalty under whatever system it might be authorized or imposed. Surely if every nullifying jury had been interrogated at the time and had it been proved to everyone’s satisfaction that all or a large part of the nullifying verdicts occurred because certain members of these juries had been opposed to the death penalty in any form, rather than because the juries involved were reluctant to impose the death penalty on the particular defendants before them, it could not be concluded that either those juries or the country had condemned mandatory punishments as distinguished from the death penalty itself. The plurality nevertheless draws such an inference even though there is no more reason to infer that jury nullification occurred because of opposition to the death penalty in particular cases than because one or more of the 12 jurors on the critical juries were opposed to the death penalty in any form and stubbornly refused to participate in a guilty verdict. Of course, the plurality does not conclude that the death penalty was itself placed beyond legislative resuscitation either by jury nullification under mandatory statutes or by the erosion of the death penalty under the discretionary-sentencing systems that led to the judgment in Furman v. Georgia. I see no more basis for arriving at a contrary conclusion with respect to the mandatory statutes.

Louisiana and North Carolina have returned to the *362mandatory capital punishment system for certain crimes.8 Their legislatures have not deemed mandatory punishment, once the crime is proved, to be unacceptable; nor have their juries rejected it, for the death penalty has been imposed with some regularity. Perhaps we would *363prefer that these States had adopted a different system, but the issue is not our individual preferences but the constitutionality of the mandatory systems chosen by these two States. I see no warrant under the Eighth Amendment for refusing to uphold these statutes.

Indeed, the more fundamental objection than the plurality’s muddled reasoning is that in Gregg v. Georgia, ante, at 174-176, it lectures us at length about the role and place of the judiciary and then proceeds to ignore its own advice, the net effect being to suggest that observers of this institution should pay more attention to what we do than, what we say. The plurality claims that it has not forgotten what the past has taught about the limits of judicial review; but I fear that it has again surrendered to the temptation to make policy for and to attempt to govern the country through a misuse of the powers given this Court under the Constitution.

Y

I conclude that § 14:30 of the Louisiana statutes imposing the death penalty for first-degree murder is not unconstitutional under the Eighth Amendment. I am not impressed with the argument that this result reduces the Amendment to little more than mild advice from the Framers to state legislators. Weems, Trop, and Furman bear witness to the contrary.

For the foregoing reasons, I dissent.

Mr. Justice Blackmun,

dissenting.

I dissent for the reasons set forth in my dissent in Furman v. Georgia, 408 U. S. 238, 405-414 (1972), and in the other dissenting opinions I joined in that case. Id., at 375, 414, and 465.

Mr. Justice White

announced the judgment of the Court and filed an opinion in which Mr. Justice Stewart, Mr. Justice Blackmun, and Mr. Justice Stevens, joined.

Georgia Code Ann. § 26-2001 (1972) provides that “ [a] person convicted of rape shall be punished by death or by imprisonment for life, or by imprisonment for not less than one nor more than 20 years." 1 Punishment is determined by a jury in a separate sentencing proceeding in which at least one of the statutory aggravating circumstances must be found before the death penalty may be imposed.2 Petitioner Coker was convicted of rape and sentenced to death. Both the conviction and the sentence were affirmed by the Georgia Supreme Court. Coker was granted a writ of certiorari, 429 U. S. 815, limited to the single claim, rejected by the Georgia court, that the punishment of death for rape violates the Eighth Amendment, which proscribes “cruel and unusual punishments” and which must be observed by the States as well as the Federal Government. Robinson v. California, 370 U. S. 660 (1962).

*587I

While serving various sentences for murder, rape, kidnap-ing, and aggravated assault, petitioner escaped from the Ware Correctional Institution near Waycross, Ga., on September 2, 1974. At approximately 11 o’clock that night, petitioner entered the house of Allen and Elnita Carver through an unlocked kitchen door. Threatening the couple with a “board,” he tied up Mr. Carver in the bathroom, obtained a knife from the kitchen, and took Mr. Carver’s money and the keys to the family car. Brandishing the knife and saying “you know what’s going to happen to you if you try anything, don’t you,” Coker then raped Mrs. Carver. Soon thereafter, petitioner drove away in the Carver car, taking Mrs. Carver with him. Mr. Carver, freeing himself, notified the police; and not long thereafter petitioner was apprehended. Mrs. Carver was unharmed.

Petitioner was charged with escape, armed robbery, motor vehicle theft, kidnaping, and rape. Counsel was appointed to represent him. Having been found competent to stand trial, he was tried. The jury returned a verdict of guilty, rejecting his general plea of insanity. A sentencing hearing was then conducted in accordance with the procedures dealt with at length in Gregg v. Georgia, 428 U. S. 153 (1976), where this Court sustained the death penalty for murder when imposed pursuant to the statutory procedures.3 The jury was *588instructed that it could consider as aggravating circumstances whether the rape had been committed by a person with a prior record of conviction for a capital felony and whether the rape *589had been committed in the course of committing another capital felony, namely, the armed robbery of Allen Carver. The court also instructed, pursuant to statute, that even if *590aggravating circumstances were present, the death penalty need not be imposed if the jury found they were outweighed by mitigating circumstances, that is, circumstances not constituting justification or excuse for the offense in question, *591“but which, in fairness and mercy, may be considered as extenuating or reducing the degree” of moral culpability or punishment. App. 300. The jury’s verdict on the rape count was death by electrocution. Both aggravating circumstances on which the court instructed were found to be present by the jury.

II

Furman v. Georgia, 408 U. S. 238 (1972), and the Court’s decisions last Term in Gregg v. Georgia, 428 U. S. 153 (1976) ; Proffitt v. Florida, 428 U. S. 242 (1976); Jurek v. Texas, 428 U. S. 262 (1976); Woodson v. North Carolina, 428 U. S. 280 (1976); and Roberts v. Louisiana, 428 U. S. 325 (1976), make unnecessary the recanvassing of certain critical aspects of the controversy about the constitutionality of capital punishment. It is now settled that the death penalty is not invariably cruel and unusual punishment within the meaning of the Eighth Amendment; it is not inherently barbaric or an unacceptable mode of punishment for crime; neither is it always disproportionate to the crime for which it is imposed. It is also established that imposing capital punishment, at least for murder, in accordance with the procedures provided under the Georgia statutes saves the sentence from the infirmities which led the Court to invalidate the prior Georgia capital punishment statute in Furman v. Georgia, supra.

In sustaining the imposition of the death penalty in Gregg, *592however, the Court firmly embraced the holdings and dicta from prior cases, Furman v. Georgia, supra; Robinson v. California, 370 U. S. 660 (1962); Trop v. Dulles, 356 U. S. 86 (1958); and Weems v. United States, 217 U. S. 349 (1910), to the effect that the Eighth Amendment bars not only those punishments that are “barbaric” but also those that are “excessive” in relation to the crime committed. Under Gregg, a punishment is “excessive” and unconstitutional if it (1) makes no measurable contribution to acceptable goals of punishment and hence is nothing more than the purposeless and needless imposition of pain and suffering; or (2) is grossly out of proportion to the severity of the crime. A punishment might fail the test on either ground. Furthermore, these Eighth Amendment judgments should not be, or appear to be, merely the subjective views of individual Justices; judgment should be informed by objective factors to the maximum possible extent. To this end, attention must be given to the public attitudes concerning a particular sentence — history and precedent, legislative attitudes, and the response of juries reflected in their sentencing decisions are to be consulted. In Gregg, after giving due regard to such sources, the Court’s judgment was that the death penalty for deliberate murder was neither the purposeless imposition of severe punishment nor a punishment grossly disproportionate to the crime. But the Court reserved the question of the constitutionality of the death penalty when imposed for other crimes. 428 U. S., at 187 n. 35.

Ill

That question, with respect to rape of an adult woman, is now before us. We have concluded that a sentence of death is grossly disproportionate and excessive punishment for the crime of rape and is therefore forbidden by the Eighth Amendment as cruel and unusual punishment.4

*593A

As advised by recent cases, we seek guidance in history and from the objective evidence of the country’s present judgment concerning the acceptability of death as a penalty for rape of an adult woman. At no time in the last 50 years have a majority of the States authorized death as a punishment for rape. In 1925, 18 States, the District of Columbia, and the Federal Government authorized capital punishment for the rape of an adult female.5 By 1971 just prior to the decision in Furman v. Georgia, that number had declined, but not substantially, to 16 States plus the Federal Government.6 Furman then invalidated most of the capital punishment statutes in this country, including the rape statutes, because, among other reasons, of the manner in which the death penalty was imposed and utilized under those laws.

With their death penalty statutes for the most part invalidated, the States were faced with the choice of enacting modified capital punishment laws in an attempt to satisfy the requirements of Furman or of being satisfied with life imprisonment as the ultimate punishment for any offense. Thirty-*594five. States immediately reinstituted the death penalty for at least limited kinds of crime. Gregg v. Georgia, 428 U. S., at 179 n. 23. This public judgment as to the acceptability of capital punishment, evidenced by the immediate, post-Furman legislative reaction in a large majority of the States, heavily influenced the Court to sustain the death penalty for murder in Gregg v. Georgia, supra, at 179-182.

But if the “most marked indication of society’s endorsement of the death penalty for murder is the legislative response to Furman,” Gregg v. Georgia, supra, at 179-180, it should also be a telling datum that the public judgment with respect to rape, as reflected in the statutes providing the punishment for that crime, has been dramatically different. In reviving death penalty laws to satisfy Furman’s mandate, none of the States that had not previously authorized death for rape chose to include rape among capital felonies. Of the 16 States in which rape had been a capital offense, only three provided the death penalty for rape of an adult woman in their revised statutes — Georgia, North Carolina, and Louisiana. In the latter two States, the death penalty was mandan tory for those found guilty, and those laws were invalidated by Woodson and Roberts. When Louisiana and North Carolina, responding to those decisions, again revised their capital punishment laws, they re-enacted the death penalty for murder but not for rape; none of the seven other legislatures that to our knowledge have amended or replaced their death penalty statutes since July 2, 1976, including four States (in addition to Louisiana and North Carolina) that had authorized the death sentence for rape prior to 1972 and had reacted to Furman with mandatory statutes, included rape among the crimes for which death was an authorized punishment.7

*595Georgia argues that 11 of the 16 States that authorized death for rape in 1972 attempted to comply with Furman by enacting arguably mandatory death penalty legislation and that it is very likely that, aside from Louisiana and North Carolina, these States simply chose to eliminate rape as a capital offense rather than to require death for each and every instance of rape.8 The argument is not without force; but 4 of the 16 States did not take the mandatory course and also did not continue rape of an adult woman as a capital offense. Further, as we have indicated, the legislatures of 6 of the 11 arguably mandatory States have revised their death penalty laws since Woodson and Roberts without enacting a new death penalty for rape. And this is to say nothing of 19 other States that enacted nonmandatory, post-Furman statutes and chose not to sentence rapists to death.

It should be noted that Florida, Mississippi, and Tennessee also authorized the death penalty in some rape cases, but only where the victim was a child and the rapist an adults.9 The Tennessee statute has since been invalidated because the death sentence was mandatory. Collins v. State, 550 S. W. 2d 643 (Tenn. 1977). The upshot is that Georgia is the sole jurisdic*596tion in the United States at the present time that authorizes a sentence of death when the rape victim is an adult woman, and only two other jurisdictions provide capital punishment when the victim is a child.

The current judgment with respect to the death penalty for rape is not wholly unanimous among state legislatures, but it obviously weighs very heavily on the side of rejecting capital punishment as a suitable penalty for raping an adult woman.10

B

It was also observed in Gregg that “[t]he jury . . . is a significant and reliable objective index of contemporary values because it is so directly involved,” 428 U. S., at 181, and that it is thus important to look to the sentencing decisions that juries have made in the course of assessing whether capital punishment is an appropriate penalty for the crime being tried. Of course, the jury’s judgment is meaningful only where the jury has an appropriate measure of choice as to whether the death penalty is to be imposed. As far as execution for rape is concerned, this is now true only in Georgia and in Florida; and in the latter State, capital punishment is authorized only for the rape of children.

According to the factual submissions in this Court, out of all rape convictions in Georgia since 1973 — and that total number has not been tendered — 63 cases had been reviewed by the Georgia Supreme Court as of the time of oral argument; and of these, 6 involved a death sentence, 1 of which was set aside, leaving 5 convicted rapists now under sentence *597of death in the State of Georgia. Georgia juries have thus sentenced rapists to death six times since 1973. This obviously is not a negligible number; and the State argues that as a practical matter juries simply reserve the extreme sanction for extreme, cases of rape and that recent experience surely does not prove that jurors consider the death penalty to be a disproportionate punishment for every conceivable instance of rape, no matter how aggravated. Nevertheless, it is true that in the vast majority of cases, at least 9 out of 10, juries have not imposed the death sentence.

IV

These recent events evidencing the attitude of state legislatures and sentencing juries do not wholly determine this controversy, for the Constitution contemplates that in the end our own judgment will be brought to bear on the question of the acceptability of the death penalty under the Eighth Amendment. Nevertheless, the legislative rejection of capital punishment for rape strongly confirms our own judgment, which is that death is indeed a disproportionate penalty for the crime of raping an adult woman.

We do not discount the seriousness of rape as a crime. ..It is highly reprehensible, both in a moral sense and in its almost total contempt for the personal integrity and autonomy of the female victim and for the latter’s privilege of choosing those with whom intimate relationships are to be established. Short of homicide, it is the “ultimate violation of self.”11 It is also a violent crime because it normally involves force, or the threat of force or intimidation, to overcome the will and the capacity of the victim to resist. Rape is very often aecom-*598panied by physical injury to the female and can also inflict mental and psychological damage.12 Because it undermines the community’s sense of security, there is public injury as well.

Rape is without doubt deserving of serious punishment; but in terms of moral depravity and of the injury to the person and to the public, it does not compare with murder, which does involve the unjustified taking of human life. Although it may be accompanied by another crime, rape by definition does not include the death of or even the serious injury to another person.13 The murderer kills; the rapist, if no more than that, does not. Life is over for the victim of the murderer; for the rape victim, life may not be nearly so happy as it was, but it is not over and normally is not beyond repair. We have the abiding conviction that the death penalty, which “is unique in its severity and irrevocability,” Gregg v. Georgia, 428 U. S., at 187, is an excessive penalty for the rapist who, as such, does not take human life.

This does not end the matter; for under Georgia law, death may not be imposed for any capital offense, including rape, unless the jury or judge finds one of the statutory aggravating circumstances and then elects to impose that sentence. Ga. Code §26-3102 (1977); Gregg v. Georgia, supra, at 165-166. For the rapist to be executed in Georgia, it must therefore be found not only that he committed rape but also that one or more of the following aggravating circumstances were present: (1) that the rape was committed by a person with a prior record of conviction for a capital felony; (2) that the rape was committed while the offender was engaged in the commission of another capital felony, or aggravated battery; or (3) the rape “was outrageously or wantonly vile, horrible or *599inhuman in that it involved torture, depravity of mind, or aggravated battery to the victim.” 14 Here, the first two of these aggravating circumstances were alleged and found by the jury.

Neither of these circumstances, nor both of them together, change our conclusion that the death sentence imposed on Coker is a disproportionate punishment for rape. Coker had prior convictions for capital felonies — rape, murder, and kid-naping — but these prior convictions do not change the fact that the instant crime being punished is a rape not involving the taking of life.

It is also true that the present rape occurred while Coker was committing armed robbery, a felony for which the Georgia statutes authorize the death penalty.16 But Coker was tried for the robbery offense as well as for rape and received a separate life sentence for this crime; the jury did not deem the robbery itself deserving of the death penalty, even though accompanied by the aggravating circumstance, which was stipulated, that Coker had been convicted of a prior capital crime.15

*600We note finally that in Georgia a person commits murder when he unlawfully and with malice aforethought, either express or implied, causes the death of another human being. He also commits that crime when in the commission of a felony he causes the death of another human being, irrespective of malice. But even where the killing is deliberate, it is not punishable by death absent proof of aggravating circumstances. It is difficult to accept the notion, and we do not, that the rapist, with or without aggravating circumstances, should be punished more heavily than the deliberate killer as long as the rapist does not himself take the life of his victim. The judgment of the Georgia Supreme Court upholding the death sentence is reversed, and the case is remanded to that court for further proceedings not inconsistent with this opinion.

So ordered.

Me. Justice Brennan-,

concurring in the judgment.

Adhering to my view that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, Gregg v. Georgia, 428 U. S. 153, 227 (1976) (dissenting opinion), I concur in the judgment of the Court setting aside the death sentence imposed under the Georgia rape statute.

Mr. Justice Marshall,

concurring in the judgment.

In Gregg v. Georgia, 428 U. S. 153, 231 (1976) (dissenting opinion), I stated: “In Furman v. Georgia, 408 U. S. 238, 314 (1972) (concurring opinion), I set forth at some length my views on the basic issue presented to the Court in these cases. The death penalty, I concluded, is a cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments. That continues to be my view.”

*601I then explained in some detail my reasons for reaffirming my position. I continue to adhere to those views in concurring in the judgment of the Court in this case.

Mr. Justice Powell,

concurring in the judgment in part and dissenting in part.

I concur in the judgment of the Court on the facts of this case, and also in the plurality’s reasoning supporting the view that ordinarily death is disproportionate punishment for the crime of raping an adult woman. Although rape invariably is a reprehensible crime, there is no indication that petitioner’s offense was committed with excessive brutality or that the victim sustained serious or lasting injury. The plurality, however, does not limit its holding to the case before us or to similar cases. Rather, in an opinion that ranges well beyond what is necessary, it holds that capital punishment always— regardless of the circumstances — is a disproportionate penalty for the crime of rape.

The Georgia statute, sustained. in Gregg v. Georgia, 428 U. S. 153 (1976), specifies aggravating circumstances that may be considered by the jury when appropriate. With respect to the crime of rape, only three such circumstances are specified: (i) the offense was committed by a person with a prior record of conviction for a capital felony; (ii) the offense was committed while the offender was engaged in another capital felony or in aggravated battery; and (iii) the offense was “outrageously or wantonly vile, horrible or inhuman in that it involved torture, depravity of mind, or an aggravated battery to the victim.” Ante, at 588-589, n. 3. Only the third circumstance describes in general the offense of aggravated rape, often identified as a separate and more heinous offense than rape. See, e. g., ALI, Model Penal Code § 207.4, Comment, p. 246 (Tent. Draft No. 4, 1955); ALI, Model Penal Code § 213.1 (Prop. Off. Draft, 1962); Nev. Rev. Stat. § 200.363 (1975). That third circumstance was not sub*602mitted to the jury in this case, as the evidence would not have supported such a finding. It is therefore quite unnecessary for the plurality to write in terms so sweeping as to foreclose each of the 50 state legislatures from creating a narrowly defined substantive crime of aggravated rape punishable by death.1

In accord with our decisions last Term, the plurality opinion states:

“[T]he death penalty is not invariably cruel and unusual punishment within the meaning of the Eighth Amend*603ment; it is not inherently barbaric or an unacceptable mode of punishment for crime; neither is it always disproportionate to the crime for which it is imposed.” Ante, at 591.

Thus, capital punishment may be imposed on those sentenced in accordance with the procedures identified in Gregg and Woodson v. North Carolina, 428 U. S. 280 (1976), at least when the offender is convicted of murder, the crime involved in all five of last Term’s capital cases.

Today, in a case that does not require such an expansive pronouncement, the plurality draws a bright line between murder and all rapes — regardless of the degree of brutality of the rape or the effect upon the victim. I dissent because I am not persuaded that such a bright line is appropriate. As noted in Snider v. Peyton, 356 F. 2d 626, 627 (CA4 1966), “[tjhere is extreme variation in the degree of culpability of rapists.” The deliberate viciousness of the rapist may be greater than that of the murderer. Rape is never an act committed accidentally. Rarely can it be said to be unpremeditated. There also is wide variation in the effect on the victim. The plurality opinion says that “[ljife is over for the victim of the murderer; for the rape victim, life may not be nearly so happy as it was, but it is not over and normally is not beyond repair.” Ante, at 598. But there is indeed “extreme variation” in the crime of rape. Some victims are so grievously injured physically or psychologically that life is beyond repair.

Thus, it may be that the death penalty is not disproportionate punishment for the crime of aggravated rape. Final resolution of the question must await careful inquiry into objective indicators of society’s “evolving standards of decency,” particularly legislative enactments and the responses of juries in capital cases.2 See Gregg v. Georgia, supra, at 173-182 *604(joint opinion of Stewart, Powell, and Stevens, JJ.); Woodson v. North Carolina, supra, at 294—295 (plurality opinion); Furman v. Georgia, 408 U. S. 238, 436-443 (1972) (Powell, J., dissenting). The plurality properly examines these indicia, which do support the conclusion that society finds the death penalty unacceptable for the crime of rape in the absence of excessive brutality or severe injury. But it has not been shown that society finds the penalty disproportionate for all rapes. In a proper case a more discriminating inquiry than the plurality undertakes well might discover that both juries and legislatures have reserved the ultimate penalty for the case of an outrageous rape resulting in serious, lasting harm to the victim. I would not prejudge the issue. To this extent, I respectfully dissent.

Mr. Chief Justice Burger,

with whom Mr. Justice Rehnquist joins, dissenting.

In a case such as this, confusion often arises as to the Court’s proper role in reaching a decision. Our task is not to give effect to our individual views on capital punishment; rather, we must determine what the Constitution permits a State to do under its reserved powers. In striking down the death penalty imposed upon the petitioner in this case, the Court has overstepped the bounds of proper constitutional adjudication by substituting its policy judgment for that of the state legislature. I accept that the Eighth Amendment’s concept of disproportionality bars the death penalty for minor crimes. But rape is not a minor crime; hence the Cruel and Unusual Punishments Clause does not give the Members of this Court license to engraft their conceptions of proper public policy onto the considered legislative judgments of the States. Since I cannot agree that Georgia lacked the consti*605tutional power to impose the penalty of death for rape, I dissent from the Court’s judgment.

On December 5, 1971, the petitioner, Ehrlich Anthony Coker, raped and then stabbed to death a young woman. Less than eight months later Coker kidnaped and raped a second young woman. After twice raping this 16-year-old victim, he stripped her, severely beat her with a club, and dragged her into a wooded area where he left her for dead. He was apprehended and pleaded guilty to offenses stemming from these incidents. He was sentenced by three separate courts to three life terms, two 20-year terms, and one 8-year term of imprisonment.1 Each judgment specified that the sentences it imposed were to run consecutively rather than concurrently. Approximately 1% years later, on September 2, 1974, petitioner escaped from the state prison where he was serving these sentences. He promptly raped another 16-year-old woman in the presence of her husband, abducted her from her home, and threatened her with death and serious bodily harm. It is this crime for which the sentence now under review was imposed.

The Court today holds that the State of Georgia may not impose the death penalty on Coker. In so doing, it prevents the State from imposing any effective punishment upon Coker for his latest rape. The Court’s holding, moreover, bars Georgia from guaranteeing its citizens that they *606will suffer no further attacks by this habitual rapist. In fact, given the lengthy sentences Coker must serve for the crimes he has already committed, the Court's holding assures that petitioner- — as well as others in his position- — will henceforth feel no compunction whatsoever about committing further rapes as frequently as he may be able to escape from confinement and indeed even within the walls of the prison itself. To what extent we have left States “elbowroom” to protect innocent persons from depraved human beings like Coker remains in doubt.

(2)

My first disagreement with the Court’s holding is its unnecessary breadth. The narrow issue here presented is whether the State of Georgia may constitutionally execute this petitioner for the particular rape which he has committed, in light of all the facts and circumstances shown by this record. The plurality opinion goes to great lengths to consider societal mores and attitudes toward the generic crime of rape and the punishment for it; however, the opinion gives little attention to the special circumstances which bear directly on whether imposition of the death penalty is an appropriate societal response to Coker’s criminal acts: (a) On account of his prior offenses, Coker is already serving such lengthy prison sentences that imposition of additional periods of imprisonment would have no- incremental punitive effect; (b) by his life pattern Coker has shown that he presents a particular danger .to the safety, welfare, and chastity of women, and on his record the likelihood is therefore great that he will repeat his crime at the first opportunity; (c) petitioner escaped from prison, only a year and a half after he commenced serving his latest sentences; he has nothing to lose by further escape attempts; and (d) should he again succeed in escaping from prison, it is reasonably predictable that he will repeat his pattern of attacks on *607women — and with impunity since the threat of added prison sentences will be no deterrent.

Unlike the plurality, I would narrow the inquiry in this case to the question actually presented: Does the Eighth Amendment’s ban against cruel and unusual punishment prohibit the State of Georgia from executing a person who has, within the space of three years, raped three separate women, killing one and attempting to kill another, who is serving prison terms exceeding his probable lifetime and who has not hesitated to escape confinement at the first available opportunity? Whatever one’s view may be as to the State’s constitutional power to impose the death penalty upon a rapist who stands before a court convicted for the first time, this case reveals a chronic rapist whose continuing danger to the community is abundantly clear.

Mr. Justice Powell would hold the death sentence inappropriate in this case because “there is no indication that petitioner’s offense was committed with excessive brutality or that the victim sustained serious or lasting injury.” Ante, at 601.2 Apart from the reality that rape is inherently one *608of the most egregiously brutal acts one human being can inflict upon another, there is nothing in the Eighth Amendment that so narrowly limits the factors which may be considered by a state legislature in determining whether a particular punishment is grossly excessive. Surely recidivism, especially the repeated commission of heinous crimes, is a factor which may properly be weighed as an aggravating circumstance, permitting the imposition of a punishment more severe than for one isolated offense. For example, as a matter of national policy, Congress has expressed its will that a person who has committed two felonies will suffer enhanced punishment for a third one, 18 U. S. C. § 3575 (e) (1); Congress has also declared that a second conviction for assault on a mail carrier may be punished more seriously than a first such conviction, 18 U. S. C. § 2114. Many States *609provide an increased penalty for habitual criminality. See, e. g., Wis. Stat. Ann. §939.62 (1958); see also Annot., 58 A. L. R. 20 (1929); 82 A. L. R. 345 (1933); 79 A. L. R. 2d 826 (1961).3 As a factual matter, the plurality opinion is correct in stating that Coker’s “prior convictions do not change the fact that the instant crime being punished is a rape not involving the taking of life,” ante, at 599; however, it cannot be disputed that the existence of these prior convictions makes Coker a substantially more serious menace to society than a first-time offender: 4

“There is a widely held view that those who present the strongest case for severe measures of incapacitation are not murderers as a group (their offenses often are situational) but rather those who have repeatedly engaged in violent, combative behavior. A well-demonstrated *610propensity for life-endangering behavior is thought to provide a more solid basis for infliction of the most severe measures of incapacitation than does the fortuity of a single homicidal incident.” Packer, Making the Punishment Fit the Crime, 77 Harv. L. Rev. 1071, 1080 (1964). (Emphasis added.)

In my view, the Eighth Amendment does not prevent the State from taking an individual’s “well-demonstrated propensity for life-endangering behavior” into account in devising punitive measures which will prevent inflicting further harm upon innocent victims. See Gregg v. Georgia, 428 U. S. 153, 183 n. 28 (1976). Only one year ago Mr. Justice White succinctly noted: “[D]eath finally forecloses the possibility that a prisoner will commit further crimes, whereas life imprisonment does not.” Roberts v. Louisiana, 428 U. S. 325, 354 (1976) (dissenting opinion); see also Furman v. Georgia, 408 U. S., at 311 (White, J., concurring).

In sum, once the Court has held that “the punishment of death does not invariably violate the Constitution,” Gregg v. Georgia, supra, at 169, it seriously impinges upon the State’s legislative judgment to hold that it may not impose such sentence upon an individual who has shown total and repeated disregard for the welfare, safety, personal integrity, and human worth of others, and who seemingly cannot be deterred from continuing such conduct.5 I therefore would *611hold that the death sentence here imposed is within the power reserved to the State and leave for another day the question of whether such sanction would be proper under other circumstances. The dangers which inhere whenever the Court casts its constitutional decisons in terms sweeping beyond the facts of the case presented, are magnified in the context of the Eighth Amendment. In Furman v. Georgia, supra, at 431, Me. Justice Powell, in dissent, stated:

“[W]here, as here, the language of the applicable [constitutional] provision provides great leeway and where the underlying social policies are felt to be of vital importance, the temptation to read personal preference into the Constitution is understandably great. It is too easy to propound our subjective standards of wise policy under the rubric of more or less universally held standards of decency.” (Emphasis added.)

Since the Court now invalidates the death penalty as a sanction for all rapes of adults at all times under all circumstances,6 I reluctantly turn to what I see as the broader issues raised by this holding.

(3)

The plurality, ante, at 597-598, acknowledges the gross nature of the crime of rape. A rapist not only violates a victim’s privacy and personal integrity, but inevitably causes serious psychological as well as physical harm in the process. The long-range effect upon the victim’s life and health is likely *612to be irreparable; it is impossible to measure the harm which results. Volumes have been written by victims, physicians, and psychiatric specialists on the lasting injury suffered by rape victims. Rape is not a mere physical attack — it is destructive of the human personality. The remainder of the victim’s life may be gravely affected, and this in turn may have a serious detrimental effect upon her husband and any children she may have. I therefore wholly agree with Mr. Justice White’s conclusion as far as it goes — that “[s]hort of homicide, [rape] is the 'ultimate violation of self.’ ” Ante, at 597. Victims may recover from the physical damage of knife or bullet wounds, or a beating with fists or a club, but recovery from such a gross assault on the. human personality is not healed by medicine or surgery. To speak blandly, as the plurality does, of rape victims who are “unharmed,” or to classify the human outrage of rape, as dees Mr. Justice Powell, in terms of “excessively brutal,” ante, at 601, versus “moderately brutal,” takes too little account of the profound suffering the crime imposes upon the victims and their loved ones.

Despite its strong condemnation of rape, the Court reaches the inexplicable conclusion that “the death penalty ... is an excessive penalty” for the perpetrator of this heinous offense.7 This, the Court holds, is true even though in Georgia the death penalty may be imposed only where the rape is coupled with one or more aggravating circumstances. The process by which this conclusion is reached is as startling as it is disquieting. It represents a clear departure from precedent by making this Court “under the aegis of the Cruel and Unusual Punishments Clause, the ultimate arbiter of the standards of criminal responsibility in diverse areas of the *613criminal law, throughout the country.” Powell v. Texas, 392 U. S. 514, 533 (1968) (opinion of Marshall, J.).8 This seriously strains and distorts our federal system, removing much of the flexibility from which it has drawn strength for two centuries.

The analysis of the plurality opinion is divided into two parts: (a) an “objective” determination that most American jurisdictions do not presently make rape a capital offense, and (b) a subjective judgment that death is an excessive punishment for rape because the crime does not, in and of itself, cause the death' of the victim. I take issue with each of these points.

(a)

The plurality opinion bases its analysis, in part, on the fact that “Georgia is the sole jurisdiction in the United States at the present time that authorizes a sentence of death when the rape victim is an adult woman.” Ante, at 595-596. Surely, however, this statistic cannot be deemed determinative, or even particularly relevant. As the opinion concedes, ante, at 594, two other States — Louisiana and North Carolina — have enacted death penalty statutes for adult rape since this Court’s 1972 decision in Furman v. Georgia, 408 U. S. 238. If the Court is to rely on some “public opinion” process, does this not suggest the beginning of a “trend”?

*614More to the point, however, it is myopic to base sweeping constitutional principles upon the narrow experience of the past five years. Considerable uncertainty was introduced into this area of the law by this Court’s Furman decision. A large number of States found their death penalty statutes invalidated; legislatures were left in serious doubt by the expressions vacillating between discretionary and mandatory death penalties, as to whether this Court would sustain any statute imposing death as a criminal sanction.9 Failure of more States to enact statutes imposing death for rape of an adult woman may thus reflect hasty legislative compromise occasioned by time pressures following Furman, a desire to wait on the experience of those States which did enact such statutes, or simply an accurate forecast of today’s holding.

In any case, when considered in light of the experience since the turn of this century, where more than one-third of American jurisdictions have consistently provided the death penalty for rape, the plurality’s focus on the experience of the immediate past must be viewed as truly disingenuous. Having in mind the swift changes in positions of some Members of this Court in the short span of five years, can it rationally be considered a relevant indicator of what our society deems “cruel and unusual” to look solely to what legislatures have refrained from doing under conditions of great uncertainty arising from our less than lucid holdings on the Eighth Amendment? Far more representative of societal mores of the 20th century is the accepted *615practice in a substantial number of jurisdictions preceding the Furman decision. “[The] problem ... is the suddenness of the Court’s perception of progress in the human attitude since decisions of only a short while ago.” Furman v. Georgia, supra, at 410 (Blackmun, J., dissenting). Cf. Rudolph v. Alabama, 375 U. S. 889 (1963).

However, even were one to give the most charitable acceptance to the plurality’s statistical analysis, it still does not, to my mind, support its conclusion. The most that can be claimed is that for the past year Georgia has been the only State whose adult rape death penalty statute has not otherwise been invalidated; two other state legislatures had enacted rape death penalty statutes in the last five years, but these were invalidated for reasons unrelated to rape under the Court’s decisions last Term. Woodson v. North Carolina, 428 U. S. 280 (1976); Roberts v. Louisiana, 428 U. S. 325 (1976). Even if these figures could be read as indicating that no other States view the death penalty as an appropriate punishment for the rape of an adult woman, it would not necessarily follow that Georgia’s imposition of such sanction violates the Eighth Amendment.

The Court has repeatedly pointed to the reserve strength of our federal system which allows state legislatures, within broad limits, to experiment with laws, both criminal and civil, in the effort to achieve socially desirable results. See, e. g., Whalen v. Roe, 429 U. S. 589, 597-598, and n. 22 (1977) ; Johnson v. Louisiana, 406 U. S. 356, 376 (1972) (opinion of Powell, J.); California v. Green, 399 U. S. 149, 184-185 (1970) (Harlan, J., concurring); Fay v. New York, 332 U. S. 261, 296 (1947). Various provisions of the Constitution, including the Eighth Amendment and the Due Process Clause, of course place substantive limitations on the type of experimentation a State may undertake. However, as the plurality admits, the crime of rape is second perhaps only to murder in its gravity. It follows then that Georgia did not approach *616such substantive constraints by enacting the statute here in question. See also infra, at 619-622.

Statutory provisions in criminal justice applied in one part of the country can be carefully watched by other state legislatures, so that the experience of one State becomes available to all. Although human lives are in the balance, it must be remembered that failure to allow flexibility may also jeopardize human lives — those of the victims of undeterred criminal conduct. See infra, at 620. Our concern for the accused ought not foreclose legislative judgments showing a modicum of consideration for the potential victims.

Three state legislatures have, in the past five years, determined that the taking of human fife and the devastating consequences of rape will be minimized if rapists may, in a limited class of cases, be executed for their offenses.10 That these States are presently a minority does not, in my view, make their judgment less worthy of deference. Our concern for human fife must not be confined to the guilty; a state legislature is not to be thought insensitive to human values because it acts firmly to protect the lives and related values of the innocent. In this area, the choices for legislatures are at best painful and difficult and deserve a high degree of deference. Only last Term Me. Justice White observed:

“It will not do to denigrate these legislative judgments as some form of vestigial savagery or as purely retributive in motivation; for they are solemn judgments, reasonably based, that imposition of the death penalty will save the lives of innocent persons. This concern for life and human values and the sincere efforts of the States to pursue them are matters of the greatest moment with which the judiciary should he most reluc*617tant to interfere.” Roberts v. Louisiana, supra, at 355 (dissenting opinion). (Emphasis added.)

The question of whether the death penalty is an appropriate punishment for rape is surely an open one. It is arguable that many prospective rapists would be deterred by the possibility that they could suffer death for their offense; it is also arguable that the death penalty would have only minimal deterrent effect.11 It may well be that rape victims would become more willing to report the crime and aid in the apprehension of the criminals if they knew that community disapproval of rapists was sufficiently strong to inflict the extreme penalty; or perhaps they would be reluctant to cooperate in the prosecution of rapists if they knew that a conviction might result in the imposition of the death penalty. Quite possibly, the occasional, well-publicized execution of egregious rapists may cause citizens to feel greater security in their daily lives;12 or, on the contrary, it may be that members of a civilized community will suffer the pangs of a heavy conscience because such punishment will be perceived as excessive.13 We cannot know which among *618this range of possibilities is correct, but today’s holding forecloses the very exploration we have said federalism was intended to foster. It is difficult to believe that Georgia would long remain alone in punishing rape by death if the next decade demonstrated a drastic reduction in its incidence of rape, an increased cooperation by rape victims in the apprehension and prosecution of rapists, and a greater confidence in the rule of law on the part of the populace.

In order for Georgia’s legislative program to develop it must be given time to take effect so that data may be evaluated for' comparison with the experience of States which have not enacted death penalty statutes. Today, the Court repudiates the State’s solemn judgment on how best to deal with the crime of rape before anyone can know whether the death penalty is an effective deterrent for one of the most horrible of all crimes. And this is done a few short years after Mr. Justice Powell’s excellent statement:

“In a period in our country’s history when the frequency of [rape] is increasing alarmingly, it is indeed a grave event for the Court to take from the States whatever deterrent and retributive weight the death penalty retains.” Furman v. Georgia, 408 U. S., at 459 (dissenting opinion) (footnote omitted).

To deprive States of this authority as the Court does, on the basis that “[t]he current judgment with respect to the death penalty for rape . . . weighs very heavily on the side of rejecting capital punishment as a suitable penalty for raping an adult woman,” ante, at 596, is impermissibly rash. The current judgment of some Members of this Court has undergone significant change in the short time since Furman,14 Social change on great issues generally reveals itself in small increments, and the “current judgment” of many States could *619well be altered on the basis of Georgia's experience, were we to allow its statute to stand.15

(b)

The subjective judgment that the death penalty is simply disproportionate to the crime of rape is even more disturbing than the “objective” analysis discussed supra. The plurality’s conclusion on this point is based upon the bare fact that murder necessarily results in the physical death of the victim, while rape does not. Ante, at 598-599, 600. However, no Member of the Court explains why this distinction has relevance, much less constitutional significance. It is, after all, not irrational — nor constitutionally impermissible — for a legislature to make the penalty more severe than the criminal act it punishes16 in the hope it would deter wrongdoing:

“We may not require the legislature to select the least severe penalty possible so long as the penalty selected is not cruelly inhumane or disproportionate to the crime involved.” Gregg v. Georgia, 428 U. S., at 175.

Accord, Furman v. Georgia, supra, at 451 (Powell, J., dissenting).

It begs the question to state, as does the plurality opinion:

“Life is over for the victim of the murderer; for the rape victim, life may not be nearly so happy as it was, but it is not over and normally is not beyond repair.” Ante, at 598.

*620Until now, the issue under the Eighth Amendment has not been the state of any particular victim after the crime, but rather whether the punishment imposed is grossly disproportionate to the evil committed by the perpetrator. See Gregg v. Georgia, supra, at 173, Furman v. Georgia, supra, at 458 (Powell, J., dissenting). As a matter of constitutional principle, that test cannot have the primitive simplicity of “life for life, eye for eye, tooth for tooth.” Rather States must be permitted to engage in a more sophisticated weighing of values in dealing with criminal activity which consistently poses serious danger of death or grave bodily harm. If innocent life and limb are to be preserved I see no constitutional barrier in punishing by death all who engage in such activity, regardless of whether the risk comes to fruition in any particular instance. See Packer, 77 Harv. L. Rev., at 1077-1079.

Only one year ago the Court held it constitutionally permissible to impose the death penalty for the crime of murder, provided that certain procedural safeguards are followed. Compare Gregg v. Georgia, supra; Proffitt v. Florida, 428 U. S. 242 (1976), and Jurek v. Texas, 428 U. S. 262 (1976), with Roberts v. Louisiana, 428 U. S. 325 (1976), and Woodson v. North Carolina, 428 U. S. 280 (1976). Today, the plurality readily admits that “[s]hort of homicide, [rape] is the ‘ultimate violation of self.’ ” Ante, at 597. Moreover, as stated by Mr. Justicie Powell:

“The threat of serious injury is implicit in the definition of rape; the victim is either forced into submission by physical violence or by the threat of violence.” Furman v. Georgia, supra, at 460 (dissenting opinion).

Rape thus is not a crime “light years” removed from murder in the degree of its heinousness; it certainly poses a serious potential danger to the life and safety of innocent victims— apart from the devastating psychic consequences. It would *621seem to follow therefore that, affording the States proper leeway under the broad standard of the Eighth Amendment,17 if murder is properly punishable by death, rape should be also, if that is the considered judgment of the legislators.

The Court’s conclusion to the contrary is very disturbing indeed. The clear implication of today’s holding appears to be that the death penalty may be properly imposed only as to crimes resulting in death of the victim. This casts serious doubt upon the constitutional validity of statutes imposing the death penalty for a variety of conduct which, though dangerous, may not necessarily result in any immediate death, e. g., treason, airplane hijacking, and kid-naping. In that respect, today’s holding does even more harm than is initially apparent. We cannot avoid taking judicial notice that crimes such as airplane hijacking, kidnaping, and mass terrorist activity constitute a serious and increasing danger to the safety of the public. It would be unfortunate indeed if the effect of today’s holding were to inhibit States and the Federal Government from experimenting with various remedies — including possibly imposition of the penalty of death — to prevent and deter such crimes.

*622Some sound observations, made only a few years ago, deserve repetition:

“Our task here, as must so frequently be emphasized and re-emphasized, is to pass upon the constitutionality of legislation that has been enacted and that is challenged. This is the sole task for judges. We should not allow our personal preferences as to the wisdom of legislative and congressional action, or our distaste for such action, to guide our judicial decision in cases such as these. The temptations to cross that policy line are very great. In fact, as today’s decision reveals, they are almost irresistible.” Furman v. Georgia, 408 U. S., at 411 (Blackmun, J., dissenting).

Whatever our individual views as to the wisdom of capital punishment, I cannot agree that it is constitutionally impermissible for a state legislature to make the “solemn judgment” to impose such penalty for the crime of rape. Accordingly, I would leave to the States the task of legislating in this area of the law.

Justice White

delivered the opinion of the Court.

I

The facts of this case, taken principally from the opinion of the Florida Supreme Court, are as follows. . On April 1, *7841975, at approximately 7:45 a. m., Thomas and Eunice Kersey, aged 86 and 74, were robbed and fatally shot at their farmhouse in central Florida. The evidence showed that Sampson and Jeanette Armstrong had gone to the back door of the Kersey house and asked for water for an overheated car. When Mr. Kersey came out of the house, Sampson Armstrong grabbed him, pointed a gun at him, and told Jeanette Armstrong to take his money. Mr. Kersey cried for help, and his wife came out of the house with a gun and shot Jeanette Armstrong, wounding her. Sampson Armstrong, and perhaps Jeanette Armstrong, then shot and killed both of the Kerseys, dragged them into the kitchen, and took their money and fled.

Two witnesses testified that they drove past the Kersey house between 7:30 and 7:40 a. m. and saw a large cream- or yellow-colored car parked beside the road about 200 yards from the house and that a man was sitting in the car. Another witness testified that at approximately 6:45 a. m. he saw Ida Jean Shaw, petitioner’s common-law wife and Jeanette Armstrong’s mother, driving a yellow Buick with a vinyl top which belonged to her and petitioner Earl Enmund. Enmund was a passenger in the car along with an unidentified woman. At about 8 a. m. the same witness saw the car return at a high rate of speed. Enmund was driving, Ida Jean Shaw was in the front seat, and one of the other two people in the car was lying down across the back seat.

Enmund, Sampson Armstrong, and Jeanette Armstrong were indicted for the first-degree murder and robbery of the Kerseys. Enmund and Sampson Armstrong were tried together.1 The prosecutor maintained in his closing argument that “Sampson Armstrong killed the old people.” Record 1577. The judge instructed the jury that “[t]he killing of a *785human being while engaged in the perpetration of or in the attempt to perpetrate the offense of robbery is murder in the first degree even though there is no premeditated design or intent to kill.” App. 6. He went on to instruct them that

“[i]n order to sustain a conviction of first degree murder while engaging in the perpetration of or in the attempted perpetration of the crime of robbery, the evidence must establish beyond a reasonable doubt that the defendant was actually present and was actively aiding and abetting the robbery or attempted robbery, and that the unlawful killing occurred in the perpetration of or in the attempted perpetration of the robbery.” Id., at 9.

The jury found both Enmund and Sampson Armstrong guilty of two counts of first-degree murder and one count of robbery. A separate sentencing hearing was held and the jury recommended the death penalty for both defendants under the Florida procedure whereby the jury advises the trial judge whether to impose the death penalty. See Fla. Stat. §921.141(2) (1981). The trial judge then sentenced Enmund to death on the two counts of first-degree murder. Enmund appealed, and the Florida Supreme Court remanded for written findings as required by Fla. Stat. §921.141(3) (1981). The trial judge found four statutory aggravating circumstances: the capital felony was committed while Enmund was engaged in or was an accomplice in the commission of an armed robbery, Fla. Stat. §921.141(5)(d) (1981); the capital felony was committed for pecuniary gain, §921.141(5)(f); it was especially heinous, atrocious, or cruel, § 921.141(5)(h); and Enmund was previously convicted of a felony involving the use or threat of violence, §921.141(5)(b). 399 So. 2d 1362, 1371-1372 (Fla. 1981). The court found that “none of the statutory mitigating circumstances applied” to Enmund and that the aggravating circumstances outweighed the mitigating circumstances. Id., at 1372. Enmund was therefore sentenced to death on each of the murder counts.

*786The Florida Supreme Court affirmed Enmund’s conviction and sentences. It found that “[t]here was no direct evidence at trial that Earl Enmund was present at the back door of the Kersey home when the plan to rob the elderly couple led to their being murdered.” Id., at 1370. However, it rejected petitioner’s argument that at most he could be found guilty of second-degree murder under Florida’s felony-murder rule. The court explained that the interaction of the “ ‘felony murder rule and the law of principals combine to make a felon generally responsible for the lethal acts of his co-felon.’” Id., at 1369, quoting Adams v. State, 341 So. 2d 765, 768-769 (Fla. 1976), cert. denied, 434 U. S. 878 (1977). Although petitioner could be convicted of second-degree murder only if he were an accessory before the fact rather than a principal, the Florida Supreme Court reasoned:

“[T]he only evidence of the degree of his participation is the jury’s likely inference that he was the person in the car by the side of the road near the scene of the crimes. The jury could have concluded that he was there, a few hundred feet away, waiting to help the robbers escape with the Kerseys’ money. The evidence, therefore, was sufficient to find that the appellant was a principal of the second degree, constructively present aiding and abetting the commission of the crime of robbery. This conclusion supports the verdicts of murder in the first degree on the basis of the felony murder portion of section 782.04(1)(a).” 399 So. 2d, at 1370.2

*787The State Supreme Court rejected two of the four statutory aggravating circumstances found by the trial court. It held that the findings that the murders were committed in the course of a robbery and that they were committed for pecuniary gain referred to the same aspect of petitioner’s crime and must be treated as only one aggravating circumstance. Id., at 1373. In addition, the court held that “[t]he recited circumstance, that the murders were especially heinous, atrocious, and cruel, cannot be approved.” Ibid., citing Armstrong v. State, 399 So. 2d 953 (Fla. 1981).3 However, because there were two aggravating circumstances and no mitigating circumstances, the death sentence was affirmed. In so doing, the court expressly rejected Enmund’s submission that because the evidence did not establish that he intended to take life, the death penalty was barred by the Eighth Amendment of the United States Constitution. 399 So. 2d, at 1371.

We granted Enmund’s petition for certiorari, 454 U. S. 939 (1981), presenting the question whether death is a valid penalty under the Eighth and Fourteenth Amendments for one who neither took life, attempted to take life, nor intended to take life.4

*788II

As recounted above, the Florida Supreme Court held that the record supported no more than the inference that Enmund was the person in the car by the side of the road at the time of the killings, waiting to help the robbers escape. This was enough under Florida law to make Enmund a constructive aider and abettor and hence a principal in first-degree murder upon whom the death penalty could be imposed. It was thus irrelevant to Enmund’s challenge to the death sentence that he did not himself kill and was not present at the killings; also beside the point was whether he intended that the Kerseys be killed or anticipated that lethal force would or might be used if necessary to effectuate the robbery or a safe escape. We have concluded that imposition of the death penalty in these circumstances is inconsistent with the Eighth and Fourteenth Amendments.

A

The Cruel and Unusual Punishments Clause of the Eighth Amendment is directed, in part, “‘against all punishments which by their excessive length or severity are greatly disproportioned to the offenses charged.’” Weems v. United States, 217 U. S. 349, 371 (1910), quoting O’Neil v. Vermont, 144 U. S. 323, 339-340 (1892) (Field, J., dissenting). This Court most recently held a punishment excessive in relation to the crime charged in Coker v. Georgia, 433 U. S. 584 (1977). There the plurality opinion concluded that the imposition of the death penalty for the rape of an adult woman “is grossly disproportionate and excessive punishment for the crime of rape and is therefore forbidden by the Eighth Amendment as cruel and unusual punishment.” Id., at 592. In reaching this conclusion, it was stressed that our judgment “should be informed by objective factors to the maximum possible extent.” Ibid. Accordingly, the Court looked to the historical development of the punishment at issue, legislative judgments, international opinion, and the sentencing decisions juries have made before bringing its *789own judgment to bear on the matter. We proceed to analyze the punishment at issue in this case in a similar manner.

B

The Coker plurality observed that “[a]t no time in the last 50 years have a majority of the States authorized death as a punishment for rape.” Id., at 593. More importantly, in reenacting death penalty laws in order to satisfy the criteria established in Furman v. Georgia, 408 U. S. 238 (1972), only three States provided the death penalty for the rape of an adult woman in their revised statutes. 433 U. S., at 594. The plurality therefore concluded that “[t]he current judgment with respect to the death penalty for rape is not wholly unanimous among state legislatures, but it obviously weighs very heavily on the side of rejecting capital punishment as a suitable penalty for raping an adult woman.” Id., at 596 (footnote omitted).

Thirty-six state and federal jurisdictions presently authorize the death penalty. Of these, only eight jurisdictions authorize imposition of the death penalty solely for participation in a robbery in which another robber takes life.5 Of the remaining.28 jurisdictions, in 4 felony murder is not a capital crime.6 Eleven States require some culpable mental state *790with respect to the homicide as a prerequisite to conviction of a crime for which the death penalty is authorized. Of these 11 States, 8 make knowing, intentional, purposeful, or premeditated killing an element of capital murder.7 Three other States require proof of a culpable mental state short of intent, such as recklessness or extreme indifference to human life, before the death penalty may be imposed.8 In these 11 States, therefore, the actors in a felony murder are not subject to the death penalty without proof of their mental state, proof which was not required with respect to Enmund *791either under the trial court’s instructions or under the law announced by the Florida Supreme Court.

Four additional jurisdictions do not permit a defendant such as Enmund to be put to death. Of these, one State flatly prohibits capital punishment in cases where the defendant did not actually commit murder.9 Two jurisdictions preclude the death penalty in cases such as this one where the defendant “was a principal in the offense, which was committed by another, but his participation was relatively minor, although not so minor as to constitute a defense to prosecution.”10 One other State limits the death penalty in felony murders to narrow circumstances not involved here.11

Nine of the remaining States deal with the imposition of the death penalty for a vicarious felony murder in their capital sentencing statutes. In each of these States, a defendant may not be executed solely for participating in a felony in which a person was killed if the defendant did not actually cause the victim’s death. For a defendant to be executed in these States, typically the statutory aggravating circumstances which are present must outweigh mitigating factors. To be sure, a vicarious felony murderer may be sentenced to death in these jurisdictions absent an intent to kill if sufficient aggravating circumstances are present. However, six *792of these nine States make it a statutory mitigating circumstance that the defendant was an accomplice in a capital felony committed by another person and his participation was relatively minor.12 By making minimal participation in a capital felony committed by another person a mitigating circumstance, these sentencing statutes reduce the likelihood that a person will be executed for vicarious felony murder. The remaining three jurisdictions exclude felony murder from their lists of aggravating circumstances that will support a death sentence.13 In each of these nine States, a nontriggerman guilty of felony murder cannot be sentenced to death for the felony murder absent aggravating circumstances above and beyond the felony murder itself.

Thus only a small minority of jurisdictions — eight—allow the death penalty to be imposed solely because the defendant somehow participated in a robbery in the course of which a murder was committed. Even if the nine States are included where such a defendant could be executed for an unintended felony murder if sufficient aggravating circumstances are present to outweigh mitigating circumstances — which often include the defendant’s minimal participation in the murder — only about a third of American jurisdictions would ever permit a defendant who somehow participated in a robbery where a murder occurred to be sentenced to die. Moreover, of the eight States which have enacted new death penalty statutes since 1978, none authorize capital punishment in such circumstances.14 While the current legislative judg*793ment with respect to imposition of the death penalty where a defendant did not take life, attempt to take it, or intend to take life is neither “wholly unanimous among state legislatures,” Coker v. Georgia, 433 U. S., at 596, nor as compelling as the legislative judgments considered in Coker, it nevertheless weighs on the side of rejecting capital punishment for the crime at issue.15

*794c

Society’s rejection of the death penalty for accomplice liability in felony murders is also indicated by the sentencing decisions that juries have made. As we have previously observed, “‘[t]he jury ... is a significant and reliable objective index of contemporary values because it is so directly involved.’” Coker v. Georgia, supra, at 596, quoting Gregg v. Georgia, 428 U. S. 153, 181 (1976). The evidence is overwhelming that American juries have repudiated imposition of the death penalty for crimes such as petitioner’s. First, according to the petitioner, a search of all reported appellate court decisions since 1954 in cases where a defendant was executed for homicide shows that of the 362 executions, in 339 the person executed personally committed a homicidal assault.16 In 2 cases the person executed had another person commit the homicide for him, and in 16 cases the facts were not reported in sufficient detail to determine whether the person executed committed the homicide.17 The survey revealed only 6 cases out of 362 where a nontriggerman felony murderer was executed. All six executions took place in *7951955. By contrast, there were 72 executions for rape in this country between 1955 and this Court’s decision in Coker v. Georgia in 1977.18

That juries have rejected the death penalty in cases such as this one where the defendant did not commit the homicide, was not present when the killing took place, and did not participate in a plot or scheme to murder is also shown by petitioner’s survey of the Nation’s death-row population.19 As of October 1, 1981, there were 796 inmates under sentences of death for homicide. Of the 739 for whom sufficient data are available, only 41 did not participate in the fatal assault on the victim. Of the 40 among the 41 for whom sufficient information was available, only 16 were not physically present when the fatal assault was committed. These 16 prisoners included only 3, including petitioner, who were sentenced to die absent a finding that they hired or solicited someone else to kill the victim or participated in a scheme designed to kill the victim. The figures for Florida are similar.20 Forty-five felony murderers are currently on death row. The Florida Supreme Court either found or affirmed a trial court or jury finding that the defendant intended life to be taken in 36 cases. In eight cases the courts made no finding with respect to intent, but the defendant was the triggerman in each case. In only one case — Enmund’s—there was no finding of an intent to kill and the defendant was not the triggerman.21 *796The State does not challenge this analysis of the Florida cases.

The dissent criticizes these statistics on the ground that they do not reveal the percentage of homicides that were charged as felony murders or the percentage of cases where the State sought the death penalty for an accomplice guilty of felony murder. Post, at 818-819. We doubt whether it is possible to gather such information, and at any rate, it would be relevant if prosecutors rarely sought the death penalty for accomplice felony murder, for it would tend to indicate that prosecutors, who represent society’s interest in punishing crime, consider the death penalty excessive for accomplice felony murder. The fact remains that we are not aware of a single person convicted of felony murder over the past quarter century who did not kill or attempt to kill, and did not intend the death of the victim, who has been executed, and that only three persons in that category are presently sentenced to die. Nor can these figures be discounted by attributing to petitioner the argument that “death is an unconstitutional penalty absent an intent to kill,” post, at 819, and observing that the statistics are incomplete with respect to intent. Petitioner’s argument is that because he did not kill, attempt to kill, and he did not intend to kill, the death penalty is disproportionate as applied to him, and the statistics he cites are adequately tailored to demonstrate that juries— and perhaps prosecutors as well — consider death a disproportionate penalty for those who fall within his category.22

*797III

Although the judgments of legislatures, juries, and prosecutors weigh heavily in the balance, it is for us ultimately to judge whether the Eighth Amendment permits imposition of the death penalty on one such as Enmund who aids and abets a felony in the course of which a murder is committed by others but who does not himself kill, attempt to kill, or intend that a killing take place or that lethal force will be employed. We have concluded, along with most legislatures and juries, that it does not.

We have no doubt that robbery is a serious crime deserving serious punishment. It is not, however, a crime “so grievous an affront to humanity that the only adequate response may be the penalty of death.” Gregg v. Georgia, 428 U. S., at 184 (footnote omitted). “[I]t does not compare with murder, which does involve the unjustified taking of human life. Although it may be accompanied by another crime, [robbery] by definition does not include the death of or even the serious injury to another person. The murderer kills; the [robber], if no more than that, does not. Life is over for the victim of the murderer; for the [robbery] victim, life ... is not over and normally is not beyond repair.” Coker v. Georgia, 433 U. S., at 598 (footnote omitted). As was said of the crime of rape in Coker, we have the abiding conviction that the death penalty, which is “unique in its severity and irrevocability,” Gregg v. Georgia, supra, at 187, is an excessive penalty for the robber who, as such, does not take human life.

*798Here the robbers did commit murder; but they were subjected to the death penalty only because they killed as well as robbed. The question before us is not the disproportionality of death as a penalty for murder, but rather the validity of capital punishment for Enmund’s own conduct. The focus must be on Ms culpability, not on that of those who committed the robbery and shot the victims, for we insist on “individualized consideration as a constitutional requirement in imposing the death sentence,” Lockett v. Ohio, 438 U. S. 586, 605 (1978) (footnote omitted), which means that we must focus on “relevant facets of the character and record of the individual offender.” Woodson v. North Carolina, 428 U. S. 280, 304 (1976). Enmund himself did not kill or attempt to kill; and, as construed by the Florida Supreme Court, the record before us does not warrant a finding that Enmund had any intention of participating in or facilitating a murder. Yet under Florida law death was an authorized penalty because Enmund aided and abetted a robbery in the course of which murder was committed. It is fundamental that “causing harm intentionally must be punished more severely than causing the same harm unintentionally.” H. Hart, Punishment and Responsibility 162 (1968). Enmund did not kill or intend to kill and thus his culpability is plainly different from that of the robbers who killed; yet the State treated them alike and attributed to Enmund the culpability of those who killed the Kerseys. This was impermissible under the Eighth Amendment.

In Gregg v. Georgia the opinion announcing the judgment observed that “[t]he death penalty is said to serve two principal social purposes: retribution and deterrence of capital crimes by prospective offenders.” 428 U. S., at 183 (footnote omitted). Unless the death penalty when applied to those in Enmund’s position measurably contributes to one or both of these goals, it “is nothing more than the purposeless and needless imposition of pain and suffering,” and hence an unconstitutional punishment. Coker v. Georgia, supra, at 592. We are quite unconvinced, however, that the threat *799that the death penalty will be imposed for murder will measurably deter one who does not kill and has no intention or purpose that life will be taken. Instead, it seems likely that “capital punishment can serve as a deterrent only when murder is the result of premeditation and deliberation,” Fisher v. United States, 328 U. S. 463, 484 (1946) (Frankfurter, J., dissenting), for if a person does not intend that life be taken or contemplate that lethal force will be employed by others, the possibility that the death penalty will be imposed for vicarious felony murder will not “enter into the cold calculus that precedes the decision to act.” Gregg v. Georgia, supra, at 186 (footnote omitted).

It would be very different if the likelihood of a killing in the course of a robbery were so substantial that one should share the blame for the killing if he somehow participated in the felony. But competent observers have concluded that there is no basis in experience for the notion that death so frequently occurs in the course of a felony for which killing is not an essential ingredient that the death penalty should be considered as a justifiable deterrent to the felony itself. Model Penal Code §210.2, Comment, p. 38, and n. 96. This conclusion was based on three comparisons of robbery statistics, each of which showed that only about one-half of one percent of robberies resulted in homicide.23 The most recent national *800crime statistics strongly support this conclusion.24 In addition to the evidence that killings only rarely occur during robberies is the fact, already noted, that however often death occurs in the course of a felony such as robbery, the death penalty is rarely imposed on one only vicariously guilty of the murder, a fact which further attenuates its possible utility as an effective deterrence.

As for retribution as a justification for executing Enmund, we think this very much depends on the degree of Enmund’s culpability — what Enmund’s intentions, expectations, and actions were. American criminal law has long considered a defendant’s intention — and therefore his moral guilt — to be critical to “the degree of [his] criminal culpability,” Mullaney v. Wilbur, 421 U. S. 684, 698 (1975), and the Court has found criminal penalties to be unconstitutionally excessive in the absence of intentional wrongdoing. In Robinson v. California, 370 U. S. 660, 667 (1962), a statute making narcotics addiction a crime, even though such addiction “is apparently an illness which may be contracted innocently or involuntarily,” was struck down under the Eighth Amendment. Similarly, in Weems v. United States, the Court invalidated a statute making it a crime for a public official to make a false entry in a public record but not requiring the offender to “injur[e] any one by his act or inten[d] to injure any one.” 217 U. S., at 363. The Court employed a similar approach in Godfrey v. Georgia, 446 U. S. 420, 433 (1980), reversing a death sentence based on the existence of an aggravating circumstance because the defendant’s crime did not reflect “a consciousness *801materially more ‘depraved’ than that of any person guilty of murder.”

For purposes of imposing the death penalty, Enmund’s criminal culpability must be limited to his participation in the robbery, and his punishment must be tailored to his personal responsibility and moral guilt. Putting Enmund to death to avenge two killings that he did not commit and had no intention of committing or causing does not measurably contribute to the retributive end of ensuring that the criminal gets his just deserts. This is the judgment of most of the legislatures that have recently addressed the matter, and we have no reason to disagree with that judgment for purposes of construing and applying the Eighth Amendment.

IV

Because the Florida Supreme Court affirmed the death penalty in this case in the absence of proof that Enmund killed or attempted to kill, and regardless of whether Enmund intended or contemplated that life would be taken, we reverse the judgment upholding the death penalty and remand for further proceedings not inconsistent with this opinion.

So ordered.

Justice Brennan,

concurring.

I join the Court’s opinion. However, I adhere to my view that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments. See Gregg v. Georgia, 428 U. S. 153, 227 (1976) (dissenting opinion).

Justice O’Connor,

with whom The Chief Justice, Justice Powell, and Justice Rehnquist join,

dissenting.

Today the Court holds that the Eighth Amendment prohibits a State from executing a convicted felony murderer. I dissent from this holding not only because I believe that it is not supported by the analysis in our previous cases, but also *802because today’s holding interferes with state criteria for assessing legal guilt by recasting intent as a matter of federal constitutional law.

I

The evidence at trial showed that at approximately 7:30 a. m. on April 1, 1975, Sampson and Jeanette Armstrong approached the back door of Thomas and Eunice Kersey’s farmhouse on the pretext of obtaining water for their overheated car.1 When Thomas Kersey retrieved a water jug to help the Armstrongs, Sampson Armstrong grabbed him, held a gun to him, and told Jeanette Armstrong to take his wallet. Hearing her husband’s cries for help, Eunice Kersey came around the side of the house with a gun and shot Jeanette Armstrong. Sampson Armstrong, and perhaps Jeanette Armstrong, returned the fire, killing both of the Kerseys.2 The Armstrongs dragged the bodies into the kitchen, took Thomas Kersey’s money, and fled to a nearby car, where the petitioner, Earl Enmund, was waiting to help the Armstrongs escape. Record 1348-1351.3

Ida Jean Shaw4 testified that on March 31 the petitioner and the two Armstrongs were staying at her house. When she awoke on April 1, the day of the murders, the petitioner, *803Jeanette, and Sampson, as well as Shaw’s 1969 yellow Buick, were gone. Id., at 1185-1186. A little after eight o’clock, either the petitioner or Sampson Armstrong entered the house and told her that Jeanette had been shot. Id., at 1187-1188. After learning that Jeanette had been shot during a robbery, Shaw asked the petitioner “[w]hy he did it.” Enmund answered that he had decided to rob Thomas Kersey after he had seen Kersey’s money a few weeks earlier. Id., at 1205.5 At the same time, Sampson Armstrong volunteered that he had made sure that the Kerseys were dead. Id., at 1207-1208.

Ida Jean Shaw also testified that, pursuant to the petitioner’s and Sampson Armstrong’s instructions, she had disposed of a .22-caliber pistol that she normally kept in her car, as well as a .38-caliber pistol belonging to the Armstrongs. Id., at 1198-1202. The murder weapons were never recovered.6

In his closing argument, the prosecutor did not argue that Earl Enmund had killed the Kerseys. Instead, he maintained that the petitioner had initiated and planned the *804armed robbery, and was in the car during the killings. According to the prosecutor, “Sampson Armstrong killed the old people.” Id., at 1577.7

After deliberating for four hours, the jury found Sampson Armstrong and the petitioner each guilty of two counts of first-degree murder8 and one count of robbery.9 The jury *805then heard evidence pertaining to the appropriate sentence for the two defendants, and recommended the death penalty for each defendant on each of the murder counts.10

In its sentencing findings,11 the trial court found four statutory aggravating circumstances regarding the petitioner’s involvement in the murder: (1) the petitioner previously had been convicted of a felony involving the use of violence (an armed robbery in 1957), Fla. Stat. §921.141(5)(b) (1981); (2) the murders were committed during the course of a robbery, §921.141(5)(d); (3) the murders were committed for pecuniary gain, §921.141(5)(f); and (4) the murders were especially heinous, atrocious, or cruel because the Kerseys had been shot in a prone position in an effort to eliminate them as witnesses, §921.141(5)(h). App. 30-31; 399 So. 2d 1362, 1371-1372 (Fla. 1981).12

*806The trial court also found that “none of the statutory mitigating circumstances applied” to the petitioner. App. 32 (emphasis in original). Most notably, the court concluded that the evidence clearly showed that the petitioner was an accomplice to the capital felony and that his participation had not been “relatively minor,” but had been major in that he “planned the capital felony and actively participated in an attempt to avoid detection by disposing of the murder weapons.” Ibid.; 399 So. 2d, at 1373. See Fla. Stat. § 921.141(6)(d) (1981).13

Considering these factors, the trial court concluded that the “aggravating circumstances of these capital felonies outweigh the mitigating circumstances,” and imposed the death penalty for each count of murder. App. 32; 399 So. 2d, at 1373. The court sentenced the petitioner to life imprisonment for the robbery. App. 28.14

*807On appeal, the Florida Supreme Court affirmed the petitioner’s convictions and sentences.15 In challenging-his convictions for first-degree murder, the petitioner claimed that there was no evidence that he had committed premeditated murder, or that he had been present aiding and abetting the robbery when the Kerseys were shot. He argued that since the jury properly could have concluded only that he was in the car on the highway when the murders were committed, he could be found guilty at most of second-degree murder under the State’s felony-murder rule.16

The court rejected this argument. Quoting from an earlier case, the Florida Supreme Court held:

“ ‘[A]n individual who personally kills another during the perpetration or attempt to perpetrate one of the enumerated felonies is guilty of first degree murder. . . . Moreover, the felon’s liability for first degree murder extends to all of his co-felons who are personally present. As perpetrators of the underlying felony, they are principals in the homicide. In Florida, as in the majority of jurisdictions, the felony murder rule and the law of principals combine to make a felon generally responsible for the lethal acts of his co-felon. Only if the felon is an accessory before the fact and not personally present does liability attach- under the second degree murder provision of the applicable statute in the instant ease.’” 399 So. 2d, at 1369 (quoting Adams v. State, 341 So. 2d 765, 768-769 (Fla. 1976) (footnote omitted), cert. denied, 434 U. S. 878 (1977)).

*808Consequently, the critical issue regarding liability was whether the petitioner’s conduct would make him a principal or merely an accessory before the fact to the underlying robbery. Under Florida law at the time of the murders, “if the accused was present aiding and abetting the commission or attempt of one of the violent felonies listed in the first-degree murder statute, he is equally guilty, with the actual perpetrator of the underlying felony, of first-degree murder.” 399 So. 2d, at 1370. Moreover,

“‘the presence of the aider and abetter need not have been actual, but it is sufficient if he was constructively present, provided the aider, pursuant to a previous understanding, is sufficiently near and so situated as to abet or encourage, or to render assistance to, the actual perpetrator in committing the felonious act or in escaping after its commission.’” Ibid, (quoting Pope v. State, 84 Fla. 428, 446, 94 So. 865, 871 (1922)).

The court noted that there “was no direct evidence at trial that Earl Enmund was present at the back door of the Kersey home when the plan to rob the elderly couple led to their being murdered.” 399 So. 2d, at 1370.17 Instead,

“the only evidence of the degree of his participation is the jury’s likely inference that he was the person in the car by the side of the road near the scene of the crimes. The jury could have concluded that he was there, a few hundred feet away, waiting to help the robbers escape with the Kerseys’ money.” Ibid.

This evidence, the court concluded, was sufficient to find the petitioner to be a principal under state law, “constructively present aiding and abetting the commission of the crime of robbery,” and thus guilty of first-degree murder. Ibid.

*809Turning to the trial court’s written sentencing findings, the State Supreme Court rejected two of the four aggravating circumstances. First, the court held that two of the trial judge’s findings — that the murders were committed both in the course of robbery and for pecuniary gain — referred to the same aspect of the petitioner’s crime. Consequently, these facts supported only one aggravating circumstance. Second, citing Armstrong v. State, 399 So. 2d 953 (Fla. 1981), the court held that “[t]he recited circumstance, that the murders were especially heinous, atrocious, and cruel, cannot be approved.” 399 So. 2d, at 1373.18 The court affirmed the trial court’s findings that none of the statutory mitigating circumstances applied. Ibid. Because one of those findings was that Enmund’s participation in the capital felony was not minor, due to his role in planning the robbery, the State Supreme Court implicitly affirmed the finding that Enmund had planned the robbery.

Regarding the petitioner’s claim that imposition of the death penalty, absent a showing that he intended to kill, would violate the Eighth Amendment’s ban on cruel and unusual punishments, the court simply stated that the petitioner “offers us no binding legal authority that directly supports this proposition, and we therefore reject it.” Id., at 1371.

*810II

Earl Enmund’s claim in this Court is that the death sentence imposed by the Florida trial court, and affirmed by the Florida Supreme Court, is unconstitutionally disproportionate to the role he played in the robbery and murders of the Kerseys.19 In particular, he contends that because he had no actual intent to kill the victims — in effect, because his behavior and intent were no more blameworthy than that of any robber — capital punishment is too extreme a penalty.20

In Gregg v. Georgia, 428 U. S. 153 (1976), a majority of this Court concluded that the death penalty does not invariably violate the Cruel and Unusual Punishments Clause of the Eighth Amendment.21 See id., at 187 (opinion of Stewart, Powell, and Stevens, JJ.) (“[W]hen a life has been taken deliberately by the offender, we cannot say that the punishment is invariably disproportionate to the crime. It is an extreme sanction, suitable to the most extreme of crimes”) (footnote omitted); id., at 226 (opinion of White, J.) (rejecting the argument that “the death penalty, however imposed and for whatever crime, is cruel and unusual punishment”); *811id., at 227 (Blackmun, J., concurring in judgment). In no case since Gregg and its companion cases,22 has this Court retreated from that position.23 Recognizing the constitutional*812ity of the death penalty, however, only marks the beginning of the inquiry, for Earl Enmund was not convicted of murder as it is ordinarily envisioned — a deliberate and premeditated, unlawful killing. Rather, through the doctrine of accessorial liability, the petitioner has been convicted of two murders that he did not specifically intend.24 Thus, it is necessary to examine the concept of proportionality as enunciated in this Court’s cases to determine whether the penalty imposed on Earl Enmund is unconstitutionally disproportionate to his crimes.

A

The Eighth Amendment concept of proportionality was first fully expressed in Weems v. United States, 217 U. S. 349 (1910). In that case, defendant Weems was sentenced to 15 years at hard labor for falsifying a public document. *813After remarking that “it is a precept of justice that punishment for crime should be graduated and proportioned to offense,” id., at 367, and after comparing Weems’ punishment to the punishments for other crimes, the Court concluded that the sentence was cruel and unusual. Id., at 381.

Not until two-thirds of a century later, in Coker v. Georgia, 433 U. S. 584 (1977), did the Court declare another punishment to be unconstitutionally disproportionate to the crime. Writing for himself and three other Members of the Court, Justice White concluded that death is a disproportionate penalty for the crime of raping an adult woman. Id., at 597.25 In reaching this conclusion, the plurality was careful to inform its judgment “by objective factors to the maximum possible extent [by giving attention] to the public attitudes concerning a particular sentence — history and precedent, legislative attitudes, and the response of juries reflected in their sentencing decisions.” Id., at 592. The plurality’s resort to objective factors was no doubt an effort to derive “from the evolving standards of decency that mark the progress of a maturing society” the meaning of the requirement of proportionality contained within the Eighth Amendment. Trop v. Dulles, 356 U. S. 86, 101 (1958) (opinion of Warren, C. J.).

The plurality noted that within the previous 50 years a majority of the States had never authorized death as a punishment for rape. More significantly to the plurality, only 3 of the 35 States that immediately reinstituted the death penalty following the Court’s judgment in Furman v. Georgia, 408 U. S. 238 (1972) (invalidating nearly all state capital punish*814ment statutes), defined rape as a capital offense.26 The plurality also considered “the sentencing decisions that juries have made in the course of assessing whether capital punishment is an appropriate penalty for the crime being tried.” 433 U. S., at 596. See Gregg v. Georgia, 428 U. S., at 181 (opinion of Stewart, Powell, and Stevens, JJ.) (“The jury also is a significant and reliable objective index of contemporary values because it is so directly involved”). From the available data, the plurality concluded that in at least 90% of the rape convictions since 1973, juries in Georgia had declined to impose the death penalty. 433 U. S., at 597.

Thus, the conclusion reached in Coker rested in part on the Court’s observation that both legislatures and juries firmly rejected the penalty of death for the crime of rape. See Woodson v. North Carolina, 428 U. S. 280, 293 (1976) (opinion of Stewart, Powell, and Stevens, JJ.) (concluding that the State’s mandatory death penalty statute violates the Eighth Amendment because the “two crucial indicators of evolving standards of decency respecting the imposition of punishment in our society — jury determinations and legislative enactments — both point conclusively to the repudiation of automatic death sentences”).

In addition to ascertaining “contemporary standards,” the plurality opinion also considered qualitative factors bearing on the question whether the death penalty was disproportionate, for “the Constitution contemplates that in the end our own judgment will be brought to bear on the question of the acceptability of the death penalty under the Eighth Amendment.” 433 U. S., at 597. The plurality acknowledged that a rapist is almost as blameworthy as a murderer, describing *815the crime of rape as “highly reprehensible, both in a moral sense and in its almost total contempt for the personal integrity and autonomy of the female victim.” Ibid. Despite the enormity of the crime of rape, however, the Court concluded that the death penalty was “grossly out of proportion to the severity of the crime,” id., at 592, in part because the harm caused by a rape “does not compare with murder, which does involve the unjustified taking of human life.” Id., at 598.

Coker teaches, therefore, that proportionality — at least as regards capital punishment — not only requires an inquiry into contemporary standards as expressed by legislators and jurors, but also involves the notion that the magnitude of the punishment imposed must be related to the degree of the harm inflicted on the victim, as well as to the degree of the defendant’s blameworthiness.27 Moreover, because they turn on considerations unique to each defendant’s case, these latter factors underlying the concept of proportionality are reflected in this Court’s conclusion in Lockett v. Ohio, 438 U. S. 586, 605 (1978), that “individualized consideration [is] a constitutional requirement in imposing the death sentence” (opinion of Burger, C. J.) (footnote omitted). See id., at 613 (opinion of Blackmun, J.) (“the Ohio judgment in this case improperly provided the death sentence for a defendant who only aided and abetted a murder, without permitting any consideration by the sentencing authority of the extent of her involvement, or the degree of her mens rea, in the commission of the homicide”).

*816In sum, in considering the petitioner’s challenge, the Court should decide not only whether the petitioner’s sentence of death offends contemporary standards as reflected in the responses of legislatures and juries, but also whether it is disproportionate to the harm that the petitioner caused and to the petitioner’s involvement in the crime, as well as whether the procedures under which the petitioner was sentenced satisfied the constitutional requirement of individualized consideration set forth in Lockett.

B

Following the analysis set forth in Coker, the petitioner examines the historical development of the felony-murder rule, as well as contemporary legislation and jury verdicts in capital cases, in an effort to show that imposition of the death penalty on him would violate the Eighth Amendment. This effort fails, however, for the available data do not show that society has rejected conclusively the death penalty for felony murderers.

As the petitioner acknowledges, the felony-murder doctrine, and its corresponding capital penalty, originated hundreds of years ago,28 and was a fixture of English common law until 1957 when Parliament declared that an unintentional killing during a felony would be classified as manslaughter.29 The common-law rule was transplanted to the American Col*817onies, and its use continued largely unabated into the 20th century, although legislative reforms often restricted capital felony murder to enumerated violent felonies.30

The petitioner discounts the weight of this historical precedent by arguing that jurors and judges widely resisted the application of capital punishment by acquitting defendants in felony-murder cases or by convicting them of non-capital manslaughter.31 The force of the petitioner’s argument is speculative at best, however, for it is unclear what fraction of the jury nullification in this country resulted from dissatisfaction with the capital felony-murder rule. Much of it, surely, was a reaction to the mandatory death penalty, and the failure of the common law and early state statutes to classify murder by degree. In fact, it was in response to juror attitudes toward capital punishment that most jurisdictions by the early part of this century replaced their mandatory death penalty statutes with statutes allowing juries the discretion to decide whether to impose or to recommend the death penalty. See Woodson v. North Carolina, 428 U. S., at 291-292 (opinion of Stewart, Powell, and Stevens, JJ.).32 Thus, it simply is not possible to conclude that histori*818cally this country conclusively has rejected capital punishment for homicides committed during the course of a felony.

The petitioner and the Court turn to jury verdicts in an effort to show that, by present standards at least, capital punishment is grossly out of proportion to the crimes that the petitioner committed. Surveying all reported appellate court opinions since 1954 involving executions, the petitioner has found that of the 362 individuals executed for homicide, 339 personally committed the homicidal assault, and two others each had another person commit the homicide on his behalf. Only six persons executed were “non-triggermen.”33 A similar trend can be seen in the petitioner’s survey of the current death row population.34 Of the 739 prisoners for whom sufficient data are available, only 40 did not participate in the homicidal assault, and of those, only 3 (including the petitioner) were sentenced to death absent a finding that they had collaborated with the killer in a specific plan to kill. Brief for Petitioner 35-36. See also App. to Reply Brief for Petitioner (showing that of the 45 felony murderers currently on death row in Florida, 36 were found by the State Supreme Court or a trial court to have had the intent to kill; in 8 cases, the state courts made no finding, but the defendant was the triggerman; and in 1, the petitioner’s case, the defendant was not the triggerman, and there was no finding of intent to kill).

Impressive as these statistics are at first glance, they cannot be accepted uncritically. So stated, the data do not reveal the number or fraction of homicides that were charged as felony murders, or the number or fraction of cases in which the State sought the death penalty for an accomplice guilty of *819felony murder. Consequently, we cannot know the fraction of cases in which juries rejected the death penalty for accomplice felony murder. Moreover, as Justice Blackmun pointed out in his concurring opinion in Lockett v. Ohio, 438 U. S., at 615, n. 2, many of these data classify defendants by whether they “personally committed a homicidal assault,” and do not show the fraction of capital defendants who were shown to have an intent to kill. While the petitioner relies on the fact that he did not pull the trigger, his principal argument is, and must be, that death is an unconstitutional penalty absent an intent to kill, for otherwise defendants who hire others to kill would escape the death penalty. See n. 20, supra. Thus, the data he presents are not entirely relevant. Even accepting the petitioner’s facts as meaningful, they may only reflect that sentencers are especially cautious in imposing the death penalty, and reserve that punishment for those defendants who are sufficiently involved in the homicide, whether or not there was specific intent to kill.

Finally, as the petitioner acknowledges, the jury verdict statistics cannot be viewed in isolation from state death penalty legislation. The petitioner and the Court therefore review recent legislation in order to support the conclusion that society has rejected capital felony murder. Of the 35 States that presently have a death penalty, however, fully 31 authorize a sentencer to impose a death sentence for a death that occurs during the course of a robbery.35 The States are not uniform in delimiting the circumstances under which the *820death penalty may be imposed for felony murder, but each state statute can be classified as one of three types. The first category, containing 20 statutes, includes those States that permit imposition of the death penalty for felony murder even though the defendant did not commit the homicidal act, and even though he had no actual intent to kill.36 Three addi*821tional States, while requiring some finding of intent, do not require the intent to kill that the petitioner believes is constitutionally mandated before the death sentence may be imposed.37 The second category, containing seven statutes, includes those States that authorize the death penalty only if the defendant had the specific intent (or some rough equivalent) to kill the victim.38 The third class of statutes, from *822only three States, restricts application of the death penalty to those felony murderers who actually commit the homicide.39

The Court’s curious method of counting the States that authorize imposition of the death penalty for felony murder cannot hide the fact that 23 States permit a sentencer to impose the death penalty even though the felony murderer has neither killed nor intended to kill his victim. While the Court acknowledges that eight state statutes follow the Florida death penalty scheme, see ante, at 789, n. 5, it also concedes that 15 other , statutes permit imposition of the death penalty where the defendant neither intended to kill or actually killed the victims. See ante, at 790, n. 8 (Arkansas, Delaware, and Kentucky); ante, at 793-794, n. 15 (New Mexico); ante, at 791, n. 10 (Colorado); ante, at 791, n. 11 (Vermont); ante, *823at 792, n. 12 (Arizona, Connecticut, Indiana, Montana, Nebraska, and North Carolina); ante, at 792, n. 13 (Idaho, Oklahoma, and South Dakota). Not all of the statutes list the same aggravating circumstances. Nevertheless, the question before the Court is not whether a particular species of death penalty statute is unconstitutional, but whether a scheme that permits imposition of the death penalty, absent a finding that the defendant either killed or intended to kill the victims, is unconstitutional. In short, the Court’s peculiar statutory analysis cannot withstand closer scrutiny.

Thus, in nearly half of the States, and in two-thirds of the States that permit the death penalty for murder, a defendant who neither killed the victim nor specifically intended that the victim die may be sentenced to death for his participation in the robbery-murder. Far from “weighting] very heavily on the side of rejecting capital punishment as a suitable penalty for” felony murder, Coker v. Georgia, 443 U. S., at 596, these legislative judgments indicate that our “evolving standards of decency” still embrace capital punishment for this crime. For this reason, I conclude that the petitioner has failed to meet the standards in Coker and Woodson that the “two crucial indicators of evolving standards of decency . . . —jury determinations and legislative enactments — both point conclusively to the repudiation” of capital punishment for felony murder. 428 U. S., at 293 (emphasis added). In short, the death penalty for felony murder does not fall short of our national “standards of decency.”

C

As I noted earlier, the Eighth Amendment concept of proportionality involves more than merely a measurement of contemporary standards of decency. It requires in addition that the penalty imposed in a capital case be proportional to the harm caused and the defendant’s blameworthiness. Critical to the holding in Coker, for example, was that “in terms of moral depravity and of the injury to the person and *824to the public, [rape] does not compare with murder, which . . . involve[s] the unjustified taking of human life.” 433 U. S., at 598.

Although the Court disingenuously seeks to characterize Enmund as only a “robber,” ante, at 797, it cannot be disputed that he is responsible, along with Sampson and Jeanette Armstrong, for the murders of the Kerseys. There is no dispute that their lives were unjustifiably taken, and that the petitioner, as one who aided and abetted the armed robbery, is legally liable for their deaths.40 Quite unlike the defendant in Coker, the petitioner cannot claim that the penalty imposed is “grossly out of proportion” to the harm for which he admittedly is at least partly responsible.

The Court’s holding today is especially disturbing because it makes intent a matter of federal constitutional law, requiring this Court both to review highly subjective definitional problems customarily left to state criminal law and to develop an Eighth Amendment meaning of intent. As Justice Blackmun pointed out in his concurring opinion in Lockett, the Court’s holding substantially “interfere[s] with the States’ individual statutory categories for assessing legal *825guilt.” 438 U. S., at 616.41 See also id., at 635-636 (opinion of Rehnquist, J.) (rejecting the idea that intent to kill must be proved before the State can impose the death penalty). Although the Court’s opinion suggests that intent can be ascertained as if it were some historical fact, in fact it is a legal concept, not easily defined. Thus, while proportionality requires a nexus between the punishment imposed and the defendant’s blameworthiness, the Court fails to explain why the Eighth Amendment concept of proportionality requires rejection of standards of blameworthiness based on other levels of intent, such as, for example, the intent to commit an armed robbery coupled with the knowledge that armed robberies involve substantial risk of death or serious injury to other persons. Moreover, the intent-to-kill requirement is crudely crafted; it fails to take into account the complex picture of the defendant’s knowledge of his accomplice’s intent and whether he was armed, the defendant’s contribution to the planning and success of the crime, and the defendant’s actual participation during the commission of the crime. Under the circumstances, the determination of the degree of blameworthiness is best left to the sentencer, who can sift through the facts unique to each case. Consequently, while the type of mens rea of the defendant must be considered carefully in assessing the proper penalty, it is not so critical a factor in determining blameworthiness as to require a finding of intent to kill in order to impose the death penalty for felony murder.

In sum, the petitioner and the Court have failed to show that contemporary standards, as reflected in both jury determinations and legislative enactments, preclude imposition of *826the death penalty for accomplice felony murder. Moreover, examination of the qualitative factors underlying the concept of proportionality do not show that the death penalty is disproportionate as applied to Earl Enmund. In contrast to the crime in Coker, the petitioner’s crime involves the very type of harm that this Court has held justifies the death penalty. Finally, because of the unique and complex mixture of facts involving a defendant’s actions, knowledge, motives, and participation during the commission of a felony murder, I believe that the factfinder is best able to assess the defendant’s blameworthiness. Accordingly, I conclude that the death penalty is not disproportionate to the crime of felony murder, even though the defendant did not actually kill or intend to kill his victims.42

*827III

Although I conclude that the death penalty is not disproportionate to the crime of felony murder, I believe that, in light of the State Supreme Court’s rejection of critical factual findings, our previous opinions require a remand for a new sentencing hearing.43 Repeatedly, this Court has emphasized that capital sentencing decisions must focus “on the circumstances of each individual homicide and individual defendant.” Proffitt v. Florida, 428 U. S. 242, 258 (1976) (opinion of Stewart, Powell, and Stevens, JJ.). In striking down the mandatory capital punishment statute in Woodson v. North Carolina, 428 U. S., at 304, a plurality of the Court wrote:

“A process that accords no significance to relevant facets of the character and record of the individual offender or the circumstances of the particular offense excludes from consideration in fixing the ultimate punishment of death the possibility of compassionate or mitigating factors stemming from the diverse frailties of humankind. It treats all persons convicted of a designated offense not as uniquely individual human beings, but as members of a faceless, undifferentiated mass to be subjected to the blind infliction of the penalty of death.

“. . . [W]e believe that in capital cases the fundamental respect for humanity underlying the Eighth Amendment requires consideration of the character and record of the individual offender and the circumstances of the particular offense as a constitutionally indispensable part of the process of inflicting the penalty of death.”

In Lockett v. Ohio, 438 U. S., at 605, a plurality of this Court concluded:

“Given that the imposition of death by public authority is so profoundly different from all other penalties, we can*828not avoid the conclusion that an individualized decision is essential in capital cases. The need for treating each defendant in a capital case with that degree of respect due the uniqueness of the individual is far more important than in noncapital cases. . . . The nonavailability of corrective or modifying mechanisms with respect to an executed capital sentence underscores the need for individualized consideration as a constitutional requirement in imposing the death sentence” (footnote omitted).

Accordingly, “the sentencer, in all but the rarest kind of capital case, [may] not be precluded from considering, as a mitigating factor, any aspect of the defendant’s character or record and any of the circumstances of the offense that the defendant proffers as a basis for a sentence less than death” (footnotes omitted). Id., at 604. See id., at 613 (opinion of Blackmun, J.) (concluding that the Ohio capital sentencing statute is unconstitutional because it “provided the death sentence for a defendant who only aided and abetted a murder, without permitting any consideration by the sentencing authority of the extent of her involvement, or the degree of her mens rea, in the commission of the homicide”); Green v. Georgia, 442 U. S. 95, 97 (1979) (per curiam) (holding that the exclusion of evidence, from the capital sentencing proceeding, that the petitioner was not present when the victim was killed violated due process because “[t]he excluded testimony was highly relevant to a critical issue in the punishment phase of the trial”); Eddings v. Oklahoma, 455 U. S. 104 (1982) (adopting the plurality’s rule in Lockett). Thus, in deciding whether or not to impose capital punishment on a felony murderer, a sentencer must consider any relevant evidence or arguments that the death penalty is inappropriate for a particular defendant because of his relative lack of mens rea and his peripheral participation in the murder. Because of the peculiar circumstances of this case, I conclude that the trial court did not give sufficient consideration to the petitioner’s role in the crimes, and thus did not consider the miti*829gating circumstances proffered by the defendant at his sentencing hearing.44

In sentencing the petitioner, the trial court found four statutory aggravating circumstances: the petitioner had been convicted previously of a violent felony; the murders had been committed during the course of a robbery; the murders had been committed for pecuniary gain; and the murders were especially heinous, atrocious, or cruel. In its factual findings, the trial court stated that the “armed robbery . . . was planned ahead of time by the defendant Enmund,” App. 30, and that he had shot each of the victims while they lay prone in order to eliminate them as witnesses. Id., at 30-31. The court expressly found that “none of the statutory mitigating circumstances applied” to the petitioner. Id., at 32 (emphasis in original). Among other findings, the court rejected Enmund’s claim that his participation in the murders had been “relatively minor,” and found instead that “his participation in the capital felony was major. The defendant Enmund planned the capital felony and actively participated in an attempt to avoid detection by disposing of the murder weapons.” Ibid.

The Florida Supreme Court rejected these findings in part. The court noted that there “was no direct evidence at trial that Earl Enmund was present at the back door of the Kersey home when the plan to rob the elderly couple led to their being murdered.” 399 So. 2d, at 1370. Rather,

“the only evidence of the degree of his participation is the jury’s likely inference that he was the person in the car by the side of the road near the scene of the crimes. The jury could have concluded that he was there, a few hundred feet away, waiting to help the robbers escape with the Kerseys’ money.” Ibid.

*830Consequently, the court expressly rejected the trial court’s finding that Enmund personally had committed the homicides. Reviewing the aggravating circumstances, the Supreme Court consolidated two of them, and rejected the trial court’s conclusion that the murders had been “heinous, atrocious, or cruel,” since the evidence showed that the Armstrongs had killed the Kerseys in a gun battle arising from Mrs. Kersey’s armed resistance, and not that the petitioner had killed them in an effort to eliminate them as witnesses. See Armstrong v. State, 399 So. 2d, at 963.

Although the state statutory procedures did not prevent the trial judge from considering any mitigating circumstances,45 the trial judge’s view of the facts, in part rejected by the State Supreme Court, effectively prevented such consideration. In his erroneous belief that the petitioner had shot both of the victims while they lay in a prone position in order to eliminate them as witnesses, the trial judge necessarily rejected the only argument offered in mitigation — that the petitioner’s role in the capital felonies was minor, undeserving of the death penalty, because the petitioner was in the car when the fatal shots were fired. This fundamental misunderstanding of the petitioner’s role in the crimes prevented the trial court from considering the “circumstances of the particular offense” in imposing sentence. Woodson v. North Carolina, 428 U. S., at 304. Moreover, this error was not so insignificant that we can be sure its effect on the *831sentencing judge’s decision was negligible.46 Accordingly, I would vacate the decision below insofar as it affirms the death sentence, and remand the case for a new sentencing hearing.

Justice Marshall

announced the judgment of the Court and delivered the opinion of the Court with respect to Parts I and II and an opinion with respect to Parts III, IV, and V, in which Justice Brennan, Justice Blackmun, and Justice Stevens join.

For centuries no jurisdiction has countenanced the execution of the insane, yet this Court has never decided whether the Constitution forbids the practice. Today we keep faith with our common-law heritage in holding that it does.

H

Alvin Bernard Ford was convicted of murder in 1974 and sentenced to death. There is no suggestion that he was incompetent at the time of his offense, at trial, or at sentenc*402ing. In early 1982, however, Ford began to manifest gradual changes in behavior. They began as an occasional peculiar idea or confused perception, but became more serious over time. After reading in the newspaper that the Ku Klux Klan had held a rally in nearby Jacksonville, Florida, Ford developed an obsession focused upon the Klan. His letters to various people reveal endless brooding about his “Klan work,” and an increasingly pervasive delusion that he had become the target of a complex conspiracy, involving the Klan and assorted others, designed to force him to commit suicide. He believed that the prison guards, part of the conspiracy, had been killing people and putting the bodies in the concrete enclosures used for beds. Later, he began to believe that his women relatives were being tortured and sexually abused somewhere in the prison. This notion developed into a delusion that the people who were tormenting him at the prison had taken members of Ford’s family hostage. The hostage delusion took firm hold and expanded, until Ford was reporting that 135 of his friends and family were being held hostage in the prison, and that only he could help them. By “day 287” of the “hostage crisis,” the list of hostages had expanded to include “senators, Senator Kennedy, and many other leaders.” App. 53. In a letter to the Attorney General of Florida, written in 1983, Ford appeared to assume authority for ending the “crisis,” claiming to have fired a number of prison officials. He began to refer to himself as “Pope John Paul, III,” and reported having appointed nine new justices to the Florida Supreme Court. Id., at 59.

Counsel for Ford asked a psychiatrist who had examined Ford earlier, Dr. Jamal Amin, to continue seeing him and to recommend appropriate treatment. On the basis of roughly 14 months of evaluation, taped conversations between Ford and his attorneys, letters written by Ford, interviews with Ford’s acquaintances, and various medical records, Dr. Amin concluded in 1983 that Ford suffered from “a severe, uncontrollable, mental disease which closely resembles ‘Paranoid *403Schizophrenia With Suicide Potential’” — a “major mental disorder . . . severe enough to substantially affect Mr. Ford’s present ability to assist in the defense of his life.” Id., at 91.

Ford subsequently refused to see Dr. Amin again, believing him to have joined the conspiracy against him, and Ford’s counsel sought assistance from Dr. Harold Kaufman, who interviewed Ford in November 1983. Ford told Dr. Kaufman that “I know there is some sort of death penalty, but I’m free to go whenever I want because it would be illegal and the executioner would be executed.” Id., at 65. When asked if he would be executed, Ford replied: “I can’t be executed because of the landmark case. I won. Ford v. State will prevent executions all over.” Id., at 66. These statements appeared amidst long streams of seemingly unrelated thoughts in rapid succession. Dr. Kaufman concluded that Ford had no understanding of why he was being executed, made no connection between the homicide of which he had been convicted and the death penalty, and indeed sincerely believed that he would not be executed because he owned the prisons and could control the Governor through mind waves. Id., at 67. Dr. Kaufman found that there was “no reasonable possibility that Mr. Ford was dissembling, malingering or otherwise putting on a performance . . . .” Id., at 65. The following month, in an interview with his attorneys, Ford regressed further into nearly complete incomprehensibility, speaking only in a code characterized by intermittent use of the word “one,” making statements such as “Hands one, face one. Mafia one. God one, father one, Pope one. Pope one. Leader one.” Id., at 72.

Counsel for Ford invoked the procedures of Florida law governing the determination of competency of a condemned inmate, Fla. Stat. § 922.07 (1985). Following the procedures set forth in the statute, the Governor of Florida appointed a panel of three psychiatrists to evaluate whether, under §922.07(2), Ford had “the mental capacity to understand the nature of the death penalty and the reasons why it was im*404posed upon him.” At a single meeting, the three psychiatrists together interviewed Ford for approximately 30 minutes. Each doctor then filed a separate two- or three-page report with the Governor, to whom the statute delegates the final decision. One doctor concluded that Ford suffered from “psychosis with paranoia” but had “enough cognitive functioning to understand the nature and the effects of the death penalty, and why it is to be imposed on him.” App. 103. Another found that, although Ford was “psychotic,” he did “know fully what can happen to him.” Id., at 105-106. The third concluded that Ford had a “severe adaptational disorder,” but did “comprehend his total situation including being sentenced to death, and all of the implications of that penalty.” Id., at 99-100. He believed that Ford’s disorder, “although severe, seem[ed] contrived and recently learned.” Id., at 100. Thus, the interview produced three different diagnoses, but accord on the question of sanity as defined by state law.

The Governor’s decision was announced on April 30, 1984, when, without explanation or statement, he signed a death warrant for Ford’s execution. Ford’s attorneys unsuccessfully sought a hearing in state court to determine anew Ford’s competency to suffer execution. Ford v. Wainwright, 451 So. 2d 471, 475 (Fla. 1984). Counsel then filed a petition for habeas corpus in the United States District Court for the Southern District of Florida, seeking an evidentiary hearing on the question of Ford’s sanity, proffering the conflicting findings of the Governor-appointed commission and subsequent challenges to their methods by other psychiatrists. The District Court denied the petition without a hearing. The Court of Appeals granted a certificate of probable cause and stayed Ford’s execution, Ford v. Strickland, 734 F. 2d 538 (CA11 1984), and we rejected the State’s effort to vacate the stay of execution. Wainwright v. Ford, 467 U. S. 1220 (1984). The Court of Appeals then addressed the merits of Ford’s claim and a divided panel affirmed the Dis*405trict Court’s denial of the writ. 752 F. 2d 526 (CA11 1985). This Court granted Ford’s petition for certiorari in order to resolve the important issue whether the Eighth Amendment prohibits the execution of the insane and, if so, whether the District Court should have held a hearing on petitioner’s claim. 474 U. S. 1019 (1985).

I — I

Since this Court last had occasion to consider the infliction of the death penalty upon the insane, our interpretations of the Due Process Clause and the Eighth Amendment have evolved substantially. In Solesbee v. Balkcom, 339 U. S. 9 (1950), a condemned prisoner claimed a due process right to a judicial determination of his sanity, yet the Court did not consider the possible existence of a right under the Eighth Amendment, which had not yet been applied to the States. The sole question the Court addressed was whether Georgia’s procedure for ascertaining sanity adequately effectuated that State’s own policy of sparing the insane from execution. See also Caritativo v. California, 357 U. S. 549 (1958); United States ex rel. Smith v. Baldi, 344 U. S. 561 (1953); Phyle v. Duffy, 334 U. S. 431 (1948); Nobles v. Georgia, 168 U. S. 398 (1897). Now that the Eighth Amendment has been recognized to affect significantly both the procedural and the substantive aspects of the death penalty, the question of executing the insane takes on a wholly different complexion. The adequacy of the procedures chosen by a State to determine sanity, therefore, will depend upon an issue that this Court has never addressed: whether the Constitution places a substantive restriction on the State’s power to take the fife of an insane prisoner.

There is now little room for doubt that the Eighth Amendment’s ban on cruel and unusual punishment embraces, at a minimum, those modes or acts of punishment that had been considered cruel and unusual at the time that the Bill of Rights was adopted. See Solem v. Helm, 463 U. S. 277, 285-286 (1983); id., at 312-313 (Burger, C. J., joined by *406White, Rehnquist, and O’Connor, JJ., dissenting); Fur-man v. Georgia, 408 U. S. 238, 264 (1972) (Brennan, J., concurring); McGautha v. California, 402 U. S. 183, 226 (1971) (Black, J., concurring). “Although the Framers may have intended the Eighth Amendment to go beyond the scope of its English counterpart, their use of the language of the English Bill of Rights is convincing proof that they intended to provide at least the same protection . . . .” Solern v. Helm, supra, at 286.

Moreover, the Eighth Amendment’s proscriptions are not limited to those practices condemned by the common law in 1789. See Gregg v. Georgia, 428 U. S. 153, 171 (1976) (opinion of Stewart, Powell, and Stevens, JJ.). Not bound by the sparing humanitarian concessions of our forebears, the Amendment also recognizes the “evolving standards of decency that mark the progress of a maturing society.” Trap v. Dulles, 356 U. S. 86, 101 (1958) (plurality opinion). In addition to considering the barbarous methods generally outlawed in the 18th century, therefore, this Court takes into account objective evidence of contemporary values before determining whether a particular punishment comports with the fundamental human dignity that the Amendment protects. See Coker v. Georgia, 433 U. S. 584, 597 (1977) (plurality opinion).

A

We begin, then, with the common law. The bar against executing a prisoner who has lost his sanity bears impressive historical credentials; the practice consistently has been branded “savage and inhuman.” 4 W. Blackstone, Commentaries *24-*25 (hereinafter Blackstone). Blackstone explained:

“[IJdiots and lunatics are not chargeable for their own acts, if committed when under these incapacities: no, not even for treason itself. Also, if a man in his sound memory commits a capital offence, and before arraignment for it, he becomes mad, he ought not to be arraigned for *407it: because he is not able to plead to it with that advice and caution that he ought. And if, after he has pleaded, the prisoner becomes mad, he shall not be tried: for how can he make his defence? If, after he be tried and found guilty, he loses his senses before judgment, judgment shall not be pronounced; and if, after judgment, he becomes of nonsane memory, execution shall be stayed: for peradventure, says the humanity of the English law, had the prisoner been of sound memory, he might have alleged something in stay of judgment or execution.” Ibid, (footnotes omitted).

Sir Edward Coke had earlier expressed the same view of the common law of England: “[B]y intendment of Law the execution of the offender is for example, . . . but so it is not when a mad man is executed, but should be a miserable spectacle, both against Law, and of extream inhumanity and cruelty, and can be no example to others.” 3 E. Coke, Institutes 6 (6th ed. 1680) (hereinafter Coke). Other recorders of the common law concurred. See 1 M. Hale, Pleas of the Crown 36 (1736) (hereinafter Hale); 1 W. Hawkins, Pleas of the Crown 2 (7th ed. 1795) (hereinafter Hawkins); Hawles, Remarks on the Trial of Mr. Charles Bateman, 11 How. St. Tr. 474, 477 (1685) (hereinafter Hawles).

As is often true of common-law principles, see 0. Holmes, The Common Law 5 (1881), the reasons for the rule are less sure and less uniform than the rule itself. One explanation is that the execution of an insane person simply offends humanity, Coke 6; another, that it provides no example to others and thus contributes nothing to whatever deterrence value is intended to be served by capital punishment. Ibid. Other commentators postulate religious underpinnings: that it is uncharitable to dispatch an offender “into another world, when he is not of a capacity to fit himself for it,” Hawles 477. It is also said that execution serves no purpose in these cases because madness is its own punishment: furiosus *408solo furore punitur. Blackstone *395. More recent commentators opine that the community’s quest for “retribution” — the need to offset a criminal act by a punishment of equivalent “moral quality” — is not served by execution of an insane person, which has a “lesser value” than that of the crime for which he is to be punished. Hazard & Louisell, Death, the State, and the Insane: Stay of Execution, 9 UCLA L. Rev. 381, 387 (1962). Unanimity of rationale, therefore, we do not find. “But whatever the reason of the law is, it is plain the law is so.” Hawles 477. We know of virtually no authority condoning the execution of the insane at English common law.1

Further indications suggest that this solid proscription was carried to America, where it was early observed that “the judge is bound” to stay the execution upon insanity of the prisoner. 1 J. Chitty, A Practical Treatise on the Criminal Law *761; see 1 F. Wharton, A Treatise on Criminal Law §59 (8th ed. 1880).

B

This ancestral legacy has not outlived its time. Today, no State in the Union permits the execution of the insane.2 It *409is clear that the ancient and humane limitation upon the State’s ability to execute its sentences has as firm a hold upon the jurisprudence of today as it had centuries ago in England. The various reasons put forth in support of the common-law restriction have no less logical, moral, and practical force than they did when first voiced. For today, no less than before, we may seriously question the retributive value of executing a person who has no comprehension of why he has been singled out and stripped of his fundamental right to life. See Note, The Eighth Amendment and the Execution of the Presently Incompetent, 32 Stan. L. Rev. 765, 777, n. 58 (1980). Similarly, the natural abhorrence civilized societies feel at killing one who has no capacity to come to grips with his own conscience or deity is still vivid today. And the intuition that such an execution simply offends humanity is evidently shared across this Nation. Faced with such widespread evidence of a restriction upon sovereign power, this Court is compelled to conclude that the Eighth Amendment *410prohibits a State from carrying out a sentence of death upon a prisoner who is insane. Whether its aim be to protect the condemned from fear and pain without comfort of understanding, or to protect the dignity of society itself from the barbarity of exacting mindless vengeance, the restriction finds enforcement in the Eighth Amendment.

I — I 1 — I HH

The Eighth Amendment prohibits the State from inflicting the penalty of death upon a prisoner who is insane. Petitioner’s allegation of insanity in his habeas corpus petition, if proved, therefore, would bar his execution. The question before us is whether the District Court was under an obligation to hold an evidentiary hearing on the question of Ford’s sanity. In answering that question, we bear in mind that, while the underlying social values encompassed by the Eighth Amendment are rooted in historical traditions, the manner in which our judicial system protects those values is purely a matter of contemporary law. Once a substantive right or restriction is recognized in the Constitution, therefore, its enforcement is in no way confined to the rudimentary process deemed adequate in ages past.

A

In a habeas corpus proceeding, “a federal evidentiary hearing is required unless the state-court trier of fact has after a full hearing reliably found the relevant facts.” Townsend v. Sain, 372 U. S. 293, 312-313 (1963). The habeas corpus statute, following this Court’s decision in Townsend, provides that, in general, “a determination after a hearing on the merits of a factual issue, made by a State court of competent jurisdiction . . . , shall be presumed to be correct,” and an evidentiary hearing not required. 28 U. S. C. § 2254(d). In this case, it is clear that no state court has issued any determination to which that presumption of correctness could be said to attach; indeed, no court played any role in the rejection of petitioner’s claim of insanity. Thus, quite simply, *411Townsend and §2254 require the District Court to grant a hearing de novo on that question.

But our examination does not stop there. For even when a state court has rendered judgment, a federal court is obliged to hold an evidentiary hearing on habeas corpus if, among other factors, “the factfinding procedure employed by the State court was not adequate to afford a full and fair hearing,” § 2254(d)(2); or “the material facts were not adequately developed at the State court hearing,” § 2254(d)(3); or “the applicant did not receive a full, fair, and adequate hearing in the State court proceeding.” § 2254(d)(6). If federal factfinding is to be avoided, then, in addition to providing a court judgment on the constitutional question, the State must also ensure that its procedures are adequate for the purpose of finding the facts.

B

The adequacy of a state-court procedure under Townsend is largely a function of the circumstances and the interests at stake. In capital proceedings generally, this Court has demanded that factfinding procedures aspire to a heightened standard of reliability. See, e. g., Spaziano v. Florida, 468 U. S. 447, 456 (1984). This especial concern is a natural consequence of the knowledge that execution is the most irremediable and unfathomable of penalties; that death is different. See Woodson v. North Carolina, 428 U. S. 280, 305 (1976) (opinion of Stewart, Powell, and Stevens, JJ.).

Although the condemned prisoner does not enjoy the same presumptions accorded a defendant who has yet to be convicted or sentenced, he has not lost the protection of the Constitution altogether; if the Constitution renders the fact or timing of his execution contingent upon establishment of a further fact, then that fact must be determined with the high regard for truth that befits a decision affecting the life or death of a human being. Thus, the ascertainment of a prisoner’s sanity as a predicate to lawful execution calls for no less stringent standards than those demanded in any *412other aspect of a capital proceeding. Indeed, a particularly acute need for guarding against error inheres in a determination that “in the present state of the mental sciences is at best a hazardous guess however conscientious.” Solesbee v. Balkcom, 339 U. S., at 23 (Frankfurter, J., dissenting). That need is greater still because the ultimate decision will turn on the finding of a single fact, not on a range of equitable considerations. Cf. Woodson v. North Carolina, supra, at 304. In light of these concerns, the procedures employed in petitioner’s case do not fare well.

C

Florida law directs the Governor, when informed that a person under sentence of death may be insane, to stay the execution and appoint a commission of three psychiatrists to examine the prisoner. Fla. Stat. §922.07 (1985 and Supp. 1986). “The examination of the convicted person shall take place with all three psychiatrists present at the same time.” Ibid. After receiving the report of the commission, the Governor must determine whether “the convicted person has the mental capacity to understand the nature of the death penalty and the reasons why it was imposed on him.” Ibid. If the Governor finds that the prisoner has that capacity, then a death warrant is issued; if not, then the prisoner is committed to a mental health facility. The procedure is conducted wholly within the executive branch, ex parte, and provides the exclusive means for determining sanity. Ford v. Wainwright, 451 So. 2d, at 475.

Petitioner received the statutory process. The Governor selected three psychiatrists, who together interviewed Ford for a total of 30 minutes, in the presence of eight other people, including Ford’s counsel, the State’s attorneys, and correctional officials. The Governor’s order specifically directed that the attorneys should not participate in the examination in any adversarial manner. This order was consistent with the present Governor’s “publicly announced pol*413icy of excluding all advocacy on the part of the condemned from the process of determining whether a person under a sentence of death is insane.” Goode v. Wainwright, 448 So. 2d 999, 1001 (Fla. 1984).

After submission of the reports of the three examining psychiatrists, reaching conflicting diagnoses but agreeing on the ultimate issue of competency, Ford’s counsel attempted to submit to the Governor some other written materials, including the reports of the two other psychiatrists who had examined Ford at greater length, one of whom had concluded that the prisoner was not competent to suffer execution. The Governor’s office refused to inform counsel whether the submission would be considered. The Governor subsequently issued his decision in the form of a death warrant. That this most cursory form of procedural review fails to achieve even the minimal degree of reliability required for the protection of any constitutional interest, and thus falls short of adequacy under Townsend, is self-evident.

IV

A

The first deficiency in Florida’s procedure lies in its failure to include the prisoner in the truth-seeking process. Notwithstanding this Court’s longstanding pronouncement that “[t]he fundamental requisite of due process of law is the opportunity to be heard,” Grannis v. Órdean, 234 U. S. 385, 394 (1914), state practice does not permit any material relevant to the ultimate decision to be submitted on behalf of the prisoner facing execution. In all other proceedings leading to the execution of an accused, we have said that the factfinder must “have before it all possible relevant information about the individual defendant whose fate it must determine.” Jurek v. Texas, 428 U. S. 262, 276 (1976) (plurality opinion). And we have forbidden States to limit the capital defendant’s submission of relevant evidence in mitigation of the sentence. Skipper v. South Carolina, 476 U. S. 1, 8 *414(1986); Lockett v. Ohio, 438 U. S. 586, 604 (1978) (joint opinion). It would be odd were we now to abandon our insistence upon unfettered presentation of relevant information, before the final fact antecedent to execution has been found.

Rather, consistent with the heightened concern for fairness and accuracy that has characterized our review of the process requisite to the taking of a human life, we believe that any procedure that precludes the prisoner or his counsel from presenting material relevant to his sanity or bars consideration of that material by the factfinder is necessarily inadequate. “[T]he minimum assurance that the life-and-death guess will be a truly informed guess requires respect for the basic ingredient of due process, namely, an opportunity to be allowed to substantiate a claim before it is rejected.” Solesbee v. Balkcom, supra, at 23 (Frankfurter, J., dissenting).

We recently had occasion to underscore the value to be derived from a factfinder’s consideration of differing psychiatric opinions when resolving contested issues of mental state. In Ake v. Oklahoma, 470 U. S. 68 (1985), we recognized that, because “psychiatrists disagree widely and frequently on what constitutes mental illness [and] on the appropriate diagnosis to be attached to given behavior and symptoms,” the factfinder must resolve differences in opinion within the psychiatric profession “on the basis of the evidence offered by each party” when a defendant’s sanity is at issue in a criminal trial. Id., at 81. The same holds true after conviction; without any adversarial assistance from the prisoner’s representative — especially when the psychiatric opinion he proffers is based on much more extensive evaluation than that of the state-appointed commission — the factfinder loses the substantial benefit of potentially probative information. The result is a much greater likelihood of an erroneous decision.

*415B

A related flaw in the Florida procedure is the denial of any opportunity to challenge or impeach the state-appointed psychiatrists’ opinions. “[C]ross-examination ... is beyond any doubt the greatest legal engine ever invented for the discovery of truth.” 5 J. Wigmore, Evidence § 1367 (J. Chadbourn rev. 1974). Cross-examination of the psychiatrists, or perhaps a less formal equivalent, would contribute markedly to the process of seeking truth in sanity disputes by bringing to light the bases for each expert’s beliefs, the precise factors underlying those beliefs, any history of error or caprice of the examiner, any personal bias with respect to the issue of capital punishment, the expert’s degree of certainty about his or her own conclusions, and the precise meaning of ambiguous words used in the report. Without some questioning of the experts concerning their technical conclusions, a factfinder simply cannot be expected to evaluate the various opinions, particularly when they are themselves inconsistent. See Barefoot v. Estelle, 463 U. S. 880, 899 (1983). The failure of the Florida procedure to afford the prisoner’s representative any opportunity to clarify or challenge the state experts’ opinions or methods creates a significant possibility that the ultimate decision made in reliance on those experts will be distorted.3

*416c

Perhaps the most striking defect in the procedures of Fla. Stat. §922.07 (1985 and Supp. 1986), as noted earlier, is the State’s placement of the decision wholly within the executive branch. Under this procedure, the person who appoints the experts and ultimately decides whether the State will be able to carry out the sentence that it has long sought is the Governor, whose subordinates have been responsible for initiating every stage of the prosecution of the condemned from arrest through sentencing. The commander of the State’s corps of prosecutors cannot be said to have the neutrality that is necessary for reliability in the factfinding proceeding.

Historically, delay of execution on account of insanity was not a matter of executive clemency (ex mandato regis) or judicial discretion (ex arbitrio judiéis); rather, it was required by law (ex necessitate legis). 1 N. Walker, Crime and Insanity in England 196 (1968). Thus, history affords no better basis than does logic for placing the final determination of a fact, critical to the trigger of a constitutional limitation upon the State’s power, in the hands of the State’s own chief executive. In no other circumstance of which we are aware is the vindication of a constitutional right entrusted to the unreviewable discretion of an administrative tribunal.

V

A

Having identified various failings of the Florida scheme, we must conclude that the State’s procedures for determining sanity are inadequate to preclude federal redetermination of the constitutional issue. We do not here suggest that only a full trial on the issue of sanity will suffice to protect the federal interests; we leave to the State the task of developing appropriate ways to enforce the constitutional restriction *417upon its execution of sentences.4 It may be that some high threshold showing on behalf of the prisoner will be found a necessary means to control the number of nonmeritorious or repetitive claims of insanity. Cf. Pate v. Robinson, 383 U. S. 375, 387 (1966) (hearing on competency to stand trial required if “sufficient doubt” of competency exists). Other legitimate pragmatic considerations may also supply the boundaries of the procedural safeguards that feasibly can be provided.

Yet the lodestar of any effort to devise a procedure must be the overriding dual imperative of providing redress for those with substantial claims and of encouraging accuracy in the factfinding determination. The stakes are high, and the “evidence” will always be imprecise. It is all the more important that the adversary presentation of relevant information be as unrestricted as possible. Also essential is that the manner of selecting and using the experts responsible for producing that “evidence” be conducive to the formation of neutral, sound, and professional judgments as to the prisoner’s ability to comprehend the nature of the penalty. Fidelity to these principles is the solemn obligation of a civilized society.

B

Today we have explicitly recognized in our law a principle that has long resided there. It is no less abhorrent today than it has been for centuries to exact in penance the life of one whose mental illness prevents him from comprehending the reasons for the penalty or its implications. In fight of the *418clear need for trustworthiness in any factual finding that will prevent or permit the carrying out of an execution, we hold that Fla. Stat. §922.07 (1985 and Supp. 1986) provides inadequate assurances of accuracy to satisfy the requirements of Townsend v. Sain, 372 U. S. 293 (1963). Having been denied a factfinding procedure “adequate to afford a full and fair hearing” on the critical issue, 28 U. S. C. § 2254(d)(2), petitioner is entitled to an evidentiary hearing in the District Court, de novo, on the question of his competence to be executed. Townsend v. Sain, swpra, at 312.

The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion.

It is so ordered.

Justice Powell,

concurring in part and concurring in the judgment.

I join Parts I and II of the Court’s opinion. As Justice Marshall ably demonstrates, execution of the insane was barred at common law precisely because it was considered cruel and unusual. In Solem v. Helm, 463 U. S. 277 (1983), we explained that while the Framers “may have intended the Eighth Amendment to go beyond the scope of its English counterpart, their use of the language of the English Bill of Rights is convincing proof that they intended to provide at least the same protection.” Id., at 286. It follows that the practice of executing the insane is barred by our own Constitution.

That conclusion leaves two issues for our determination: (i) the meaning of insanity in this context, and (ii) the procedures States must follow in order to avoid the necessity of de novo review in federal courts under 28 U. S. C. § 2254(d). The Court’s opinion does not address the first of these issues, and as to the second, my views differ substantially from Justice Marshall’s. I therefore write separately.

*419I

The Court holds today that the Eighth Amendment bars execution of a category of defendants defined by their mental state. The bounds of that category are necessarily governed by federal constitutional law. I therefore turn to the same sources that give rise to the substantive right to determine its precise definition: chiefly, our common-law heritage and the modern practices of the States, which are indicative of our “evolving standards of decency.” Trop v. Dulles, 356 U. S. 86, 101 (1958) (plurality opinion). See Solem v. Helm, supra, at 284-286; Gregg v. Georgia, 428 U. S. 153, 175-176 (1976) (opinion of Stewart, Powell, and Stevens, JJ.).

A

As the Court recognizes, ante, at 407-408, the ancient prohibition on execution of the insane rested on differing theories. Those theories do not provide a common answer when it comes to defining the mental awareness required by the Eighth Amendment as a prerequisite to a defendant’s execution. On the one hand, some authorities contended that the prohibition against executing the insane was justified as a way of preserving the defendant’s ability to make arguments on his own behalf. See 1 M. Hale, Pleas of the Crown 35 (1736) (“if after judgment he become of non sane memory, his execution shall be spared; for were he of sound memory he might allege somewhat in stay of judgment or execution”); accord 4 W. Blackstone, Commentaries *388-*389. Other authorities suggest, however, that the prohibition derives from more straightforward humanitarian concerns. Coke expressed the view that execution was intended to be an “example” to the living, but that the execution of “a mad man” was such “a miserable spectacle ... of extream inhumanity and cruelty” that it “can be no example to others.” 3 E. Coke, Institutes 6 (1794). Hawles added that it is “against Christian charity to send a great offender quick . . . into another world, when he is not of a capacity to fit himself for it.” *420Hawles, Remarks on the Trial of Mr. Charles Bateman, 11 How. St. Tr. 474, 477 (1685).

The first of these justifications has slight merit today. Modern practice provides far more extensive review of convictions and sentences than did the common law, including not only direct appeal but ordinarily both state and federal collateral review.1 Throughout this process, the defendant has access to counsel, by constitutional right at trial, and by employment or appointment at other stages of the process whenever the defendant raises substantial claims. Nor does the defendant merely have the right to counsel’s assistance; he also has the right to the effective assistance of counsel at trial and on appeal. Evitts v. Lucey, 469 U. S. 387 (1985); Strickland v. Washington, 466 U. S. 668 (1984). See Kimmelman v. Morrison, ante, at 392-393 (Powell, J., concurring in judgment). These guarantees are far broader than those enjoyed by criminal defendants at common law. It is thus unlikely indeed that a defendant today could go to his death with knowledge of undiscovered trial error that might set him free.

In addition, in cases tried at common law execution often followed fairly quickly after trial, so that incompetence at the *421time of execution was linked as a practical matter with incompetence at the trial itself. Our decisions already recognize, however, that a defendant must be competent to stand trial, and thus the notion that a defendant must be able to assist in his defense is largely provided for. See Drope v. Missouri, 420 U. S. 162 (1975).2

B

The more general concern of the common law — that executions of the insane are simply cruel — retains its vitality. It is as true today as when Coke lived that most men and women value the opportunity to prepare, mentally and spiritually, for their death. Moreover, today as at common law, one of the death penalty’s critical justifications, its retributive force, depends on the defendant’s awareness of the penalty’s existence and purpose. Thus, it remains true that executions of the insane both impose a uniquely cruel penalty and are inconsistent with one of the chief purposes of executions generally. For precisely these reasons, Florida requires the Governor to stay executions of those who “d[o] not have the mental capacity to understand the nature of the death penalty and why it was imposed” on them. Fla. Stat. §922.07 (1985 and Supp. 1986). See also Ill. Rev. Stat., ch. 38, ¶ 1005-2-3(a) (1985) (“A person is unfit to be executed if because of a mental condition he is unable to understand the nature and purpose of such sentence”); State v. Pastet, 169 Conn. 13, 28, 363 A. 2d 41, 49 (question is “whether the defendant was able to understand the nature of the sentencing proceedings, i. e., why he was being punished and the nature of his punishment”), cert, denied, 423 U. S. 937 (1975). A number of *422States have more rigorous standards,3 but none disputes the need to require that those who are executed know the fact of their impending execution and the reason for it.

Such a standard appropriately defines the kind of mental deficiency that should trigger the Eighth Amendment prohibition. If the defendant perceives the connection between his crime and his punishment, the retributive goal of the criminal law is satisfied. And only if the defendant is aware that his death is approaching can he prepare himself for his passing. Accordingly, I would hold that the Eighth Amendment forbids the execution only of those who are unaware of the punishment they are about to suffer and why they are to suffer it.

Petitioner’s claim of insanity plainly fits within this standard. According to petitioner’s proffered psychiatric examination, petitioner does not know that he is to be executed, but rather believes that the death penalty has been invalidated. App. 65-67. If this assessment is correct, petitioner *423cannot connect his execution to the crime for which he was convicted. Thus, the question is whether petitioner’s evidence entitles him to a hearing in Federal District Court on his claim.

II

Petitioner concedes that the Governor of Florida has determined that he is not insane under the standard prescribed by Florida’s statute, which is the same as the standard just described. Petitioner further concedes that there is expert evidence that supports the Governor’s finding. Thus, if that finding is entitled to a presumption of correctness under 28 U. S. C. § 2254(d), there is no ground for holding a hearing on petitioner’s federal habeas corpus petition.

I agree with Justice Marshall that the Governor’s finding is not entitled to a presumption of correctness under § 2254(d). I reach this conclusion for two independent reasons. First, § 2254(d) requires deference to the factual findings of “a State court of competent jurisdiction.” The term “State court” may have a certain amount of flexibility,4 but no amount of stretching can extend it to include the Governor. The essence of a “court” is independence from the prosecutorial arm of government and, as Justice Marshall correctly notes, the Governor is “[t]he commander of the State’s corps of prosecutors.” Ante, at 416. Unless the relevant language is to be read out of the statute, I see no basis for affording any deference to the Governor’s determination.

Second, the presumption of correctness does not attach to the Governor’s implicit finding of sanity because the State did not give petitioner’s claim “a full and fair hearing,” 28 U. S. C. § 2254(d)(2). This statutory phrase apparently was drawn from the Court’s opinion in Townsend v. Sain, 372 U. S. 293, 313 (1963). There, the Court concluded that where the state court’s “fact-finding procedure . . . was not *424adequate for reaching reasonably correct results,” or where the process “appeared] to be seriously inadequate for the ascertainment of the truth,” no presumption of correctness would attach to the state court’s findings when those findings were challenged on federal habeas corpus. Id., at 316.

At least in the context of competency determinations prior to execution, this standard is no different from the protection afforded by procedural due process. It is clear that an insane defendant’s Eighth Amendment interest in forestalling his execution unless or until he recovers his sanity cannot be deprived without a “fair hearing.” Indeed, fundamental fairness is the hallmark of the procedural protections afforded by the Due Process Clause. See Lassiter v. Department of Social Services of Durham County, 452 U. S. 18, 24-25 (1981). Thus, the question in this case is whether Florida’s procedures for determining petitioner’s sanity comport with the requirements of due process.

Together with Justice Marshall and Justice O’Con-nor, I would hold that they do not. As Justice O’Connor states, “[i]f there is one ‘fundamental requisite’ of due process, it is that an individual is entitled to an ‘opportunity to be heard.’” Post, at 430 (quoting Grannis v. Ordean, 234 U. S. 385, 394 (1914)). In this case, petitioner was deprived of that opportunity. The Florida statute does not require the Governor to consider materials submitted by the prisoner, and the present Governor has a “publicly announced policy of excluding” such materials from his consideration. Goode v. Wainwright, 448 So. 2d 999, 1001 (Fla. 1984). Thus, the determination of petitioner’s sanity appears to have been made solely on the basis of the examinations performed by state-appointed psychiatrists. Such a procedure invites arbitrariness and error by preventing the affected parties from offering contrary medical evidence or even from explaining the inadequacies of the State’s examinations. It does not, therefore, comport with due process. It follows that the State’s procedure was not “fair,” and that the Dis*425trict Court on remand must consider the question of petitioner’s competency to be executed.

I — H I — I I — I

While the procedures followed by Florida in this case do not comport with basic fairness, I would not require the kind of full-scale “sanity trial” that Justice Marshall appears to find necessary. See ante, at 413-416, 418. Due process is a flexible concept, requiring only “such procedural protections as the particular situation demands.” Mathews v. Eldridge, 424 U. S. 319, 334 (1976); Morrissey v. Brewer, 408 U. S. 471, 481 (1972). See also post, at 429 (O’Connor, J., concurring in result in part and dissenting in part). In this instance, a number of considerations support the conclusion that the requirements of due process are not as elaborate as Justice Marshall suggests.

First, the Eighth Amendment claim at issue can arise only after the prisoner has been validly convicted of a capital crime and sentenced to death. Thus, in this case the State has a substantial and legitimate interest in taking petitioner’s life as punishment for his crime. That interest is not called into question by petitioner’s claim. Rather, the only question raised is not whether, but when, his execution may take place.5 This question is important, but it is not comparable to the antecedent question whether petitioner should be executed at all. It follows that this Court’s decisions imposing heightened procedural requirements on capital trials and sentencing proceedings — e. g., Lockett v. Ohio, 438 U. S. 586 (1978) (plurality opinion); Turner v. Murray, 476 U. S. 28 (1986) — do not apply in this context.

Second, petitioner does not make his claim of insanity against a neutral background. On the contrary, in order to *426have been convicted and sentenced, petitioner must have been judged competent to stand trial, or his competency-must have been sufficiently clear as not to raise a serious question for the trial court. The State therefore may properly presume that petitioner remains sane at the time sentence is to be carried out,6 and may require a substantial threshold showing of insanity merely to trigger the hearing process. Cf. Ake v. Oklahoma, 470 U. S. 68, 82-83 (1985).

Finally, the sanity issue in this type of case does not resemble the basic issues at trial or sentencing. Unlike issues of historical fact, the question of petitioner’s sanity calls for a basically subjective judgment. See Addington v. Texas, 441 U. S. 418, 429-430 (1979); cf. Barefoot v. Estelle, 463 U. S. 880, 898-901 (1983). And unlike the determination of whether the death penalty is appropriate in a particular case, the competency determination depends substantially on expert analysis in a discipline fraught with “subtleties and nuances.” Addington, supra, at 430. This combination of factors means that ordinary adversarial procedures — complete with live testimony, cross-examination, and oral argument by counsel — are not necessarily the best means of arriving at sound, consistent judgments as to a defendant’s sanity. Cf. Parham v. J. R., 442 U. S. 584, 609 (1979) (“Common human experience and scholarly opinions suggest that the supposed protections of an adversary proceeding to determine the appropriateness of medical decisions for the commitment and treatment of mental and emotional illness may well be more illusory than real”).

*427We need not determine the precise limits that due process imposes in this area. In general, however, my view is that a constitutionally acceptable procedure may be far less formal than a trial. The State should provide an impartial officer or board that can receive evidence and argument from the prisoner’s counsel, including expert psychiatric evidence that may differ from the State’s own psychiatric examination. Beyond these basic requirements, the States should have substantial leeway to determine what process best balances the various interests at stake. As long as basic fairness is observed, I would find due process satisfied, and would apply the presumption of correctness of § 2254(d) on federal habeas corpus.

> HH

Because petitioner has raised a viable claim under the Eighth Amendment, and because that claim was not adjudicated fairly within the meaning of due process or of § 2254(d), petitioner is entitled to have his claim adjudicated by the District Court on federal habeas corpus. I therefore join the Court’s judgment.

Justice O’Connor,

with whom Justice White joins, concurring in the result in part and dissenting in part.

I am in full agreement with Justice Rehnquist’s conclusion that the Eighth Amendment does not create a substantive right not to be executed while insane. Accordingly, I do not join the Court’s reasoning or opinion. Because, however, the conclusion is for me inescapable that Florida positive law has created a protected liberty interest in avoiding execution while incompetent, and because Florida does not provide even those minimal procedural protections required by due process in this area, I would vacate the judgment and remand to the Court of Appeals with directions that the case be returned to the Florida system so that a hearing can be held in a manner consistent with the requirements of the Due Process Clause. I cannot agree, however, that the federal *428courts should have any role whatever in the substantive determination of a defendant’s competency to be executed.

As we explained in Hewitt v. Helms, 459 U. S. 460, 466 (1983), “[l]iberty interests protected by the Fourteenth Amendment may arise from two sources — the Due Process Clause itself and the laws of the States.” See also Meachum v. Fano, 427 U. S. 215, 223-227 (1976). With Justice Rehnquist, I agree that the Due Process Clause does not independently create a protected interest in avoiding the execution of a death sentence during incompetency. See also Solesbee v. Balkcom, 339 U. S. 9 (1950). The relevant provision of the Florida statute, however, provides that the Governor “shall” have the prisoner committed to a “Department of Corrections mental health treatment facility” if the prisoner “does not have the mental capacity to understand the nature of the death penalty and why it was imposed on him.” Fla. Stat. § 922.07(3) (1985 and Supp. 1986). Our cases leave no doubt that where a statute indicates with “language of an unmistakable mandatory character,” that state conduct injurious to an individual will not occur “absent specified substantive predicates,” the statute creates an expectation protected by the Due Process Clause. Hewitt v. Helms, swpra, at 471-472. See also Vitek v. Jones, 445 U. S. 480, 488-491 (1980); Greenholtz v. Nebraska Penal Inmates, 442 U. S. 1, 10 (1979) (entitlement created where under state law “there is [a] set of facts which, if shown, mandate a decision favorable to the individual”). That test is easily met here. Nor is it relevant that the statute creating the interest also specifies the procedures to be followed when the State seeks to deprive the individual of that interest. As we reaffirmed last Term, “[t]he categories of substance and procedure are distinct.” Cleveland Board of Education v. Loudermill, 470 U. S. 532, 541 (1985). Thus, regardless of the procedures the State deems adequate for determining the preconditions to adverse official action, federal law defines the kind of proc*429ess a State must afford prior to depriving an individual of a protected liberty or property interest. Id., at 541.

Although the state-created entitlement to avoid execution while insane unquestionably triggers the demands of the Due Process Clause, in my judgment those demands are minimal in this context. “It is axiomatic that due process ‘is flexible and calls for such procedural protections as the particular situation [requires].’” Greenholtz v. Nebraska Penal Inmates, supra, at 12, quoting Morrissey v. Brewer, 408 U. S. 471, 481 (1972). And there are any number of reasons for concluding that this “particular situation” warrants substantial caution before reading the Due Process Clause to mandate anything like the full panoply of trial-type procedures. The prisoner’s interest in avoiding an erroneous determination is, of course, very great. But I consider it self-evident that once society has validly convicted an individual of a crime and therefore established its right to punish, the demands of due process are reduced accordingly. Meachum v. Fano, supra, at 224. Moreover, the potential for false claims and deliberate delay in this context is obviously enormous. Nobles v. Georgia, 168 U. S. 398, 405-406 (1897). This potential is exacerbated by a unique feature of the prisoner’s protected interest in suspending the execution of a death sentence during incompetency. By definition, this interest can never be conclusively and finally determined: Regardless of the number of prior adjudications of the issue, until the very moment of execution the prisoner can claim that he has become insane sometime after the previous determination to the contrary. Hazard & Louisell, Death, the State and the Insane: Stay of Execution, 9 UCLA L. Rev. 381, 399-400 (1962). These difficulties, together with the fact that the issue arises only after conviction and sentencing, convince me that the Due Process Clause imposes few requirements on the States in this context.

Even given the broad latitude I would leave to the States in this area, however, I believe that one aspect of the Florida *430procedure for determining competency to be executed renders that procedure constitutionally deficient. If there is one “fundamental requisite” of due process, it is that an individual is entitled to an “opportunity to be heard.” Grannis v. Ordean, 234 U. S. 385, 394 (1914). As currently implemented, the Florida procedure for determining competency violates this bedrock principle. By Executive Order, the present Governor has provided that “[c]ounsel for the inmate and the State Attorney may be present [at the competency hearing] but shall not participate in the examination in any adversarial manner.” Exec. Order No. 83-137 (Dec. 9, 1983). See also Goode v. Wainwright, 448 So. 2d 999, 1001 (Fla. 1984) (describing the Governor’s “publicly announced policy of excluding all advocacy on the part of the condemned from the process of determining whether a person under a sentence of death is insane”). Indeed, respondent does not dispute that the Governor’s office has steadfastly refused to acknowledge whether it would even review the extensive psychiatric materials submitted by petitioner concerning his present mental state. While I would not invariably require oral advocacy or even cross-examination, due process at the very least requires that the decisionmaker consider the prisoner’s written submissions.

I conclude therefore that Florida law has created a protected expectation that no execution will be carried out while the prisoner lacks the “mental capacity to understand the nature of the death penalty and why it was imposed on him.” Fla. Stat. §922.07(3) (1985). Because Florida’s procedures are inadequate to satisfy even the minimal requirements of due process in this context, I would vacate the judgment below with instructions that the case be returned to Florida so that it might assess petitioner’s competency in a manner that accords with the command of the Fourteenth Amendment. In my view, however, the only federal question presented in cases such as this is whether the State’s positive law has created a liberty interest and whether its procedures *431are adequate to protect that interest from arbitrary deprivation. Once satisfied that the procedures were adequate, a federal court has no authority to second-guess a State’s substantive competency determination.

Justice Rehnquist,

with whom The Chief Justice joins, dissenting.

The Court today holds that the Eighth Amendment prohibits a State from carrying out a lawfully imposed sentence of death upon a person who is currently insane. This holding is based almost entirely on two unremarkable observations. First, the Court states that it “know[s] of virtually no authority condoning the execution of the insane at English common law.” Ante, at 408. Second, it notes that “[t]oday, no State in the Union permits the execution of the insane.” Ibid. Armed with these facts, and shielded by the claim that it is simply “keep[ing] faith with our common-law heritage,” ante, at 401, the Court proceeds to cast aside settled precedent and to significantly alter both the common-law and current practice of not executing the insane. It manages this feat by carefully ignoring the fact that the Florida scheme it finds unconstitutional, in which the Governor is assigned the ultimate responsibility of deciding whether a condemned prisoner is currently insane, is fully consistent with the “common-law heritage” and current practice on which the Court purports to rely.

The Court places great weight on the “impressive historical credentials” of the common-law bar against executing a prisoner who has lost his sanity. Ante, at 406-408. What it fails to mention, however, is the equally important and unchallenged fact that at common law it was the executive who passed upon the sanity of the condemned. See 1 N. Walker, Crime and Insanity in England 194-203 (1968). So when the Court today creates a constitutional right to a determination of sanity outside of the executive branch, it does so not in keeping with but at the expense of “our common-law heritage.”

*432In Solesbee v. Balkcom, 339 U. S. 9 (1950), a condemned prisoner claimed that he had a constitutional right to a judicial determination of his sanity. There, as here, the State did not approve the execution of insane persons and vested in the Governor the responsibility for determining, with the aid of experts, the sanity vel non of persons sentenced to death. In rejecting the prisoner’s claim, this Court stated:

“Postponement of execution because of insanity bears a close affinity not to trial for a crime but rather to reprieves of sentences in general. The power to reprieve has usually sprung from the same source as the power to pardon. Power of executive clemency in this country undoubtedly derived from the practice as it had existed in England. Such power has traditionally rested in governors or the President, although some of that power is often delegated to agencies such as pardon or parole boards. Seldom, if ever, has this power of executive clemency been subjected to review by the courts.” Id., at 11-12.

Despite references to “evolving standards of decency,” ante, at 406, and “the jurisprudence of today,” ante, at 409, the Court points to no change since Solesbee in the States’ approach to determining the sanity of a condemned prisoner. Current statutes quite often provide that initiation of inquiry into and/or final determination of postsentencing insanity is a matter for the executive or the prisoner’s custodian.* The Court’s profession of “faith to our common-law heritage” and *433“evolving standards of decency” is thus at best a half-truth. It is Florida’s scheme — which combines a prohibition against execution of the insane with executive-branch procedures for evaluating claims of insanity — that is more faithful to both traditional and modern practice. And no matter how longstanding and universal, laws providing that the State should not execute persons the executive finds insane are not themselves sufficient to create an Eighth Amendment right that sweeps away as inadequate the procedures for determining sanity crafted by those very laws.

Petitioner makes the alternative argument, not reached by the Court, that even if the Eighth Amendment does not prohibit execution of the insane, Florida’s decision to bar such executions creates a right in condemned persons to trial-type procedures to determine sanity. Here, too, Solesbee is instructive:

“Recently we have pointed out the necessary and inherent differences between trial procedures and post-conviction procedures such as sentencing. Williams v. New York, 337 U. S. 241. In that case we emphasized that certain trial procedure safeguards are not applicable to the process of sentencing. This principle applies even more forcefully to an effort to transplant every trial safeguard to a determination of sanity after conviction. As was pointed out in [Nobles v. Georgia, 168 U. S. 398 (1897)], to require judicial review every time a convicted defendant suggested insanity would make the possibility of carrying out a sentence depend upon ‘fecundity in making suggestion after suggestion of insanity.’ Nobles v. Georgia, supra, at 405-406. See also Phyle v. Duffy, [334 U. S. 431 (1948)]. To protect itself society must have power to try, convict, and execute sentences. Our legal system demands that this governmental duty be performed with scrupulous fairness to the accused. We cannot say that it offends due process to leave the question of a convicted person’s sanity to the solemn respon*434sibility of a state’s highest executive with authority to invoke the aid of the most skillful class of experts on the crucial questions involved.” 339 U. S., at 12-13.

Even the sole dissenter in Solesbee, Justice Frankfurter, agreed that if the Constitution afforded condemned prisoners no substantive right not to be executed when insane, then the State would be free to place on the Governor the responsibility for determining sanity. Id., at 15.

Petitioner argues that Solesbee is no longer controlling because it was decided “at a time when due process analysis still turned on the right-privilege distinction.” Brief for Petitioner 8. But as petitioner concedes, his due process claim turns on a showing that the Florida statute at issue here created an individual right not to be executed while insane. Even a cursory reading of the statute reveals that the only right it creates in a condemned prisoner is to inform the Governor that the prisoner may be insane. Fla. Stat. § 922.07(1) (1985). The only legitimate expectation it creates is that “[i]f the Governor decides that the convicted person does not have the mental capacity to understand the nature of the death penalty and why it was imposed on him, he shall have him committed to a Department of Corrections mental health treatment facility.” §922.07(3) (Supp. 1986) (emphasis added). Our recent cases in this area of the law may not be wholly consistent with one another. See Olim v. Wakinekona, 461 U. S. 238 (1983); Hewitt v. Helms, 459 U. S. 460 (1983); Vitek v. Jones, 445 U. S. 480 (1980); Greenholtz v. Nebraska Penal Inmates, 442 U. S. 1 (1979); Meachum v. Fano, 427 U. S. 215 (1976). I do not think this state of the law requires the conclusion that Florida has granted petitioner the sort of entitlement that gives rise to the procedural protections for which he contends.

In any event, I see no reason to reject the Solesbee Court’s conclusion that wholly executive procedures can satisfy due process in the context of a post-trial, postappeal, post-collateral-attack challenge to a State’s effort to carry out *435a lawfully imposed sentence. Creating a constitutional right to a judicial determination of sanity before that sentence may be carried out, whether through the Eighth Amendment or the Due Process Clause, needlessly complicates and postpones still further any finality in this area of the law. The defendant has already had a full trial on the issue of guilt, and a trial on the issue of penalty; the requirement of still a third adjudication offers an invitation to those who have nothing to lose by accepting it to advance entirely spurious claims of insanity. A claim of insanity may be made at any time before sentence and, once rejected, may be raised again; a prisoner found sane two days before execution might claim to have lost his sanity the next day, thus necessitating another judicial determination of his sanity and presumably another stay of his execution. See Nobles v. Georgia, 168 U. S. 398, 405-406 (1897).

Since no State sanctions execution of the insane, the real battle being fought in this case is over what procedures must accompany the inquiry into sanity. The Court reaches the result it does by examining the common law, creating a constitutional right that no State seeks to violate, and then concluding that the common-law procedures are inadequate to protect the newly created but common-law based right. I find it unnecessary to “constitutionalize” the already uniform view that the insane should not be executed, and inappropriate to “selectively incorporate” the common-law practice. I therefore dissent.

Justice O’Connor

delivered the opinion of the Court.

The question presented is whether the petitioners’ participation in the events leading up to and following the murder of four members of a family makes the sentences of death imposed by the Arizona courts constitutionally permissible although neither petitioner specifically intended to kill the victims and neither inflicted the fatal gunshot wounds. We hold that the Arizona Supreme Court applied an erroneous standard in making the findings required by Enmund v. Florida, 458 U. S. 782 (1982), and, therefore, vacate the judgments below and remand the case for further proceedings not inconsistent with this opinion.

*139W

Gary Tison was sentenced to life imprisonment as the result of a prison escape during the course of which he had killed a guard. After he had been in prison a number of years, Gary Tison’s wife, their three sons Donald, Ricky, and Raymond, Gary’s brother Joseph, and other relatives made plans to help Gary Tison escape again. See State v. Dorothy Tison, Cr. No. 108352 (Super. Ct. Maricopa County 1981). The Tison family assembled a large arsenal of weapons for this purpose. Plans for escape were discussed with Gary Tison, who insisted that his cellmate, Randy Greenawalt, also a convicted murderer, be included in the prison break. The following facts are largely evidenced by petitioners’ detailed confessions given as part of a plea bargain according to the terms of which the State agreed not to seek the death sentence. The Arizona courts interpreted the plea agreement to require that petitioners testify to the planning stages of the breakout. When they refused to do so, the bargain was rescinded and they were tried, convicted, and sentenced to death.

On July 30, 1978, the three Tison brothers entered the Arizona State Prison at Florence carrying a large ice chest filled ■with guns. The Tisons armed Greenawalt and their father, and the group, brandishing their weapons, locked the prison guards and visitors present in a storage closet. The five men fled the prison grounds in the Tisons’ Ford Galaxy automobile. No shots were fired at the prison.

After leaving the prison, the men abandoned the Ford automobile and proceeded on to an isolated house in a white Lincoln automobile that the brothers had parked at a hospital near the prison. At the house, the Lincoln automobile had a flat tire; the only spare tire was pressed into service. After two nights at the house, the group drove toward Flagstaff. As the group traveled on back roads and secondary highways through the desert, another tire blew out. The group de*140cided to flag down a passing motorist and steal a car. Raymond stood out in front of the Lincoln; the other four armed themselves and lay in wait by the side of the road. One car passed by without stopping, but a second car, a Mazda occupied by John Lyons, his wife Donnelda, his 2-year-old son Christopher, and his 15-year-old niece, Theresa Tyson, pulled over to render aid.

As Raymond showed John Lyons the flat tire on the Lincoln, the other Tisons and Greenawalt emerged. The Lyons family was forced into the backseat of the Lincoln. Raymond and Donald drove the Lincoln down a dirt road off the highway and then down a gas line service road farther into the desert; Gary Tison, Ricky Tison, and Randy Greenawalt followed in the Lyons’ Mazda. The two cars were parked trunk to trunk and the Lyons family was ordered to stand in front of the Lincoln’s headlights. The Tisons transferred their belongings from the Lincoln into the Mazda. They discovered guns and money in the Mazda which they kept, and they put the rest of the Lyons’ possessions in the Lincoln.

Gary Tison then told Raymond to drive the Lincoln still farther into the desert. Raymond did so, and, while the others guarded the Lyons and Theresa Tyson, Gary fired his shotgun into the radiator, presumably to completely disable the vehicle. The Lyons and Theresa Tyson were then escorted to the Lincoln and again ordered to stand in its headlights. Ricky Tison reported that John Lyons begged, in comments “more or less directed at everybody,” “Jesus, don’t kill me.” Gary Tison said he was “thinking about it.” App. 39, 108. John Lyons asked the Tisons and Greenawalt to “[g]ive us some water . . . just leave us out here, and you all go home.” Gary Tison then told his sons to go back to the Mazda and get some water. Raymond later explained that his father “was like in conflict with himself .... What it was, I think it was the baby being there and all this, and he wasn’t sure about what to do.” Id., at 20-21, 74.

*141The petitioners’ statements diverge to some extent, but it appears that both of them went back towards the Mazda, along with Donald, while Randy Greenawalt and Gary Tison stayed at the Lincoln guarding the victims. Raymond recalled being at the Mazda filling the water jug “when we started hearing the shots.” Id., at 21. Ricky said that the brothers gave the water jug to Gary Tison who then, with Randy Greenawalt went behind the Lincoln, where they spoke briefly, then raised the shotguns, and started firing. Id., at 41, 111. In any event, petitioners agree they saw Greenawalt and their father brutally murder their four captives with repeated blasts from their shotguns. Neither made an effort to help the victims, though both later stated they were surprised by the shooting. The Tisons got into the Mazda and drove away, continuing their flight. Physical evidence suggested that Theresa Tyson managed to crawl away from the bloodbath, severely injured. She died in the desert after the Tisons left.

Several days later the Tisons and Greenawalt were apprehended after a shootout at a police roadblock. Donald Tison was killed. Gary Tison escaped into the desert where he subsequently died of exposure. Raymond and Ricky Tison and Randy Greenawalt were captured and tried jointly for the crimes associated with the prison break itself and the shootout at the roadblock; each was convicted and sentenced.

The State then individually tried each of the petitioners for capital murder of the four victims as well as for the associated crimes of armed robbery, kidnaping, and car theft. The capital murder charges were based on Arizona felony-murder law providing that a killing occurring during the perpetration of robbery or kidnaping is capital murder, Ariz. Rev. Stat. Ann. § 13-452 (1956) (repealed 1978), and that each participant in the kidnaping or robbery is legally responsible for the acts of his accomplices. Ariz. Rev. Stat. Ann. §13-139 (1956) (repealed 1978). Each of the petitioners was con*142victed of the four murders under these accomplice liability and felony-murder statutes.1

Arizona law also provided for a capital sentencing proceeding, to be conducted without a jury, to determine whether the crime was sufficiently aggravated to warrant the death sentence. Ariz. Rev. Stat. Ann. §13-454(A) (Supp. 1973) (repealed 1978). The statute set out six aggravating and four mitigating factors. Ariz. Rev. Stat. Ann. §§ 13-454(E), (F) (Supp. 1973) (repealed 1978). The judge found three statutory aggravating factors:

(1) the Tisons had created a grave risk of death to others (not the victims);

(2) the murders had been committed for pecuniary gain;

(3) the murders were especially heinous.

The judge found no statutory mitigating factor. Importantly, the judge specifically found that the crime was not mitigated by the fact that each of the petitioners’ “participation was relatively minor.” Ariz. Rev. Stat. Ann. § 13-454(F)(3) (Supp. 1973) (repealed 1978). Rather, he found that the “participation of each [petitioner] in the crimes giving rise to the application of the felony murder rule in this case was very substantial.” App. 284-285. The trial judge also specifically found, id., at 285, that each “could reasonably have foreseen that his conduct. . . would cause or create a grave risk of . . . death.” Ariz. Rev. Stat. Ann. § 13-454(F)(4) (Supp. 1973) (repealed 1978). He did find, however, three nonstatutory mitigating factors:

(1) the petitioners’ youth — Ricky was 20 and Raymond was 19;

*143(2) neither had prior felony records;

(3) each had been convicted of the murders under the felony-murder rule.

Nevertheless, the judge sentenced both petitioners to death.

On direct appeal, the Arizona Supreme Court affirmed. The Court found:

“The record establishes that both Ricky and Raymond Tison were present when the homicides took place and that they occurred as part of and in the course of the escape and continuous attempt to prevent recapture. The deaths would not have occurred but for their assistance. That they did not specifically intend that the Lyonses and Theresa Tyson die, that they did not plot in advance that these homicides would take place, or that they did not actually pull the triggers on the guns which inflicted the fatal wounds is of little significance. ” State v. (Ricky Wayne) Tison, 129 Ariz. 526, 545, 633 P. 2d 335, 354 (1981).

In evaluating the trial court’s findings of aggravating and mitigating factors, the Arizona Supreme Court found the first aggravating factor — creation of grave risk to others — not supported by the evidence. All those killed were intended victims, and no one else was endangered. The Arizona Supreme Court, however, upheld the “pecuniary gain” and “heinousness” aggravating circumstances and the death sentences. This Court denied the Tisons’ petition for certio-rari. 459 U. S. 882 (1982).

Petitioners then collaterally attacked their death sentences in state postconviction proceedings alleging that Enmund v. Florida, 458 U. S. 782 (1982), which had been decided in the interim, required reversal. A divided Arizona Supreme Court, interpreting Enmund to require a finding of “intent to kill,” declared in Raymond Tison’s case “the dictate oí Enmund is satisfied,” writing:

*144“Intend [sic] to kill includes the situation in which the defendant intended, contemplated, or anticipated that lethal force would or might be used or that life would or might be taken in accomplishing the underlying felony. Enmund, supra; State v. Emery, [141 Ariz. 549, 554, 688 P. 2d 175, 180 (1984)] filed June 6, 1984.

“In the present case the evidence does not show that petitioner killed or attempted to kill. The evidence does demonstrate beyond a reasonable doubt, however, that petitioner intended to kill. Petitioner played an active part in preparing the breakout, including obtaining a getaway car and various weapons. At the breakout scene itself, petitioner played a crucial role by, among other things, holding a gun on prison guards. Petitioner knew that Gary Tison’s murder conviction arose out of the killing of a guard during an earlier prison escape attempt. Thus petitioner could anticipate the use of lethal force during this attempt to flee confinement; in fact, he later said that during the escape he would have been willing personally to kill in a ‘very close life or death situation,’ and that he recognized that after the escape there was a possibility of killings.

“The use of lethal force that petitioner contemplated indeed occurred when the gang abducted the people who stopped on the highway to render aid. Petitioner played an active part in the events that led to the murders. He assisted in the abduction by flagging down the victims as they drove by, while the other members of the gang remained hidden and armed. He assisted in escorting the victims to the murder site. At the site, petitioner, Ricky Tison and Greenawalt placed the gang’s possessions in the victims’ Mazda and the victims’ possessions in the gang’s disabled Lincoln Continental. After Gary Tison rendered the Lincoln inoperable by firing into its engine compartment, petitioner assisted in escorting the victims to the Lincoln. Petitioner then *145watched Gary Tison and Greenawalt fire in the direction of the victims. Petitioner did nothing to interfere. After the killings, petitioner did nothing to disassociate himself from Gary Tison and Greenawalt, but instead used the victims’ car to continue on the joint venture, a venture that lasted several more days.

“From these facts we conclude that petitioner intended to kill. Petitioner’s participation up to the moment of the firing of the fatal shots was substantially the same as that of Gary Tison and Greenawalt. . . . Petitioner, actively participated in the events leading to death by, inter alia, providing the murder weapons and helping abduct the victims. Also petitioner was present at the murder site, did nothing to interfere with the murders, and after the murders even continued on the joint venture.

“. . . In Enmund, unlike in the present case, the defendant did not actively participate in the events leading to death (by, for example, as in the present case, helping abduct the victims) and was not present at the murder site.” 142 Ariz. 454, 456-457, 690 P. 2d 755, 757-758 (1984).

In Ricky Tison’s case the Arizona Supreme Court relied on a similar recitation of facts to find intent. It found that though Ricky Tison had not said that he would have been willing to kill, he “could anticipate the use of lethal force during this attempt to flee confinement.” 142 Ariz. 446, 448, 690 P. 2d 747, 749 (1984). The court noted that Ricky Tison armed himself and hid on the side of the road with the others while Raymond flagged down the Lyons family. Ricky claimed to have a somewhat better view than Raymond did of the actual killing. Otherwise, the court noted, Ricky Tison’s participation was substantially the same as Raymond’s. Id., at 447-448, 690 P. 2d, at 748-749. We granted certiorari in *146order to consider the Arizona Supreme Court’s application of Enmund. 475 U. S. 1010 (1986).2

HH H-l

In Enmund v. Florida, this Court reversed the death sentence of a defendant convicted under Florida’s felony-murder rule. Enmund was the driver of the “getaway” car in an armed robbery of a dwelling. The occupants of the house, an elderly couple, resisted and Enmund’s accomplices killed them. The Florida Supreme Court found the inference that Enmund was the person in the car by the side of the road waiting to help his accomplices escape sufficient to support his sentence of death:

“‘[T]he only evidence of the degree of [Enmund’s] participation is the jury’s likely inference that he was the person in the car by the side of the road near the scene of the crimes. The jury could have concluded that he was there, a few hundred feet away, waiting to help the robbers escape with the Kerseys’ money. The evidence, therefore, was sufficient to find that the appellant was a principal of the second degree, constructively present aiding and abetting the commission of the crime of robbery. This conclusion supports the verdicts of murder in the first degree on the basis of the felony murder por*147tion of section 782.04(1)(a).’ 399 So. 2d, at 1370.” Enmund v. Florida, 458 U. S., at 786.

This Court, citing the weight of legislative and community opinion, found a broad societal consensus, with which it agreed, that the death penalty was disproportional to the crime of robbery-felony murder “in these circumstances.” Id., at 788. The Court noted that although 32 American jurisdictions permitted the imposition of the death penalty for felony murders under a variety of circumstances, Florida was 1 of only 8 jurisdictions that authorized the death penalty “solely for participation in a robbery in which another robber takes life.” Id., at 789. Enmund was, therefore, sentenced under a distinct minority regime, a regime that permitted the imposition of the death penalty for felony murder simpliciter. At the other end of the spectrum, eight States required a finding of intent to kill before death could be imposed in a felony-murder case and one State required actual participation in the killing. The remaining States authorizing capital punishment for felony murders fell into two somewhat overlapping middle categories: three authorized the death penalty when the defendant acted with recklessness or extreme indifference to human life, and nine others, including Arizona, required a finding of some aggravating factor beyond the fact that the killing had occurred during the course of a felony before a capital sentence might be imposed. Arizona fell into a subcategory of six States which made “minimal participation in a capital felony committed by another person a [statutory] mitigating circumstance.” Id., at 792. Two more jurisdictions required a finding that the defendant’s participation in the felony was not “relatively minor” before authorizing a capital sentence. Id., at 791.3

*148After surveying the States’ felony-murder statutes, the Enmund Court next examined the behavior of juries in cases like Enmund’s in its attempt to assess American attitudes toward capital punishment in felony-murder cases. Of 739 death row inmates, only 41 did not participate in the fatal assault. All but 16 of these were physically present at the scene of the murder and of these only 3, including Enmund, were sentenced to death in the absence of a finding that they had collaborated in a scheme designed to kill. The Court found the fact that only 3 of 739 death row inmates had been sentenced to death absent an intent to kill, physical presence, or direct participation in the fatal assault persuasive evidence that American juries considered the death sentence disproportional to felony murder simpliciter.

Against this background, the Court undertook its own proportionality analysis. Armed robbery is a serious offense, but one for which the penalty of death is plainly excessive; the imposition of the death penalty for robbery, therefore, violates the Eighth and Fourteenth Amendments’ proscription “‘against all punishments which by their excessive length or severity are greatly disproportioned to the offenses charged.’ ” Weems v. United States, 217 U. S. 349, 371 (1910) (quoting O’Neil v. Vermont, 144 U. S. 323, 339-340 (1892)); cf. Coker v. Georgia, 433 U. S. 584 (1977) (holding the death penalty disproportional to the crime of rape). Furthermore, the Court found that Enmund’s degree of participation in the murders was so tangential that it could not be said to justify a sentence of death. It found that neither the deterrent nor the retributive purposes of the death penalty were advanced by imposing the death penalty upon Enmund. The Enmund Court was unconvinced “that the threat that the death penalty will be imposed for murder will measurably deter one who does not kill and has no intention or purpose that life will be taken.” 458 U. S., at 798-799. In reaching this conclusion, the Court relied upon the fact that killing only rarely occurred during the course of robber*149ies, and such killing as did occur even more rarely resulted in death sentences if the evidence did not support an inference that the defendant intended to kill. The Court acknowledged, however, that “[i]t would be very different if the likelihood of a killing in the course of a robbery were so substantial that one should share the blame for the killing if he somehow participated in the felony.” Id., at 799.

That difference was also related to the second purpose of capital punishment, retribution. The heart of the retribution rationale is that a criminal sentence must be directly related to the personal culpability of the criminal offender. While the States generally have wide discretion in deciding how much retribution to exact in a given case, the death penalty, “unique in its severity and irrevocability,” Gregg v. Georgia, 428 U. S. 153, 187 (1976), requires the State to inquire into the relevant facets of “the character and record of the individual offender.” Woodson v. North Carolina, 428 U. S. 280, 304 (1976). Thus, in Enmund’s case, “the focus [had to] be on his culpability, not on that of those who committed the robbery and shot the victims, for we insist on ‘individualized consideration as a constitutional requirement in imposing the death sentence.’” Enmund v. Florida, supra, at 798 (quoting Lockett v. Ohio, 438 U. S. 586, 605 (1978)) (emphasis in original). Since Enmund’s own participation in the felony murder was so attenuated and since there was no proof that Enmund had any culpable mental state, Enmund v. Florida, supra, at 790-791, the death penalty was excessive retribution for his crimes.

Enmund explicitly dealt with two distinct subsets of all felony murders in assessing whether Enmund’s sentence was disproportional under the Eighth Amendment. At one pole was Enmund himself: the minor actor in an armed robbery, not on the scene, who neither intended to kill nor was found to have had any culpable mental state. Only a small minority of States even authorized the death penalty in such circumstances and even within those jurisdictions the death *150penalty was almost never exacted for such a crime. The Court held that capital punishment was disproportional in these cases. Enmund also clearly dealt with the other polar case: the felony murderer who actually killed, attempted to kill, or intended to kill. The Court clearly held that the equally small minority of jurisdictions that limited the death penalty to these circumstances could continue to exact it in accordance with local law when the circumstances warranted. The Tison brothers’ cases fall into neither of these neat categories.

Petitioners argue strenuously that they did not “intend to kill” as that concept has been generally understood in the common law. We accept this as true. Traditionally, “one intends certain consequences when he desires that his acts cause those consequences or knows that those consequences are substantially certain to result from his acts.” W. LaFave & A. Scott, Criminal Law § 28, p. 196 (1972); see Lockett v. Ohio, supra, at 625-626 (1978) (opinion of White, J.) (equating intent with purposeful conduct); see also Perkins, A Rationale of Mens Rea, 52 Harv. L. Rev. 905, 911 (1939). As petitioners point out, there is no evidence that either Ricky or Raymond Tison took any act which he desired to, or was substantially certain would, cause death.

The Arizona Supreme Court did not attempt to argue that the facts of this case supported an inference of “intent” in the traditional sense. Instead, the Arizona Supreme Court attempted to reformulate “intent to kill” as a species of foreseeability. The Arizona Supreme Court wrote:

“Intend [sic] to kill includes the situation in which the defendant intended, contemplated, or anticipated that lethal force would or might be used or that life would or might be taken in accomplishing the underlying felony.” 142 Ariz., at 456, 690 P. 2d, at 757.

This definition of intent is broader than that described by the Enmund Court. Participants in violent felonies like armed robberies can frequently “anticipate] that lethal force . . . *151might be used ... in accomplishing the underlying felony.” Enmund himself may well have so anticipated. Indeed, the possibility of bloodshed is inherent in the commission of any violent felony and this possibility is generally foreseeable and foreseen; it is one principal reason that felons arm themselves. The Arizona Supreme Court’s attempted reformulation of intent to kill amounts to little more than a restatement of the felony-murder rule itself. Petitioners do not fall within the “intent to kill” category of felony murderers for which Enmund explicitly finds the death penalty permissible under the Eighth Amendment.

On the other hand, it is equally clear that petitioners also fall outside the category of felony murderers for whom Enmund explicitly held the death penalty disproportional: their degree of participation in the crimes was major rather than minor, and the record would support a finding of the culpable mental state of reckless indifference to human life. We take the facts as the Arizona Supreme Court has given them to us. Cabana v. Bullock, 474 U. S. 376 (1986).

Raymond Tison brought an arsenal of lethal weapons into the Arizona State Prison which he then handed over to two convicted murderers, one of whom he knew had killed a prison guard in the course of a previous escape attempt. By his own admission he was prepared to kill in furtherance of the prison break. He performed the crucial role of flagging down a passing car occupied by an innocent family whose fate was then entrusted to the known killers he had previously armed. He robbed these people at their direction and then guarded the victims at gunpoint while they considered what next to do. He stood by and watched the killing, making no effort to assist the victims before, during, or after the shooting. Instead, he chose to assist the killers in their continuing criminal endeavors, ending in a gun battle with the police in the final showdown.

Ricky Tison’s behavior differs in slight details only. Like Raymond, he intentionally brought the guns into the prison *152to arm the murderers. He could have foreseen that lethal force might be used, particularly since he knew that his father’s previous escape attempt had resulted in murder. He, too, participated fully in the kidnaping and robbery and watched the killing after which he chose to aid those whom he had placed in the position to kill rather than their victims.

These facts not only indicate that the Tison brothers’ participation in the crime was anything but minor; they also would clearly support a finding that they both subjectively appreciated that their acts were likely to result in the taking of innocent life. The issue raised by this case is whether the Eighth Amendment prohibits the death penalty in the intermediate case of the defendant whose participation is major and whose mental state is one of reckless indifference to the value of human life. Enmund does not specifically address this point. We now take up the task of determining whether the Eighth Amendment proportionality requirement bars the death penalty under these circumstances.

Like the Enmund Court, we find the state legislatures’ judgment as to proportionality in these circumstances relevant to this constitutional inquiry.4 The largest number of States still fall into the two intermediate categories discussed in Enmund. Four States authorize the death penalty in *153felony-murder cases upon a showing of culpable mental state such as recklessness or extreme indifference to human life.5 Two jurisdictions require that the defendant’s participation be substantial6 and the statutes of at least six more, including Arizona, take minor participation in the felony expressly into account in mitigation of the murder.7 These requirements significantly overlap both in this case and in general, for the greater the defendant’s participation in the felony murder, the more likely that he acted with reckless indifference to human life. At a minimum, however, it can be said that all these jurisdictions, as well as six States which Enmund classified along with Florida as permitting capital punishment for felony murder simpliciter,8 and the three States which simply require some additional aggravation before imposing the death penalty upon a felony murderer,9 *154specifically authorize the death penalty in a felony-murder case where, though the defendant’s mental state fell short of intent to kill, the defendant was a major actor in a felony in which he knew death was highly likely to occur. On the other hand, even after Enmund, only 11 States authorizing capital punishment forbid imposition of the death penalty even though the defendant’s participation in the felony murder is major and the likelihood of killing is so substantial as to raise an inference of extreme recklessness.10 This substantial and recent legislative authorization of the death penalty for the crime of felony murder regardless of the absence of a finding of an intent to kill powerfully suggests that our society does not reject the death penalty as grossly excessive under these circumstances, Gregg v. Georgia, 428 U. S., at 179-181 (opinion of Stewart, Powell, and Stevens, JJ.); see also Coker v. Georgia, 433 U. S., at 594.

Moreover, a number of state courts have interpreted Enmund to permit the imposition of the death penalty in such aggravated felony murders. We do not approve or disapprove the judgments as to proportionality reached on the particular facts of these cases, but we note the apparent consensus that substantial participation in a violent felony under circumstances likely to result in the loss of innocent human life may justify the death penalty even absent an “intent to kill.” See, e. g., Clines v. State, 280 Ark. 77, 84, 656 S. W. 2d 684, 687 (1983) (armed, forced entry, nighttime robbery of private dwelling known to be occupied plus evidence that kill*155ing contemplated), cert. denied, 465 U. S. 1051 (1984); Deputy v. State, 500 A. 2d 581, 599-600 (Del. 1985) (defendant present at scene; robbed victims; conflicting evidence as to participation in killing), cert. pending, No. 85-6272; Ruffin v. State, 420 So. 2d 591, 594 (Fla. 1982) (defendant present, assisted codefendant in kidnaping, raped victim, made no effort to interfere with codefendant’s killing victim and continued on the joint venture); People v. Davis, 95 Ill. 2d 1, 52, 447 N. E. 2d 353, 378 (defendant present at the scene and had participated in other crimes with Holman, the trigger-man, during which Holman had killed under similar circumstances), cert. denied, 464 U. S. 1001 (1983); Selvage v. State, 680 S. W. 2d 17, 22 (Tex. Cr. App. 1984) (participant in jewelry store robbery during the course of which a security guard was killed; no evidence that defendant himself shot the guard but he did fire a weapon at those who gave chase); see also Allen v. State, 253 Ga. 390, 395, n. 3, 321 S. E. 2d 710, 715, n. 3 (1984) (“The result in [Enmund v. Florida] does not turn on the mere fact that Enmund was convicted of felony murder. It is important to note how attenuated was Enmund’s responsibility for the deaths of the victims in that case”), cert. denied, 470 U. S. 1059 (1985).

Against this backdrop, we now consider the proportionality of the death penalty in these midrange felony-murder cases for which the majority of American jurisdictions clearly authorize capital punishment and for which American courts have not been nearly so reluctant to impose death as they are in the case of felony murder simpliciter.11

*156A critical facet of the individualized determination of culpability required in capital cases is the mental state with which the defendant commits the crime. Deeply ingrained in our legal tradition is the idea that the more purposeful is the criminal conduct, the more serious is the offense, and, therefore, the more severely it ought to be punished. The ancient concept of malice aforethought was an early attempt to focus on mental state in order to distinguish those who deserved death from those who through “Benefit of . . . Clergy” would be spared. 23 Hen. 8, ch. 1, §§3, 4 (1531); 1 Edw. 6, ch. 12, § 10 (1547). Over time, malice aforethought came to be inferred from the mere act of killing in a variety of circumstances; in reaction, Pennsylvania became the first American jurisdiction to distinguish between degrees of murder, reserving capital punishment to “wilful, deliberate and premeditated” killings and felony murders. 3 Pa. Laws 1794, ch. 1766, pp. 186-187 (1810). More recently, in Lockett v. Ohio, 438 U. S. 586 (1978), the plurality opinion made clear that, the defendant’s mental state was critical to weighing a defendant’s culpability under a system of guided discretion, vacating a death sentence imposed under an Ohio statute that did not permit the sentencing authority to take into account “[t]he absence of direct proof that the defendant intended to cause the death of the victim.” Id., at 608 (opinion of Burger, C. J.); see also Eddings v. Oklahoma, 455 U. S. 104 (1982) (adopting position of Lockett plurality). In Enmund v. Florida, the Court recognized again the importance of mental state, explicitly permitting the death penalty in at least those cases where the felony murderer intended to kill and forbidding it in the case of a minor actor not shown to have had any culpable mental state.

*157A narrow focus on the question of whether or not a given defendant “intended to kill,” however, is a highly unsatisfactory means of definitively distinguishing the most culpable and dangerous of murderers. Many who intend to, and do, kill are not criminally hable at all — those who act in self-defense or with other justification or excuse. Other intentional homicides, though criminal, are often felt undeserving of the death penalty — those that are the result of provocation. On the other hand, some noninteritional murderers may be among the most dangerous and inhumane of all — the person who tortures another not caring whether the victim fives or dies, or the robber who shoots someone in the course of the robbery, utterly indifferent to the fact that the desire to rob may have the unintended consequence of killing the victim as well as taking the victim’s property. This reckless indifference to the value of human fife may be every bit as shocking to the moral sense as an “intent to kill.” Indeed it is for this very reason that the common law and modem criminal codes alike have classified behavior such as occurred in this case along with intentional murders. See, e.g., G. Fletcher, Rethinking Criminal Law §6.5, pp. 447-448 (1978) (“[I]n the common law, intentional killing is not the only basis for establishing the most egregious form of criminal homicide . . . . For example, the Model Penal Code treats reckless killing, ‘manifesting extreme indifference to the value of human life,’ as equivalent to purposeful and knowing killing”). Enmund held that when “intent to kill” results in its logical though not inevitable consequence — the taking of human fife — the Eighth Amendment permits the State to exact the death penalty after a careful weighing of the aggravating and mitigating circumstances. Similarly, we hold that the reckless disregard for human fife implicit in knowingly engaging in criminal activities known to carry a grave risk of death represents a highly culpable mental state, a mental state that may be taken into account in making a capi*158tal sentencing judgment when that conduct causes its natural, though also not inevitable, lethal result.

The petitioners’ own personal involvement in the crimes was not minor, but rather, as specifically found by the trial court, “substantial.” Far from merely sitting in a car away from the actual scene of the murders acting as the getaway driver to a robbery, each petitioner was actively involved in every element of the kidnaping-robbery and was physically present during the entire sequence of criminal activity culminating in the murder of the Lyons family and the subsequent flight. The Tisons’ high level of participation in these crimes further implicates them in the resulting deaths. Accordingly, they fall well within the overlapping second intermediate position which focuses on the defendant’s degree of participation in the felony.

Only a small minority of those jurisdictions imposing capital punishment for felony murder have rejected the possibility of a capital sentence absent an intent to kill, and we do not find this minority position constitutionally required. We will not attempt to precisely delineate the particular types of conduct and states of mind warranting imposition of the death penalty here. Rather, we simply hold that major participation in the felony committed, combined with reckless indifference to human life, is sufficient to satisfy the Enmund culpability requirement.12 The Arizona courts have clearly found that the former exists; we now vacate the judgments below and remand for determination of the latter in further proceedings not inconsistent with this opinion. Cabana v. Bullock, 474 U. S. 876 (1986).

It is so ordered.

*159Justice Brennan,

with whom Justice Marshall joins, and with whom Justice Blackmun and Justice Stevens join as to Parts I through IV-A, dissenting.

The murders that Gary Tison and Randy Greenawalt committed revolt and grieve all who learn of them. When the deaths of the Lyons family and Theresa Tyson were first reported, many in Arizona erupted “in a towering yell” for retribution and justice.1 Yet Gary Tison, the central figure in this tragedy, the man who had his family arrange his and Greenawalt’s escape from prison, and the man who chose, with Greenawalt, to murder this family while his sons stood by, died of exposure in the desert before society could arrest him and bring him to trial. The question this case presents is what punishment Arizona may constitutionally exact from two of Gary Tison’s sons for their role in these events. Because our precedents and our Constitution compel a different answer than the one the Court reaches today, I dissent.

HH

Under the felony-murder doctrine, a person who commits a felony is liable for any murder that occurs during the commission of that felony, regardless of whether he or she commits, attempts to commit, or intended to commit that murder. The doctrine thus imposes liability on felons for killings committed by cofelons during a felony. This curious doctrine is a living fossil from a legal era in which all felonies were punishable by death; in those circumstances, the state of mind of the felon with respect to the murder was understandably superfluous, because he or she could be executed simply for intentionally committing the felony.2 Today, in *160most American jurisdictions and in virtually all European and Commonwealth countries, a felon cannot be executed for a murder that he or she did not commit or specifically intend or attempt to commit. In some American jurisdictions, however, the authority to impose death in such circumstances still persists. Arizona is such a jurisdiction.

The proceedings below illustrate how, under the felony-murder doctrine, a defendant may be held liable and sentenced to death for a murder that he or she neither committed nor intended to commit. The prosecutor argued to the jury that it did not matter that Gary Tison and Randy Greenawalt had caused the killings, because under the felony-murder rule the Tisons could nonetheless be found legally responsible for those killings. App. 173-174, 185, 191. The trial judge’s instructions were consistent with the prosecutor’s argument. Id., at 179, 218-219. In sentencing petitioners, the trial court did not find that they had killed, attempted to kill, or intended to kill anyone. Id., at 280-289. Nevertheless, the court upheld the jury’s verdict that Ricky and Raymond Tison were liable under the felony-murder doctrine for the murders that their father and Randy Green-awalt had committed. Furthermore, the court found as an aggravating factor against petitioners the “heinous, cruel and depraved manner” in which Gary Tison and Randy Green-awalt carried out the murders. Id., at 282-283. As a result, the court imposed the death sentence.3

*161The Arizona Supreme Court affirmed. It held that the Tisons “did not specifically intend that the Lyons and Theresa Tyson die, that they did not plot in advance that these homicides would take place, [and] that they did not actually pull the triggers on the guns which inflicted the fatal wounds. ...” State v. Tison, 129 Ariz. 526, 545, 633 P. 2d 335, 354 (1981). The court found these facts to be “of little significance,” however, because “the non-participation in the shooting was not controlling since both [brothers] took part in the robbery, the kidnapping, and were present assisting in the detention of the Lyonses and Theresa Tyson while the homicides were committed.” State v. Tison, 129 Ariz. 546, 556, 633 P. 2d 355, 365 (1981). Thus, while the Arizona courts acknowledged that petitioners had neither participated in the shootings nor intended that they occur, those courts nonetheless imposed the death sentence under the theory of felony murder.

After the decision of the Arizona Supreme Court, this Court addressed, in Enmund v. Florida, 458 U. S. 782 (1982), the question “whether death is a valid penalty under the Eighth and Fourteenth Amendments for one who neither took life, attempted to take life, nor intended to take life.” Id., at 787. The question arose because the Florida Supreme Court affirmed the death sentence for Earl En-mund, an accomplice in an armed robbery in which his two cofelons had killed the two individuals that the felons had intended to rob. Enmund did not shoot anyone, and there was nothing in the record concerning Enmund’s mental state with regard to the killings, but the Florida Supreme Court had held him strictly hable for the killings under the felony-murder doctrine. Enmund v. State, 399 So. 2d 1362, 1369 (1981).

*162In reversing the Florida Supreme Court, this Court took note of the “overwhelming evidence” of “[society’s rejection of the death penalty for accomplice liability in felony murders.” 458 U. S., at 794. The Court observed that, in imposing the death penalty upon Enmund, the Florida Supreme Court had failed to focus on “Enmund’s own conduct . . . [and] on his culpability.” Id., at 798 (emphasis in original). The Court then explained, and rejected, the felony-murder doctrine as a theory of capital culpability.

“Enmund did not kill or intend to kill and thus his culpability is plainly different from that of the robbers who killed; yet the State treated them alike and attributed to Enmund the culpability of those who killed the Kerseys. This was impermissible under the Eighth Amendment.” Ibid, (emphasis added).

Enmund obviously cast considerable doubt on the constitutionality of the death sentences imposed on petitioners in this case. Following the Enmund decision, petitioners applied to the Arizona Supreme Court for postconviction review. They argued that Enmund prevented the State from imposing the death sentence because they, like Enmund, were accomplices to a felony in which killings occurred that they neither committed nor intended to commit. Despite its earlier holding that petitioners had not killed or intended to kill anyone, the Arizona Supreme Court again upheld the Tisons’ sentences. First, the court defined intent broadly, adopting a definition that equates “intent to kill” with the foreseeability of harm:

“Intend [sic] to kill includes the situation in which the defendant intended, contemplated, or anticipated that lethal force would or might be used or that life would or might be taken in accomplishing the underlying felony.” 142 Ariz. 454, 456, 690 P. 2d 755, 757 (1984).

The court then reviewed, in a passage this Court quotes at length, ante, at 144-145, petitioners’ conduct during the *163escape and subsequent flight. The court did not attempt to link any of petitioners’ statements or actions to the decision to kill the family, nor did it make any findings concerning petitioners’ mental states at the time of the shootings. Instead, the court found that each petitioner “could [have] anticipate^] the use of lethal force during this attempt to flee confinement.” 142 Ariz. 446, 448, 690 P. 2d 747, 749 (1984); 142 Ariz., at 456, 690 P. 2d, at 757. The Arizona Supreme Court then held, by a vote of 3-2, that this finding was sufficient to establish that petitioners “intended” (within the meaning of Enmund) to kill the Lyons family, and affirmed the death sentences.

The Arizona Supreme Court thus attempted to comply with Enmund by making a finding as to petitioners’ mental state. The foreseeability standard that the court applied was erroneous, however, because “the possibility of bloodshed is inherent in the commission of any violent felony and this possibility is generally foreseeable and foreseen.” Ante, at 151. Under the lower court’s standard, any participant in a violent felony during which a killing occurred, including Enmund, would be liable for the death penalty. This Court therefore properly rejects today the lower court’s misguided attempt to preserve its earlier judgment by equating intent with foreseeable harm. Ante, at 150-151. In my view, this rejection completes the analytic work necessary to decide this case, and on this basis petitioners’ sentences should have been vacated and the judgment reversed.

The Court has chosen instead to announce a new substantive standard for capital liability: a defendant’s “major participation in the felony committed, combined with reckless indifference to human life, is sufficient to satisfy the Enmund culpability requirement.” Ante, at 158. The Court then remands the case for a determination by the state court whether petitioners are culpable under this new standard. Nevertheless, the Court observes, in dictum, that “the record would support a finding of the culpable mental state of *164reckless indifference to human life.” Ante, at 151; see also ante, at 152. (“These facts . . . would clearly support a finding that [both sons] subjectively appreciated that their acts were likely to result in the taking of innocent life”).

I join no part of this. First, the Court’s dictum that its new category of mens rea is applicable to these petitioners is not supported by the record. Second, even assuming petitioners may be so categorized, objective evidence and this Court’s Eighth Amendment jurisprudence demonstrate that the death penalty is disproportionate punishment for this category of defendants. Finally, the fact that the Court reaches a different conclusion is illustrative of the profound problems that continue to plague capital sentencing.

HH HH

The facts on which the Court relies are not sufficient, in my view, to support the Court’s conclusion that petitioners acted with reckless disregard for human life.4 But even if they *165were, the Court’s decision to restrict its vision to the limited set of facts that “the Arizona Supreme Court has given . . . to us,” ante, at 151, is improper.5 By limiting itself to the facts the lower court found relevant to the foreseeability standard, this Court insulates itself from other evidence in the record directly relevant to the new standard articulated today. This evidence suggests that the question of petitioners’ mental states with respect to the shootings is very much an open one to be decided only after a thorough evidentiary hearing. I therefore stress that nothing in the Court’s opinion abrogates the State’s responsibility independently and fairly to consider all the relevant evidence before applying the Court’s new standard. See Cabana v. Bullock, 474 U. S. 376, 391 (1986) (“Considerations of federalism and comity counsel respect for the ability of state courts to carry out their role as the primary protectors of the rights of criminal defendants”).

The evidence in the record overlooked today regarding petitioners’ mental states with respect to the shootings is not trivial. For example, while the Court has found that petitioners made no effort prior to the shooting to assist the victims, the uncontradicted statements of both petitioners are *166that just prior to the shootings they were attempting to find a jug of water to give to the family. App. 20-21, 39-41, 74-75, 109. While the Court states that petitioners were on the scene during the shooting and that they watched it occur, Raymond stated that he and Ricky were still engaged in repacking the Mazda after finding the water jug when the shootings occurred. Id., at 21, 75. Ricky stated that they had returned with the water, but were still some distance (“farther than this room”) from the Lincoln when the shootings started, id., at 40-41, 111, and that the brothers then turned away from the scene and went back to the Mazda, id., at 113. Neither stated that they anticipated that the shootings would occur, or that they could have done anything to prevent them or to help the victims afterward.6 Both, however, expressed feelings of surprise, helplessness, and regret. This statement of Raymond’s is illustrative:

“Well, I just think you should know when we first came into this we had an agreement with my dad that nobody would get hurt because we [the brothers] wanted no one hurt. And when this [killing of the kidnap victims] came about we were not expecting it. And it took us by surprise as much as it took the family [the victims] by surprise because we were not expecting this to happen. And I feel bad about it happening. I wish we could [have done] something to stop it, but by the time it happened it was too late to stop it. And it’s just something *167we are going to live with the rest of our lives. It will always be there.” 142 Ariz., at 462, 690 P. 2d, at 763; see also App. 242.7

*168In light of this evidence, it is not surprising that the Arizona Supreme Court rested its judgment on the narrow ground that petitioners could have anticipated that lethal force might be used during the escape, or that the state probation officer — who reviewed at length all the facts concerning the sons’ mental states — did not recommend that the death sentence be imposed. The discrepancy between those aspects of the record on which the Court has chosen to focus and those aspects it has chosen to ignore underscores the point that a reliable and individualized Enmund determination can be made only by the trial court following an eviden-tiary hearing. See Cabana v. Bullock, 474 U. S., at 397-407 (Blackmun, J., dissenting); id., at 407-408 (Stevens, J., dissenting).

Ill

Notwithstanding the Court’s unwarranted observations on the applicability of its new standard to this case, the basic flaw in today’s decision is the Court’s failure to conduct the sort of proportionality analysis that the Constitution and past cases require. Creation of a new category of culpability is not enough to distinguish this case from Enmund. The Court must also establish that death is a proportionate punishment for individuals in this category. In other words, the Court must demonstrate that major participation in a felony with a state of mind of reckless indifference to human life deserves the same punishment as intending to commit a murder or actually committing a murder. The Court does not attempt to conduct a proportionality review of the kind performed in past cases raising a proportionality question, e. g., Solem v. Helm, 463 U. S. 277 (1983); Enmund v. Florida, 458 U. S. 782 (1982); Coker v. Georgia, 433 U. S. 584 (1977), but instead offers two reasons in support of its view.

A

One reason the Court offers for its conclusion that death is proportionate punishment for persons falling within its new *169category is that limiting the death penalty to those who intend to kill “is a highly unsatisfactory means of definitively distinguishing the most culpable and dangerous of murderers.” Ante, at 157. To illustrate that intention cannot be dispositive, the Court offers as examples “the person who tortures another not caring whether the victim lives or dies, or the robber who shoots someone in the course of the robbery, utterly indifferent to the fact that the desire to rob may have the unintended consequence of killing the victim as well as taking the victim’s property.” Ibid, (emphasis added). Influential commentators and some States have approved the use of the death penalty for persons, like those given in the Court’s examples, who kill others in circumstances manifesting an extreme indifference to the value of human life.8 Thus an exception to the requirement that only intentional murders be punished with death might be made for persons who actually commit an act of homicide; Enmund, by distinguishing from the accomplice case “those who kill,” clearly reserved that question. But the constitutionality of the death penalty for those individuals is no more relevant to this case than it was to Enmund, because this case, like Enmund, involves accomplices who did not kill. Thus, although some of the “most culpable and dangerous of murderers” may be those who killed without specifically intending to kill, it is considerably more difficult to apply that rubric con*170vincingly to those who not only did not intend to kill, but who also have not killed.9

It is precisely in this context — where the defendant has not killed — that a finding that he or she nevertheless intended to kill seems indispensable to establishing capital culpability. It is important first to note that such a defendant has not committed an act for which he or she could be sentenced to death. The applicability of the death penalty therefore turns entirely on the defendant’s mental state with regard to an act committed by another. Factors such as the defendant’s major participation in the events surrounding the killing or the defendant’s presence at the scene are relevant insofar as they illuminate the defendant’s mental state with regard to the killings. They cannot serve, however, as independent grounds for imposing the death penalty.

Second, when evaluating such a defendant’s mental state, a determination that the defendant acted with intent is qualitatively different from a determination that the defendant acted with reckless indifference to human life. The difference lies in the nature of the choice each has made. The reckless actor has not chosen to bring about the killing in the way the intentional actor has. The person who chooses to *171act recklessly and is indifferent to the possibility of fatal consequences often deserves serious punishment. But because that person has not chosen to kill, his or her moral and criminal culpability is of a different degree than that of one who killed or intended to kill.

The importance of distinguishing between these different choices is rooted in our belief in the “freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil.” Morissette v. United States, 342 U. S. 246, 250 (1952). To be faithful to this belief, which is “universal and persistent in mature systems of law,” ibid., the criminal law must ensure that the punishment an individual receives conforms to the choices that individual has made.10 Differential punishment of reckless and intentional actions is therefore essential if we are to retain “the relation between criminal liability and moral culpability” on which criminal justice depends. People v. Washington, 62 Cal. 2d 777, 783, 402 P. 2d 130, 134 (1965) (opinion of Traynor, C. J.). The State’s ultimate sanction — if it is ever to be used — must be reserved for those whose culpability is greatest. Cf. Enmund, 458 U. S., at 798 (“It is fundamental that ‘causing harm intentionally must be punished more severely than causing the same harm unintentionally’ ” (citation omitted)); United States v. United States Gypsum Co., 438 U. S. 422, 444 (1978).

Distinguishing intentional from reckless action in assessing culpability is particularly important in felony-murder cases. Justice White stressed the importance of this distinction in Lockett v. Ohio, 438 U. S. 586 (1978), a felony-murder case in *172which the petitioner’s death sentence was vacated on other grounds.

“[S]ociety has made a judgment, which has deep roots in the history of the criminal law. . . distinguishing at least for purpose of the imposition of the death penalty between the culpability of those who acted with and those who acted without a purpose to destroy life.

“[T]he type of conduct which Ohio would punish by death requires at most the degree of mens rea defined by the ALI Model Penal Code (1962) as recklessness: conduct undertaken with knowledge that death is likely to follow. Since I would hold that death may not be inflicted for killings consistent with the Eighth Amendment without a finding that the defendant engaged in conduct with the conscious purpose of producing death, these sentences must be set aside.” Id., at 626-628 (emphasis added; footnotes omitted).

In Enmund, the Court explained at length the reasons a finding of intent is a necessary prerequisite to the imposition of the death penalty. In any given case, the Court said, the death penalty must “measurably contribute]” to one or both of the two “social purposes” — deterrence and retribution— which this Court has accepted as justifications for the death penalty. Enmund, supra, at 798, citing Gregg v. Georgia, 428 U. S. 153, 183 (1976). If it does not so contribute, it “‘is nothing more than the purposeless and needless imposition of pain and suffering’ and hence an unconstitutional punishment.” Enmund, supra, at 798, quoting Coker v. Georgia, 433 U. S., at 592. Enmund’s lack of intent to commit the murder — rather than the lack of evidence as to his mental state — was the decisive factor in the Court’s decision that the death penalty served neither of the two purposes. With regard to deterrence, the Court was

*173“quite unconvinced . . . that the threat that the death penalty will be imposed for murder will measurably deter one who does not kill and has no intention or purpose that life wall be taken. Instead, it seems likely that ‘capital punishment can serve as a deterrent only when murder is the result of premeditation and deliberation’ . . . Enmund, supra, at 798-799.11

As for retribution, the Court again found that Enmund’s lack of intent, together with the fact that he did not kill the victims, was decisive. “American criminal law has long considered a defendant’s intention — and therefore his moral guilt— to be critical to the ‘degree of [his] criminal culpability.’” 458 U. S., at 800 (citation omitted). The Court concluded that “[pjutting Enmund to death to avenge two killings that he did not commit and had no intention of committing or causing does not measurably contribute to the retributive end of ensuring that the criminal gets his just deserts. ” Id., at 801. Thus, in Enmund the Court established that a finding of an intent to kill was a constitutional prerequisite for the imposition of the death penalty on an accomplice who did not kill. The Court has since reiterated that “Enmund . . . imposes a categorical rule: a person who has not in fact killed, attempted to kill, or intended that a killing take place or that lethal force be used may not be sentenced to death.” Ca*174bana v. Bullock, 474 U. S., at 386. The Court’s decision today to approve the death penalty for accomplices who lack this mental state is inconsistent with Enmund and with the only justifications this Court has put forth for imposing the death penalty in any case.

B

The Court’s second reason for abandoning the intent requirement is based on its survey of state statutes authorizing the death penalty for felony murder, and on a handful of state cases.12 On this basis, the Court concludes that “[o]nly *175a small minority of those jurisdictions imposing capital punishment for felony murder have rejected the possibility of a capital sentence absent an intent to kill, and we do not find this minority position constitutionally required.” Ante, at 158 (emphasis added). The Court would thus have us believe that “the majority of American jurisdictions clearly authorize capital punishment” in cases such as this. Ante, at 155. This is not the case. First, the Court excludes from its survey those jurisdictions that have abolished the death penalty and those that have authorized it only in circumstances different from those presented here. When these jurisdictions are included, and are considered with those jurisdictions that require a finding of intent to kill in order to impose the death sentence for felony murder, one discovers that approximately three-fifths of American jurisdictions do not authorize the death penalty for a nontriggerman absent a finding that he intended to kill. Thus, contrary to the Court’s implication that its view is consonant with that of “the majority of American jurisdictions,” ibid., the Court’s view is itself distinctly the minority position.13

*176Second, it is critical to examine not simply those jurisdictions that authorize the death penalty in a given circumstance, but those that actually impose it. Evidence that a penalty is imposed only infrequently suggests not only that jurisdictions are reluctant to apply it but also that, when it is applied, its imposition is arbitrary and therefore unconstitutional. Furman v. Georgia, 408 U. S. 238 (1972). Thus, the Court in Enmund examined the relevant statistics on the imposition of the death penalty for accomplices in a felony murder. The Court found that of all executions between 1954 and 1982, there were “only 6 cases out of 362 where a nontriggerman felony murderer was executed. All six executions took place in 1955 ” 458 U. S., at 794 (emphasis added). This evidence obviously militates against imposing the death penalty on petitioners as powerfully as it did against imposing it on Enmund.14

*177The Court in Enmund also looked at the imposition of the death penalty for felony murder within Florida, the State that had sentenced Enmund. Of the 45 murderers then on death row, 36 had been found to have “intended” to take life, and 8 of the 9 for which there was no finding of intent had been the triggerman. Thus in only one case — Enmund— had someone (such as the Tisons) who had neither killed nor intended to kill received the death sentence. Finally, the Court noted that in no Commonwealth or European country could Enmund have been executed, since all have either abolished or never employed a felony-murder doctrine. Id., at 796-797, n. 22.15

The Court today neither reviews nor updates this evidence. Had it done so, it would have discovered that, even *178including the 65 executions since Enmund, “[t]he fact remains that we are not aware of a single person convicted of felony murder over the past quarter century who did not kill or attempt to kill, and did not intend the death of the victim, who has been executed . . . 458 U. S., at 796.16 Of the 64 persons on death row in Arizona, all of those who have raised and lost an Enmund challenge in the Arizona Supreme Court have been found either to have killed or to have specifically intended to kill.17 Thus, like Enmund, the Tisons’ sentence *179appears to be an aberration within Arizona itself as well as nationally and internationally. The Court’s objective evidence that the statutes of roughly 20 States appear to authorize the death penalty for defendants in the Court’s new category is therefore an inadequate substitute for a proper proportionality analysis, and is not persuasive evidence that the punishment that was unconstitutional for Enmund is constitutional for the Tisons.

C

The Court’s failure to examine the full range of relevant evidence is troubling not simply because of what that examination would have revealed, but because until today such an examination has been treated as constitutionally required whenever the Court undertakes to determine whether a given punishment is disproportionate to the severity of a given crime. Enmund is only one of a series of cases that have framed the proportionality inquiry in this way. See, e. g., Coker v. Georgia, 433 U. S. 584 (1977). In the most recent stich case, Solem v. Helm, 463 U. S. 277, 292 (1983), the Court summarized the essence of the inquiry:

“In sum, a court’s proportionality analysis under the Eighth Amendment should be guided by objective criteria, including (i) the gravity of the offense and the harshness of the penalty; (ii) the sentences imposed on other criminals in the same jurisdiction; and (iii) the sentences *180imposed for commission of the same crime in other jurisdictions.” (Emphasis added.)

By addressing at best only the first of these criteria, the Court has ignored most of the guidance this Court has developed for evaluating the proportionality of punishment.

Such guidance is essential in determining the constitutional limits on the State’s power to punish. These limits must be defined with care, not simply because the death penalty is involved, but because the social purposes that the Court has said justify the death penalty — retribution and deterrence-are justifications that possess inadequate self-limiting principles. As Professor Packer observed, under a theory of deterrence the state may justify such punishments as “boiling people in oil; a slow and painful death may be thought more of a deterrent to crime than a quick and painless one.” Packer, Making the Punishment Fit the Crime, 77 Harv. L. Rev. 1071, 1076 (1964).18 Retribution, which has as its core logic *181the crude proportionality of “an eye for an eye,” has been regarded as a constitutionally valid basis for punishment only when the punishment is consistent with an “individualized consideration” of the defendant’s culpability, Lockett v. Ohio, 438 U. S., at 605, and when “the administration of criminal justice” works to “channel!]” society’s “instinct for retribution.” Furman v. Georgia, 408 U. S., at 308 (Stewart, J., concurring). Without such channeling, a State could impose a judgment of execution by torture as appropriate retribution for murder by torture.19 Thus, under a simple theory either of deterrence or retribution, unfettered by the Constitution, results disturbing to civil sensibilities and inconsistent with “the evolving standards of decency” in our society become rationally defensible. Cf. Trop v. Dulles, 356 U. S. 86, 101 (1958).

The Framers provided in the Eighth Amendment the limiting principles otherwise absent in the prevailing theories of punishment. One such principle is that the States may not impose punishment that is disproportionate to the severity of *182the offense or to the individual’s own conduct and culpability. Because the proportionality inquiry in this case overlooked evidence and considerations essential to such an inquiry, it is not surprising that the result appears incongruous. Ricky and Raymond Tison are similarly situated with Earl Enmund in every respect that mattered to the decision in Enmund. Like Enmund, the Tisons neither killed nor attempted or intended to kill anyone. Like Enmund, the Tisons have been sentenced to death for the intentional acts of others which the Tisons did not expect, which were not essential to the felony, and over which they had no control. Unlike Enmund, however, the Tisons will be the first individuals in over 30 years to be executed for such behavior.

I conclude that the proportionality analysis and result in this case cannot be reconciled with the analyses and results of previous cases. On this ground alone, I would dissent. But the fact that this Court’s death penalty jurisprudence can validate different results in analytically indistinguishable cases suggests that something more profoundly disturbing than faithlessness to precedent is at work in capital sentencing.

f — I <1

In 1922, “five negroes who were convicted of murder m the first degree and sentenced to death by the Court of the State of Arkansas” appealed to this Court from an order of the District Court dismissing their writ of habeas corpus. Moore v. Dempsey, 261 U. S. 86, 87 (1923). The crux of their appeal was that they “were hurried to conviction under the pressure of a mob without any regard for their rights and without according to them due process of law.” Ibid. In reversing the order, Justice Holmes stated the following for the Court:

“It certainly is true that mere mistakes of law in the course of a trial are not to be corrected [by habeas corpus]. But if the case is that the whole proceeding is a mask — that counsel, jury, and judge were swept to the fatal end by an irresistible wave of public passion, and *183that the State Courts failed to correct the wrong, neither perfection in the machinery for correction nor the possibility that the trial court and counsel saw no other way of avoiding an immediate outbreak of the mob can prevent this Court from securing to the petitioners their constitutional rights.” Id., at 91.

A

In Furman v. Georgia, supra, this Court concluded that the State’s procedural machinery was so imperfect that imposition of the death penalty had become arbitrary and therefore unconstitutional. A scant four years later, however, the Court validated Georgia’s new machinery, and in 1977 executions resumed. In this case, the State appears to have afforded petitioners all of the procedures that this Court has deemed sufficient to produce constitutional sentencing decisions. Yet in this case, as in Moore, “perfection in the [State’s] machinery for correction” has not secured to petitioners their constitutional rights. So rarely does any State (let alone any Western country other than our own) ever execute a person who neither killed nor intended to kill that “these death sentences are cruel and unusual in the same way that being struck by lightning is cruel and unusual.” Furman v. Georgia, supra, at 309 (Stewart, J., concurring). This case thus demonstrates, as Furman also did, that we have yet to achieve a system capable of “distinguishing the few cases in which the [death penalty] is imposed from the many cases in which it is not.” 408 U. S., at 313 (White, J., concurring).

What makes this a difficult case is the challenge of giving substantive content to the concept of criminal culpability. Our Constitution demands that the sentencing decision itself, and not merely the procedures that produce it, respond to the reasonable goals of punishment. But the decision to execute these petitioners, like the state courts’ decisions in Moore, and like other decisions to kill, appears responsive *184less to reason than to other, more visceral, demands. The urge to employ the felony-murder doctrine against accomplices is undoubtedly strong when the killings stir public passion and the actual murderer is beyond human grasp. And an intuition that sons and daughters must sometimes be punished for the sins of the father may be deeply rooted in our consciousness.20 Yet punishment that conforms more closely to such retributive instincts than to the Eighth Amendment is tragically anachronistic in a society governed by our Constitution.

B

This case thus illustrates the enduring truth of Justice Harlan’s observation that the tasks of identifying “those characteristics of criminal homicides and their perpetrators which call for the death penalty, and [of] expressing] these characteristics in language which can be fairly understood and applied by the sentencing authority appear to be . . . beyond present human ability.” McGautha v. California, 402 U. S. 183, 204 (1971) (emphasis added). The persistence of doctrines (such as felony murder) that allow excessive discretion in apportioning criminal culpability and of decisions (such as today’s) that do not even attempt “precisely [to] delineate the particular types of conduct and states of mind warranting imposition of the death penalty,” ante, at 158, demonstrates that this Court has still not articulated rules that will ensure that capital sentencing decisions conform to the substantive principles of the Eighth Amendment. Arbitrariness continues so to infect both the procedure and substance of capital sentencing that any decision to impose the *185death penalty remains cruel and unusual. For this reason, as well as for the reasons expressed in Gregg v. Georgia, 428 U. S., at 227, 1 adhere to my view that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments, and dissent.

*372Per Curiam.

Angel Francisco Breard is scheduled to be executed by the Commonwealth of Virginia this evening at 9 p.m. Breard, a citizen of Paraguay, came to the United States in 1986, at the *373age of 20. In 1992, Breard was charged with the attempted rape and capital murder of Ruth Dickie. At his trial in 1993, the State presented overwhelming evidence of guilt, including semen found on Dickie’s body matching Breard’s DNA profile and hairs on Dickie’s body identical in all microscopic characteristics to hair samples taken from Breard. Breard chose to take the witness stand in his defense. During his testimony, Breard confessed to killing Dickie, but explained that he had only done so because of a Satanic curse placed on him by his father-in-law. Following a jury trial in the Circuit Court of Arlington County, Virginia, Breard was convicted of both charges and sentenced to death. On appeal, the Virginia Supreme Court affirmed Breard’s convictions and sentences, Breard v. Commonwealth, 248 Va. 68, 445 S. E. 2d 670 (1994), and we denied certiorari, 513 U. S. 971 (1994). State collateral relief was subsequently denied as well.

Breard then filed a motion for habeas, relief under 28 U. S. C. §2254 in Federal District Court on August 20, 1996. In that motion, Breard argued for the first time that his convictions and sentences should be overturned because of alleged violations of the Vienna Convention on Consular Relations (Vienna Convention), April 24, 1963, [1970] 21 U. S. T. 77, T. I. A. S. No. 6820, at the time of his arrest. Specifically, Breard alleged that the Vienna Convention was violated when the arresting authorities failed to inform him that, as a foreign national, he had the right to contact the Paraguayan Consulate. The District Court rejected this claim, concluding that Breard procedurally defaulted the claim when he failed to raise it in state court and that Breard could not demonstrate cause and prejudice for this default. Breard v. Netherland, 949 F. Supp. 1255, 1266 (ED Va. 1996). The Fourth Circuit affirmed. Breard v. Pruett, 134 F. 3d 615, 620 (1998). Breard has petitioned this Court for a writ of certiorari.

*374In September 1996, the Republic of Paraguay, the Ambassador of Paraguay to the United States, and the Consul General of Paraguay to the United States (collectively Paraguay) brought suit in Federal District Court against certain Virginia officials, alleging that their separate rights under the Vienna Convention had been violated by the Commonwealth’s failure to inform Breard of his rights under the treaty and to inform the Paraguayan Consulate of Breard’s arrest, convictions, and sentences. In addition, the Consul General asserted a parallel claim under Rev. Stat. § 1979, 42 U. S. C. § 1983, alleging a denial of his rights under the Vienna Convention. The District Court concluded that it lacked subject-matter jurisdiction over these suits because Paraguay was not alleging a “continuing violation of federal law” and therefore could not bring its claims within the exception to Eleventh Amendment immunity established in Ex parte Young, 209 U. S. 123 (1908). Republic of Paraguay v. Allen, 949 F. Supp. 1269, 1272-1273 (ED Va. 1996). The Fourth Circuit affirmed on Eleventh Amendment grounds. Republic of Paraguay v. Allen, 134 F. 3d 622 (1998). Paraguay has also petitioned this Court for a writ of certiorari.

On April 3, 1998, nearly five years after Breard’s convictions became final, the Republic of Paraguay instituted proceedings against the United States in the International Court of Justice (ICJ), alleging that the United States violated the Vienna Convention at the time of Breard’s arrest. On April 9, the ICJ noted jurisdiction and issued an order requesting that the United States “take all measures at its disposal to ensure that Angel Francisco Breard is not executed pending the final decision in these proceedings . . . .” The ICJ set a briefing schedule for this matter, with oral argument likely to be held this November. Breard then filed a petition for an original writ of habeas corpus and a stay application in this Court in order to “enforce” the ICJ’s order. Paraguay filed a motion for leave to file a bill of complaint in this Court, citing this Court’s original jurisdiction *375over eases “affecting Ambassadors ... and Consuls.” U. S. Const., Art. III, §2.

It is clear that Breard procedurally defaulted his claim, if any, under the ‘Vienna Convention by failing to raise that claim in the state courts. Nevertheless, in their petitions for certiorari, both Breard and Paraguay contend that Breard’s Vienna Convention claim may be heard in federal court because the Convention is the “supreme law of the land” and thus trumps the procedural default doctrine. Pet. for Cert. in No. 97-8214, pp. 15-18; Pet. for Cert, in No. 97-1390, p. 14, n. 8. This argument is plainly incorrect for two reasons.

First, while we should give respectful consideration to the interpretation of an international treaty rendered by an international court with jurisdiction to interpret such, it has been recognized in international law that, absent a clear and express statement to the contrary, the procedural rules of the forum State govern the implementation of the treaty in that State. See Sun Oil Co. v. Wortman, 486 U. S. 717, 723 (1988); Volkswagenwerk Aktiengesellschaft v. Schlunk, 486 U. S. 694, 700 (1988); Société Nationale Industrielle Aérospatiale v. United States Dist. Court for Southern Dist. of Iowa, 482 U. S. 522, 539 (1987). This proposition is embodied in the Vienna Convention itself, which provides that the rights expressed in the Convention “shall be exercised in conformity with the laws and regulations of the receiving State,” provided that “said laws and regulations must enable full effect to be given to the purposes for which the rights accorded under this Article are intended.” Article 36(2), [1970] 21 U. S. T., at 101. It is the rule in this country that assertions- of error in criminal proceedings must first be raised in state court in order to form the basis for relief in habeas. Wainwright v. Sykes, 433 U. S. 72 (1977). Claims not so raised are considered defaulted. Ibid. By not asserting his Vienna Convention claim in state court, Breard failed to exercise his rights under the Vienna Convention *376in conformity with the laws of the United States and the Commonwealth of Virginia. Having failed to do so, he cannot raise a claim of violation of those rights now on federal habeas review.

Second, although treaties are recognized by our Constitution as the supreme law of the land, that status is no less true of provisions of the Constitution itself, to which rules of procedural default apply. We have held “that an Act of Congress ... is on a full parity with a treaty, and that when a statute which is subsequent in time is inconsistent with a treaty, the statute to the extent of conflict renders the treaty null.” Reid v. Covert, 354 U. S. 1, 18 (1957) (plurality opinion); see also Whitney v. Robertson, 124 U. S. 190, 194 (1888) (holding that if a treaty and a federal statute conflict, “the one last in date will control the other”). The Vienna Convention — which arguably confers on an individual the right to consular assistance following arrest — has continuously been in effect since 1969. But in 1996, before Breard filed his habeas petition raising claims under the Vienna Convention, Congress enacted the Antiterrorism and Effective Death Penalty Act (AEDPA), which provides that a habeas petitioner alleging that he is held in violation of “treaties of the United States” will, as a general rule, not be afforded an evidentiary hearing if he “has failed to develop the factual basis of [the] claim in State court proceedings.” 28 U. S. C. §§ 2254(a), (e)(2) (1994 ed., Supp. IV). Breard’s ability to obtain relief based on violations of the Vienna Convention is subject to this subsequently enacted rule, just as any claim arising under the United States Constitution would be. This rule prevents Breard from establishing that the violation of his Vienna Convention rights prejudiced him.' Without a hearing, Breard cannot establish how the Consul would have advised him, how the advice of his attorneys differed from the advice the Consul could have provided, and what factors he considered in electing to reject the plea bargain that the State offered him. That limitation, Breard also ar*377gues, is not justified because his Vienna Convention claims were so novel that he could not have discovered them any earlier. Assuming that were true, such novel claims would be barred on habeas review under Teague v. Lane, 489 U. S. 288 (1989).

Even were Breard’s Vienna Convention claim properly raised and proved, it is extremely doubtful that the violation should result in the overturning of a final judgment of conviction without some showing that the violation had an effect on the trial. Arizona v. Fulminante, 499 U. S. 279 (1991). In this action, no such showing could even arguably be made. Breard decided not to plead guilty and to testify at his own trial contrary to the advice of his attorneys, who were likely far better able to explain the United States legal system to him than any consular official would have been. Breard’s asserted prejudice — that had the Vienna Convention been followed, he would have accepted the State’s offer to forgo the death penalty in return for a plea of guilty— is far more speculative than the claims of prejudice courts routinely reject in those cases where an inmate alleges that his plea of guilty was infected by attorney error. See, e. g., Hill v. Lockhart, 474 U. S. 52, 59 (1985).

As for Paraguay’s suits (both the original action and the case coming to us on petition for certiorari), neither the text nor the history of the Vienna Convention clearly provides a foreign nation a private right of action in United States courts to set aside a criminal conviction and sentence for violation of consular notification provisions. The Eleventh Amendment provides a separate reason why Paraguay’s suit might not succeed. That Amendment’s “fundamental principle” that “the States, in the absence of consent, are immune from suits brought against them ... by a foreign State” was enunciated in Principality of Monaco v. Mississippi, 292 U. S. 313, 329-330 (1934). Though Paraguay claims that its suit is within an exemption dealing with continuing consequences of past violations of federal rights, see Milliken v. *378Bradley, 433 U. S. 267 (1977), we do not agree. The failure to notify the Paraguayan Consul occurred long ago and has no continuing effect. The causal link present in Milliken is absent in this suit.

Insofar as the Consul General seeks to base his claims on § 1983, his suit is not cognizable. Section 1983 provides a cause of action to any "person within the jurisdiction” of the United States for the deprivation "of any rights, privileges, or immunities secured by the Constitution and laws.” As an initial matter, it is clear that Paraguay is not authorized to bring suit under § 1983. Paraguay is not a “person” as that term is used in § 1983. See Moor v. County of Alameda, 411 U. S. 693, 699 (1973); South Carolina v. Katzenbach, 383 U. S. 301, 323-324 (1966); cf. Will v. Michigan Dept. of State Police, 491 U. S. 58 (1989). Nor is Paraguay “within the jurisdiction” of the United States. And since the Consul General is acting only in his official capacity, he has no greater ability to proceed under § 1983 than does the country he represents. Any rights that the Consul General might have by virtue of the Vienna Convention exist for the benefit of Paraguay, not for him as an individual.

It is unfortunate that this matter comes before us while proceedings are pending before the ICJ that might have been brought to that court earlier. Nonetheless, this Court must decide questions presented to it on the basis of law. The Executive Branch, on the other hand, in exercising its authority over foreign relations may, and in this case did, utilize diplomatic discussion with Paraguay. Last night the Secretary of State sent a letter to the Governor of Virginia requesting that he stay Breard’s execution. If the Governor wishes to wait for the decision of the ICJ, that is his prerogative. But nothing in our existing case law allows us to make that choice for him.

For the foregoing reasons, we deny the petition for an original writ of habeas corpus, the motion for leave to file a *379bill of complaint, the petitions for certiorari, and the accompanying stay applications filed by Breard and Paraguay.

Statement of Justice Soutee.

I agree with the Court that the lack of any reasonably arguable causal connection between the alleged treaty violations and Breard’s convictions and sentences disentitle him to relief on any theory offered. Moreover, I have substantial doubts that either Paraguay or any official acting for it is a “person” within the meaning of 42 U. S. C. § 1983 and that the Vienna Convention is enforceable in any judicial proceeding now underway. For these reasons, I believe the stay requests should be denied, with the result that Paraguay’s claims will be mooted. Accordingly, I have voted to deny Paraguay’s and Breard’s respective petitions for certio-rari (Nos. 97-1390 and 97-8214), Paraguay’s motion for leave to file a bill of complaint (No. 125, Orig.), Breard’s application for an original writ of habeas corpus (No. 97-8660), and the associated requests for a stay of execution.

Justice Stevens,

dissenting.

The Court of Appeals’ decision denying petitioner Breard’s first application for a federal writ of habeas corpus became final on February 18, 1998. Under this Court’s Rules, a timely petition for a writ of certiorari to review that decision could have been filed as late as May 19, 1998. See Rule 13.1 (“[A] petition for a writ of certiorari to review a judgment in any case, civil or criminal, entered by... a United States court of appeals ... is timely when it is filed with the Clerk of this Court within 90 days after entry of the judgment”). Ordinary review of that petition pursuant to our Rules would have given us additional time thereafter to consider its merits in the light of the response filed by the Commonwealth of Virginia. We have, however, been deprived of the normal time for considered deliberation by the Commonwealth’s decision to set the date of petitioner’s execution for today. *380There is no compelling reason for refusing to follow the procedures that we have adopted for the orderly disposition of noncapital cases. Indeed, the international aspects of this case provide an additional reason for adhering to our established Rules and procedures. I would therefore grant the applications for a stay, and I respectfully dissent from the decision to act hastily rather than with the deliberation that is appropriate in a case of this character.

Justice Ginsburg,

dissenting in No. 97-8214 (A-732).

I would grant the application for a stay of execution in order to consider in the ordinary course the instant petition, Breard’s first federal petition for writ of habeas corpus.

Justice Breyer,

dissenting.

In my view, several of the issues raised here are of sufficient difficulty to warrant less speedy consideration. Breard argues, for example, that the novelty of his Vienna Convention claim is sufficient to create “cause” for his having failed to present that claim to the Virginia state courts. Pet. for Cert. in No. 97-8214, pp. 20-22. He might add that the nature of his claim, were we to aecept it, is such as to create a “watershed rule of criminal procedure,” which might overcome the bar to consideration otherwise posed by Teague v. Lane, 489 U. S. 288, 311 (1989). He additionally says that what the Solicitor General describes as Virginia’s violation of the Convention “prejudiced” him by isolating him at a critical moment from Consular Officials who might have advised him to tiy to avoid the death penalty by pleading guilty. Pet. for Cert. in No. 97-8214, p. 22; see Brief for United States as Amicus Curiae in Nos. 97-1390 and 97-8214, p. 12 (“[T]he Executive Branch has conceded that the Vienna Convention was violated”). I cannot say, without examining the record more fully, that these arguments are obviously without merit. Nor am I willing to accept without fuller briefing and consideration the positions taken *381by the majority on all of the sometimes difficult issues that the majority addresses.

At the same time, the international aspects of the cases have provided us with the advantage of additional briefing even in the short time available. More time would likely mean additional briefing and argument, perhaps, for example, on the potential relevance of proceedings in an international forum.

Finally, as Justice Stevens points out, Virginia is now pursuing an execution schedule that leaves less time for argument and for Court consideration than the Court’s Rules provide for ordinary cases. Like Justice Stevens, I can find no special reason here to truncate the period of time that the Court’s Rules would otherwise make available.

For these reasons, taken together, I would grant the requested stay of execution and consider the petitions for certiorari in the ordinary course.

Justice Stevens

delivered the opinion of the Court.

Those mentally retarded persons who meet the law’s requirements for criminal responsibility should be tried and punished when they commit crimes. Because of their disabilities in areas of reasoning, judgment, and control of their impulses, however, they do not act with the level of moral culpability that characterizes the most serious adult criminal conduct. Moreover, their impairments can jeopardize the *307reliability and fairness of capital proceedings against mentally retarded defendants. Presumably for these reasons, in the 13 years since we decided Penry v. Lynaugh, 492 U. S. 302 (1989), the American public, legislators, scholars, and judges have deliberated over the question whether the death penalty should ever be imposed on a mentally retarded criminal. The consensus reflected in those deliberations informs our answer to the question presented by this case: whether such executions are “cruel and unusual punishments” prohibited by the Eighth Amendment to the Federal Constitution.

I

Petitioner, Daryl Renard Atkins, was convicted of abduction, armed robbery, and capital murder, and sentenced to death. At approximately midnight on August 16, 1996, Atkins and William Jones, armed with a semiautomatic handgun, abducted Eric Nesbitt, robbed him of the money on his person, drove him to an automated teller machine in his pickup truck where cameras recorded their withdrawal of additional cash, then took him to an isolated location where he was shot eight times and killed.

Jones and Atkins both testified in the guilt phase of Atkins’ trial.1 Each confirmed most of the details in the other’s account of the incident, with the important exception that each stated that the other had actually shot and killed Nesbitt. Jones’ testimony, which was both more coherent and credible than Atkins’, was obviously credited by the jury and was sufficient to establish Atkins’ guilt.2 At the penalty *308phase of the trial, the State introduced victim impact evidence and proved two aggravating circumstances: future dangerousness and “vileness of the offense.” To prove future dangerousness, the State relied on Atkins’ prior felony convictions as well as the testimony of four victims of earlier robberies and assaults. To prove the second aggravator, the prosecution relied upon the trial record, including pictures of the deceased’s body and the autopsy report.

In the penalty phase, the defense relied on one witness, Dr. Evan Nelson, a forensic psychologist who had evaluated Atkins before trial and concluded that he was “mildly mentally retarded.”3 His conclusion was based on interviews with people who knew Atkins,4 a review of school and court *309records, and the administration of a standard intelligence test which indicated that Atkins had a full scale IQ of 59.5

The jury sentenced Atkins to death, but the Virginia Supreme Court ordered a second sentencing hearing because the trial court had used a misleading verdict form. 257 Va. 160, 510 S. E. 2d 445 (1999). At the resentencing, Dr. Nelson again testified. The State presented an expert rebuttal witness, Dr. Stanton Samenow, who expressed the opinion that Atkins was not mentally retarded, but rather was of “average intelligence, at least,” and diagnosable as having antisocial personality disorder.6 App. 476. The jury again sentenced Atkins to death.

*310The Supreme Court of Virginia affirmed the imposition of the death penalty. 260 Va. 375, 385, 534 S. E. 2d 312, 318 (2000). Atkins did not argue before the Virginia Supreme Court that his sentence was disproportionate to penalties imposed for similar crimes in Virginia, but he did contend “that he is mentally retarded and thus cannot be sentenced to death.” Id., at 386, 534 S. E. 2d, at 318. The majority of the state court rejected this contention, relying on our holding in Penry. 260 Va., at 387, 534 S. E. 2d, at 319. The court was “not willing to commute Atkins’ sentence of death to life imprisonment merely because of his IQ score.” Id., at 390, 534 S. E. 2d, at 321.

Justice Hassell and Justice Koontz dissented. They rejected Dr. Samenow’s opinion that Atkins possesses average intelligence as “incredulous as a matter of law,” and concluded that “the imposition of the sentence of death upon a criminal defendant who has the mental age of a child between the ages of 9 and 12 is excessive.” Id., at 394, 395-396, 534 S. E. 2d, at 323-324. In their opinion, “it is indefensible to conclude that individuals who are mentally retarded are not to some degree less culpable for their criminal acts. By definition, such individuals have substantial limitations not shared by the general population. A moral and civilized society diminishes itself if its system of justice does not afford recognition and consideration of those limitations in a meaningful way.” Id., at 397, 534 S. E. 2d, at 325.

Because of the gravity of the concerns expressed by the dissenters, and in light of the dramatic shift in the state legislative landscape that has occurred in the past 13 years, we granted certiorari to revisit the issue that we first addressed in the Penry case. 533 U. S. 976 (2001).

*311II

The Eighth Amendment succinctly prohibits “[e]xcessive” sanctions. It provides: “Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.” In Weems v. United States, 217 U. S. 349 (1910), we held that a punishment of 12 years jailed in irons at hard and painful labor for the crime of falsifying records was excessive. We explained “that it is a precept of justice that punishment for crime should be graduated and proportioned to [the] offense.” Id., at 367. We have repeatedly applied this proportionality precept in later cases interpreting the Eighth Amendment. See Harmelin v. Michigan, 501 U. S. 957, 997-998 (1991) (Kennedy, J., concurring in part and concurring in judgment); see also id., at 1009-1011 (White, J., dissenting).7 Thus, even though “imprisonment for ninety days is not, in the abstract, a punishment which is either cruel or unusual,” it may not be imposed as a penalty for “the ‘status’ of narcotic addiction,” Robinson v. California, 370 U. S. 660, 666-667 (1962), because such a sanction would be excessive. As Justice Stewart explained in Robinson: “Even one day in prison would be a cruel and unusual punishment for the ‘crime’ of having a common cold.” Id., at 667.

A claim that punishment is excessive is judged not by the standards that prevailed in 1685 when Lord Jeffreys presided over the “Bloody Assizes” or when the Bill of Rights was adopted, but rather by those that currently prevail. As Chief Justice Warren explained in his opinion in Trop v. Dulles, 356 U. S. 86 (1958): “The basic concept underlying the Eighth Amendment is nothing less than the dignity of man. . . . The Amendment must draw its meaning from the *312evolving standards of decency that mark the progress of a maturing society.” Id., at 100-101.

Proportionality review under those evolving standards should be informed by “ ‘objective factors to the maximum possible extent,’ ” see Harmelin, 501 U. S., at 1000 (quoting Rummel v. Estelle, 445 U. S. 263, 274-275 (1980)). We have pinpointed that the “clearest and most reliable objective evidence of contemporary values is the legislation enacted by the country’s legislatures.” Penry, 492 U. S., at 331. Relying in part on such legislative evidence, we have held that death is an impermissibly excessive punishment for the rape of an adult woman, Coker v. Georgia, 433 U. S. 584, 593-596 (1977), or for a defendant who neither took life, attempted to take life, nor intended to take life, Enmund v. Florida, 458 U. S. 782, 789-793 (1982). In Coker, we focused primarily on the then-recent legislation that had been enacted in response to our decision 10 years earlier in Furman v. Georgia, 408 U. S. 238 (1972) (per curiam), to support the conclusion that the “current judgment,” though “not wholly unanimous,” weighed very heavily on the side of rejecting capital punishment as a “suitable penalty for raping an adult woman.” Coker, 433 U. S., at 596. The “current legislative judgment” relevant to our decision in Enmund was less clear than in Coker but “nevertheless weigh[ed] on the side of rejecting capital punishment for the crime at issue.” Enmund, 458 U. S., at 793.

We also acknowledged in Coker that the objective evidence, though of great importance, did not “wholly determine” the controversy, “for the Constitution contemplates that in the end our own judgment will be brought to bear on the question of the acceptability of the death penalty under the Eighth Amendment.” 433 U. S., at 597. For example, in Enmund, we concluded by expressing our own judgment about the issue:

“For purposes of imposing the death penalty, Enmund’s criminal culpability must be limited to his participation *313in the robbery, and his punishment must be tailored to his personal responsibility and moral guilt. Putting Enmund to death to avenge two killings that he did not commit and had no intention of committing or causing does not measurably contribute to the retributive end of ensuring that the criminal gets his just deserts. This is the judgment of most of the legislatures that have recently addressed the matter, and we have no reason to disagree with that judgment for purposes of construing and applying the Eighth Amendment.” 458 U. S., at 801 (emphasis added).

Thus, in cases involving a consensus, our own judgment is “brought to bear,” Coker, 438 U. S., at 597, by asking whether there is reason to disagree with the judgment reached by the citizenry and its legislators.

Guided by our approach in these cases, we shall first review the judgment of legislatures that have addressed the suitability of imposing the death penalty on the mentally retarded and then consider reasons for agreeing or disagreeing with their judgment.

III

The parties have not called our attention to any state legislative consideration of the suitability of imposing the death penalty on mentally retarded offenders prior to. 1986. In that year, the public reaction to the execution of a mentally retarded murderer in Georgia8 apparently led to the enact*314ment of the first state statute prohibiting such executions.9 In 1988, when Congress enacted legislation reinstating the federal death penalty, it expressly provided that a “sentence of death shall not be carried out upon a person who is mentally retarded.”10 In 1989, Maryland enacted a similar prohibition.11 It was in that year that we decided Penry, and concluded that those two state enactments, “even when added to the 14 States that have rejected capital punishment completely, do not provide sufficient evidence at present of a national consensus.” 492 U. S., at 334.

Much has changed since then. Responding to the national attention received by the Bowden execution and our decision in Penry, state legislatures across the country began to address the issue. In 1990, Kentucky and Tennessee enacted statutes similar to those in Georgia and Maryland, as did New Mexico in 1991, and Arkansas, Colorado, Washington, Indiana, and Kansas in 1993 and 1994.12 In 1995, when New York reinstated its death penalty, it emulated the Federal Government by expressly exempting the mentally retarded.13 Nebraska followed suit in 1998.14 There appear *315to have been no similar enactments during the next two years, but in 2000 and 2001 six more States — South Dakota, Arizona, Connecticut, Florida, Missouri, and North Carolina-joined the procession.15 The Texas Legislature unanimously adopted a similar bill,16 and bills have passed at least one house in other States, including Virginia and Nevada.17

It is not so much the number of these States that is significant, but the consistency of the direction of change.18 Given the well-known fact that anticrime legislation is far more popular than legislation providing protections for persons guilty of violent crime, the large number of States prohibiting the execution of mentally retarded persons (and the *316complete absence of States passing legislation reinstating the power to conduct such executions) provides powerful evidence that today our society views mentally retarded offenders as categorically less culpable than the average criminal. The evidence carries even greater force when it is noted that the legislatures that have addressed the issue have voted overwhelmingly in favor of the prohibition.19 Moreover, even in those States that allow the execution of mentally retarded offenders, the practice is uncommon. Some States, for example New Hampshire and New Jersey, continue to authorize executions, but none have been carried out in decades. Thus there is little need to pursue legislation barring the execution of the mentally retarded in those States. And it appears that even among those States that regularly execute offenders and that have no prohibition with regard to the mentally retarded, only five have executed offenders possessing a known IQ less than 70 since we decided Penry.20 The practice, therefore, has become truly unusual, and it is fair to say that a national consensus has developed against it.21

*317To the extent there is serious disagreement about the execution of mentally retarded offenders, it is in determining which offenders are in fact retarded. In this case, for instance, the Commonwealth of Virginia disputes that Atkins suffers from mental retardation. Not all people who claim to be mentally retarded will be so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus. As was our approach in Ford v. Wainwright, 477 U. S. 899 (1986), with regard to insanity, “we leave to the State[s] the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences.” Id., at 405, 416-417.22

IV

This consensus unquestionably reflects widespread judgment about the relative culpability of mentally retarded offenders, and the relationship between mental retardation and the penological purposes served by the death penalty. Additionally, it suggests that some characteristics of mental retardation undermine the strength of the procedural protections that our capital jurisprudence steadfastly guards.

*318As discussed above, clinical definitions of mental retardation require not only subaverage intellectual functioning, but also significant limitations in adaptive skills such as communication, self-care, and self-direction that became manifest before age 18. Mentally retarded persons frequently know the difference between right and wrong and are competent to stand trial. Because of their impairments, however, by definition they have diminished capacities to understand and process information, to communicate, to abstract from mistakes and learn from experience, to engage in logical reasoning, to control impulses, and to understand the reactions of others.23 There is no evidence that they are more likely to engage in criminal conduct than others, but there is abundant evidence that they often act on impulse rather than pursuant to a premeditated plan, and that in group settings they are followers rather than leaders.24 Their deficiencies do not warrant an exemption from criminal sanctions, but they do diminish their personal culpability.

In light of these deficiencies, our death penalty jurisprudence provides two reasons consistent with the legislative consensus that the mentally retarded should be categorically excluded from execution. First, there is a serious question as to whether either justification that we have recognized as *319a basis for the death penalty applies to mentally retarded offenders. Gregg v. Georgia, 428 U. S. 153, 183 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.), identified “retribution and deterrence of capital crimes by prospective offenders” as the social purposes served by the death penalty. Unless the imposition of the death penalty on a mentally retarded person “measurably contributes to one or both of these goals, it ‘is nothing more than the purposeless and needless imposition of pain and suffering,’ and hence an unconstitutional punishment.” Enmund, 458 U. S., at 798.

With respect to retribution — the interest in seeing that the offender gets his “just deserts” — the severity of the appropriate punishment necessarily depends on the culpability of the offender. Since Gregg, our jurisprudence has consistently confined the imposition of the death penalty to a narrow category of the most serious crimes. For example, in Godfrey v. Georgia, 446 U. S. 420 (1980), we set aside a death sentence because the petitioner’s crimes did not reflect “a consciousness materially more ‘depraved’ than that of any person guilty of murder.” Id., at 433. If the culpability of the average murderer is insufficient to justify the most extreme sanction available to the State, the lesser culpability of the mentally retarded offender surely does not merit that form of retribution. Thus, pursuant to our narrowing jurisprudence, which seeks to ensure that only the most deserving of execution are put to death, an exclusion for the mentally retarded is appropriate.

With respect to deterrence — the interest in preventing capital crimes by prospective offenders — “it seems likely that ‘capital punishment can serve as a deterrent only when murder is the result of premeditation and deliberation,’ ” Enmund, 458 U. S., at 799. Exempting the mentally retarded from that punishment will not affect the “cold calculus that precedes the decision” of other potential murderers. Gregg, 428 U. S., at 186, Indeed, that sort of calculus is at the opposite end of the spectrum from behavior of mentally retarded *320offenders. The theory of deterrence in capital sentencing is predicated upon the notion that the increased severity of the punishment will inhibit criminal actors from carrying out murderous conduct. Yet it is the same cognitive and behavioral impairments that make these defendants less morally culpable — for example, the diminished ability to understand and process information, to learn from experience, to engage in logical reasoning, or to control impulses — that also make it less likely that they can process the information of the possibility of execution as a penalty and, as a result, control their conduct based upon that information. Nor will exempting the mentally retarded from execution lessen the deterrent effect of the death penalty with respect to offenders who are not mentally retarded. Such individuals are unprotected by the exemption and will continue to face the threat of execution. Thus, executing the mentally retarded will not measurably further the goal of deterrence.

The reduced capacity of mentally retarded offenders provides a second justification for a categorical rule making such offenders ineligible for the death penalty. The risk “that the death penalty will be imposed in spite of factors which may call for a less severe penalty,” Lockett v. Ohio, 438 U. S. 586, 605 (1978), is enhanced, not only by the possibility of false confessions,25 but also by the lesser ability of mentally retarded defendants to make a persuasive showing of mitigation in the face of prosecutorial evidence of one or more aggravating factors. Mentally retarded defendants may be less able to give meaningful assistance to their counsel and *321are typically poor witnesses, and their demeanor may create an unwarranted impression of lack of remorse for their crimes. As Penry demonstrated, moreover, reliance on mental retardation as a mitigating factor can be a two-edged sword that may enhance the likelihood that the aggravating factor of future dangerousness will be found by the jury. 492 U. S., at 323-325. Mentally retarded defendants in the aggregate face a special risk of wrongful execution.

Our independent evaluation of the issue reveals no reason to disagree with the judgment of “the legislatures that have recently addressed the matter” and concluded that death is not a suitable punishment for a mentally retarded criminal. We are not persuaded that the execution of mentally retarded criminals will measurably advance the deterrent or the retributive purpose of the death penalty. Construing and applying the Eighth Amendment in the light of our “evolving standards of decency,” we therefore conclude that such punishment is excessive and that the Constitution “places a substantive restriction on the State’s power to take the life” of a mentally retarded offender. Ford, 477 U. S., at 405.

The judgment of the Virginia Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion.

It is so ordered.

Chief Justice Rehnquist,

with whom Justice Scalia and Justice Thomas join, dissenting.

The question presented by this case is whether a national consensus deprives Virginia of the constitutional power to impose the death penalty on capital murder defendants like petitioner, i. e., those defendants who indisputably are competent to stand trial, aware of the punishment they are about to suffer and why, and whose mental retardation has been found an insufficiently compelling reason to lessen their individual responsibility for the crime. The Court pronounces *322the punishment cruel and unusual primarily because 18 States recently have passed laws limiting the death eligibility of certain defendants based on mental retardation alone, despite the fact that the laws of 19 other States besides Virginia continue to leave the question of proper punishment to the individuated consideration of sentencing judges or juries familiar with the particular offender and his or her crime. See ante, at 314-315.

I agree with Justice Scalia, post, at 337-338 (dissenting opinion), that the Court’s assessment of the current legislative judgment regarding the execution of defendants like petitioner more resembles a post hoc rationalization for the majority’s subjectively preferred result rather than any objective effort to ascertain the content of an evolving standard of decency. I write separately, however, to call attention to the defects in the Court’s decision to place weight on foreign laws, the views of professional and religious organizations, and opinion polls in reaching its conclusion. See ante, at 316-317, n. 21. The Court’s suggestion that these sources are relevant to the constitutional question finds little support in our precedents and, in my view, is antithetical to considerations of federalism, which instruct that any “permanent prohibition upon all units of democratic government must [be apparent] in the operative acts (laws and the application of laws) that the people have approved.” Stanford v. Kentucky, 492 U. S. 361, 377 (1989) (plurality opinion). The Court’s uncritical acceptance of the opinion poll data brought to our attention, moreover, warrants additional comment, because we lack sufficient information to conclude that the surveys were conducted in accordance with generally accepted scientific principles or are capable of supporting valid empirical inferences about the issue before us.

In making determinations about whether a punishment is “cruel and unusual” under the evolving standards of decency embraced by the Eighth Amendment, we have emphasized that legislation is the “clearest and most reliable objective *323evidence of contemporary values.” Penry v. Lynaugh, 492 U. S. 302, 331 (1989). See also McCleskey v. Kemp, 481 U. S. 279, 300 (1987). The reason we ascribe primacy to legislative enactments follows from the constitutional role legislatures play in expressing policy of a State. “‘[I]n a democratic society legislatures, not courts, are constituted to respond to the will and consequently the moral values of the people.’ ” Gregg v. Georgia, 428 U. S. 153, 175-176 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.) (quoting Furman v. Georgia, 408 U. S. 238, 383 (1972) (Burger, C. J., dissenting)). And because the specifications of punishments are “peculiarly questions of legislative policy,” Gore v. United States, 357 U. S. 386, 393 (1958), our cases have cautioned against using “The aegis of the Cruel and Unusual Punishment Clause’ ” to cut off the normal democratic processes, Gregg, supra, at 176 (quoting Powell v. Texas, 392 U. S. 514, 533 (1968) (plurality opinion)).

Our opinions have also recognized that data concerning the actions of sentencing juries, though entitled to less weight than legislative judgments, “ ‘is a significant and reliable objective index of contemporary values,’” Coker v. Georgia, 433 U. S. 584, 596 (1977) (plurality opinion) (quoting Gregg, supra, at 181), because of the jury’s intimate involvement in the case and its function of “ ‘maintaining] a link between contemporary community values and the penal system,’ ” Gregg, supra, at 181 (quoting Witherspoon v. Illinois, 391 U. S. 510, 519, n. 15 (1968)). In Coker, supra, at 596-597, for example, we credited data showing that “at least 9 out of 10” juries in Georgia did not impose the death sentence for rape convictions. And in Enmund v. Florida, 458 U. S. 782, 793-794 (1982), where evidence of the current legislative judgment was not as “compelling” as that in Coker (but more so than that here), we were persuaded by “overwhelming [evidence] that American juries ... repudiated imposition of the death penalty” for a defendant who neither took life nor attempted or intended to take life.

*324In my view, these two sources — the work product of legislatures and sentencing jury determinations — ought to be the sole indicators by which courts ascertain the contemporary American conceptions of decency for purposes of the Eighth Amendment. They are the only objective indicia of contemporary values firmly supported by our precedents. More importantly, however, they can be reconciled with the undeniable precepts that the democratic branches of government and individual sentencing juries are, by design, better suited than courts to evaluating and giving effect to the complex societal and moral considerations that inform the selection of publicly acceptable criminal punishments.

In reaching its conclusion today, the Court does not take notice of the fact that neither petitioner nor his amici have adduced any comprehensive statistics that would conclusively prove (or disprove) whether juries routinely consider death a disproportionate punishment for mentally retarded offenders like petitioner.* Instead, it adverts to the fact that other countries have disapproved imposition of the death penalty for crimes committed by mentally retarded offenders, see ante, at 316-317, n. 21 (citing the Brief for European Union as Amicus Curiae 2). I fail to see, how*325ever, how the views of other countries regarding the punishment of their citizens provide any support for the Court’s ultimate determination. While it is true that some of our prior opinions have looked to “the climate of international opinion,” Coker, supra, at 596, n. 10, to reinforce a conclusion regarding evolving standards of decency, see Thompson v. Oklahoma, 487 U. S. 815, 830 (1988) (plurality opinion); Enmund, supra, at 796-797, n. 22; Trop v. Dulles, 356 U. S. 86, 102-103 (1958) (plurality opinion); we have since explicitly rejected the idea that the sentencing practices of other countries could “serve to establish the first Eighth Amendment prerequisite, that [a] practice is accepted among our people.” Stanford, 492 U. S., at 369, n. 1 (emphasizing that “American conceptions of decency . . . are dispositive” (emphasis in original)).

Stanford’s reasoning makes perfectly good sense, and the Court offers no basis to question it. For if it is evidence of a national consensus for which we are looking, then the viewpoints of other countries simply are not relevant. And nothing in Thompson, Enmund, Coker, or Trop suggests otherwise. Thompson, Enmund, and Coker rely only on the bare citation of international laws by the Trop plurality as authority to deem other countries’ sentencing choices germane. But the Trop plurality — representing the view of only a minority of the Court — offered no explanation for its own citation, and there is no reason to resurrect this view given our sound rejection of the argument in Stanford.

To further buttress its appraisal of contemporary societal values, the Court marshals public opinion poll results and evidence that several professional organizations and religious groups have adopted official positions opposing the imposition of the death penalty upon mentally retarded offenders. See ante, at 316-317, n. 21 (citing Brief for American Psychological Association et al. as Amici Curiae; Brief for American Association on Mental Retardation et al. as Amici Curiae; noting that “representatives of widely diverse reli*326gious communities . .. reflecting Christian, Jewish, Muslim, and Buddhist traditions ... ‘share a conviction that the execution of persons with mental retardation cannot be morally justified’ and stating that “polling data shows a widespread consensus among Americans ... that executing the mentally retarded is wrong”). In my view, none should be accorded any weight on the Eighth Amendment scale when the elected representatives of a State’s populace have not deemed them persuasive enough to prompt legislative action. In Penry, 492 U. S., at 334-335, we were cited similar data and declined to take them into consideration where the “public sentiment expressed in [them]” had yet to find expression in state law. See also Stanford, 492 U. S., at 377 (plurality opinion) (refusing “the invitation to rest constitutional law upon such uncertain foundations” as “public opinion polls, the views of interest groups, and the positions adopted by various professional associations”). For the Court to rely on such data today serves only to illustrate its willingness to proscribe by judicial fiat — at the behest of private organizations speaking only for themselves — a punishment about which no across-the-board consensus has developed through the workings of normal democratic processes in the laboratories of the States.

Even if I were to accept the legitimacy of the Court’s decision to reach beyond the product of legislatures and practices of sentencing juries to discern a national standard of decency, I would take issue with the blind-faith credence it accords the opinion polls brought to our attention. An extensive body of social science literature describes how methodological and other errors can affect the reliability and validity of estimates about the opinions and attitudes of a population derived from various sampling techniques. Everything from variations in the survey methodology, such as the choice of the target population, the sampling design used, the questions asked, and the statistical analyses used to interpret the data can skew the results. See, e. g., R. Groves, Survey *327Errors and Survey Costs (1989); 1 C. Turner & E. Martin, Surveying Subjective Phenomena (1984).

The Federal Judicial Center’s Reference Manual on Scientific Evidence 221-271 (1994) and its Manual for Complex Litigation § 21.493, pp. 101-103 (3d ed. 1995), offer helpful suggestions to judges called upon to assess the weight and admissibility of survey evidence on a factual issue before a court. Looking at the polling data (reproduced in the Appendix to this opinion) in light of these factors, one cannot help but observe how unlikely it is that the data could support a valid inference about the question presented by this case. For example, the questions reported to have been asked in the various polls do not appear designed to gauge whether the respondents might find the death penalty an acceptable punishment for mentally retarded offenders in rare cases. Most are categorical (e. g., “Do you think that persons convicted of murder who are mentally retarded should or should not receive the death penalty?”), and, as such, would not elicit whether the respondent might agree or disagree that all mentally retarded people by definition can never act with the level of culpability associated with the death penalty, régardless of the severity of their impairment or the individual circumstances of their crime. Second, none of the 27 polls cited disclose the targeted survey population or the sampling techniques used by those who conducted the research. Thus, even if one accepts that the survey instruments were adequately designed to address a relevant question, it is impossible to know whether the sample was representative enough or the methodology sufficiently sound to tell us anything about the opinions of the citizens of a particular State or the American public at large. Finally, the information provided to us does not indicate why a particular survey was conducted or, in a few cases, by whom, factors which also can bear on the objectivity of the results. In order to be credited here, such surveys should be offered as *328evidence at trial, where their sponsors can be examined and cross-examined about these matters.

* * *

There are strong reasons for limiting our inquiry into what constitutes an evolving standard of decency under the Eighth Amendment to the laws passed by legislatures and the practices of sentencing juries in America. Here, the Court goes beyond these well-established objective indicators of contemporary values. It finds “further support to [its] conclusion” that a national consensus has developed against imposing the death penalty on all mentally retarded defendants in international opinion, the views of professional and religious organizations, and opinion polls not demonstrated to be reliable. Ante, at 817, n. 21. Believing this view to be seriously mistaken, I dissent.

APPENDIX TO OPINION OF REHNQUIST, C. J.

Poll and survey results reported in Brief for American Association on Mental Retardation et al. as Amici Curiae 3a-7a, and cited by the Court, ante, at 317, n. 21:

*329

*330

*331

*332

*333

*334

*335

*336

*337

Justice Scalia,

with whom The Chief Justice and Justice Thomas join, dissenting.

Today’s decision fe the pinnacle of our Eighth Amendment death-is-different jurisprudence. Not only does it, like all of that jurisprudence, find no support in the text or history of the Eighth Amendment; it does not even have support in current social attitudes regarding the conditions that render *338an otherwise just death penalty inappropriate. Seldom has an opinion of this Court rested so obviously upon nothing but the personal views of its Members.

H-I

I begin with a brief restatement of facts that are abridged by the Court but important to understanding this case. After spending the day drinking alcohol and smoking marijuana, petitioner Daryl Renard Atkins and a partner in crime drove to a convenience store, intending to rob a customer. Their victim was Eric Nesbitt, an airman from Langley Air Force Base, whom they abducted, drove to a nearby automated teller machine, and forced to withdraw $200. They then drove him to a deserted area, ignoring his pleas to leave him unharmed. According to the co-conspirator, whose testimony the jury evidently credited, Atkins ordered Nesbitt out of the vehicle and, after he had taken only a few steps, shot him one, two, three, four, five, six, seven, eight times in the thorax, chest, abdomen, arms, and legs.

The jury convicted Atkins of capital murder. At resen-tencing (the Virginia Supreme Court affirmed his conviction but remanded for resentencing because the trial court had used an improper verdict form, 257 Va. 160, 179, 510 S. E. 2d 445, 457 (1999)), the jury heard extensive evidence of petitioner’s alleged mental retardation. A psychologist testified that petitioner was mildly mentally retarded with an IQ of 59, that he was a “slow learner,” App. 444, who showed a “lack of success in pretty much every domain of his life,” id., at 442, and that he had an “impaired” capacity to appreciate the criminality of his conduct and to conform his conduct to the law, id., at 453. Petitioner’s family members offered additional evidence in support of his mental retardation claim (e. g., that petitioner is a “follower,” id., at 421). The Commonwealth contested the evidence of retardation and presented testimony of a psychologist who found “absolutely no evidence other than the IQ score ... indicating that [peti*339tioner] was in the least bit mentally retarded” and concluded that petitioner was “of average intelligence, at least.” Id., at 476.

The jury also heard testimony about petitioner’s 16 prior felony convictions for robbery, attempted robbery, abduction, use of a firearm, and maiming. Id., at 491-522. The victims of these offenses provided graphic depictions of petitioner’s violent tendencies: He hit one over the head with a beer bottle, id., at 406; he slapped a gun across another victim’s face, clubbed her in the head with it, knocked her to the ground, and then helped her up, only to shoot her in the stomach, id., at 411-413. The jury sentenced petitioner to death. The Supreme Court of Virginia affirmed petitioner’s sentence. 260 Va. 375, 534 S. E. 2d 312 (2000).

II

As the foregoing history demonstrates, petitioner’s mental retardation was a central issue at sentencing. The jury concluded, however, that his alleged retardation was not a compelling reason to exempt him from the death penalty in light of the brutality of his crime and his long demonstrated propensity for violence. “In upsetting this particularized judgment on the basis of a constitutional absolute,” the Court concludes that no one who is even slightly mentally retarded can have sufficient “moral responsibility to be subjected to capital punishment for any crime. As a sociological and moral conclusion that is implausible; and it is doubly implausible as an interpretation of the United States Constitution.” Thompson v. Oklahoma, 487 U. S. 815, 863-864 (1988) (Scalia, J., dissenting).

Under our Eighth Amendment jurisprudence, a punishment is “cruel and unusual” if it falls within one of two categories: “those modes or acts of punishment that had been considered cruel and unusual at the time that the Bill of Rights was adopted,” Ford v. Wainwright, 477 U. S. 399, 405 (1986), and modes of punishment that are inconsistent with *340modern “‘standards of decency/” as evinced by objective indicia, the most important of which is “legislation enacted by the country’s legislatures,” Penry v. Lynaugh, 492 U. S. 302, 330-331 (1989).

The Court makes no pretense that execution of the mildly mentally retarded would have been considered “cruel and unusual” in 1791. Only the severely or profoundly mentally retarded, commonly known as “idiots,” enjoyed any special status under the law at that time. They, like lunatics, suffered a “deficiency in will” rendering them unable to tell right from wrong. 4 W. Blackstone, Commentaries on the Laws of England 24 (1769) (hereinafter Blackstone); see also Penry, 492 U. S., at 331-332 (“[T]he term ‘idiot’ was generally used to describe persons who had a total lack of reason or understanding, or an inability to distinguish between good and evil”); id., at 333 (citing sources indicating that idiots generally had an IQ of 25 or below, which would place them within the “profound” or “severe” range of mental retardation under modern standards); 2 A. Fitz-Herbert, Natura Brevium 233B (9th ed. 1794) (originally published 1534) (An idiot is “such a person who cannot account or number twenty pence, nor can tell who was his father or mother, nor how old he is, etc., so as it may appear that he hath no understanding of reason what shall be for his profit, or what for his loss”). Due to their incompetence, idiots were “ex-euse[d] from the guilt, and of course from the punishment, of any criminal action committed under such deprivation of the senses.” 4 Blackstone 25; see also Penry, supra, at 331. Instead, they were often committed to civil confinement or made wards of the State, thereby preventing them from “go[ing] loose, to the terror of the king’s subjects.” 4 Blackstone 25; see also S. Brakel, J. Parry, & B. Weiner, The Mentally Disabled and the Law 12-14 (3d ed. 1985); 1 Blackstone 292-296; 1 M. Hale, Pleas of the Crown 33 (1st Am. ed. 1847). Mentally retarded offenders with less severe impairments— those who were not “idiots” — suffered criminal prosecution *341and punishment, including capital punishment. See, e. g., I. Ray, Medical Jurisprudence of Insanity 65, 87-92 (W. Over-holser ed. 1962) (recounting the 1834 trial and execution in Concord, New Hampshire, of an apparent “imbecile” — imbecility being a less severe form of retardation which “differs from idiocy in the circumstance that while in [the idiot] there is an utter destitution of every thing like reason, [imbeciles] possess some intellectual capacity, though infinitely less than is possessed by the great mass of mankind”); A. Highmore, Law of Idiocy and Lunacy 200 (1807) (“The great difficulty in all these cases, is to determine where a person shall be said to be so far deprived of his sense and memory as not to have any of his actions imputed to him: or where notwithstanding some defects of this kind he still appears to have so much reason and understanding as will make him accountable for his actions ...”).

The Court is left to argue, therefore, that execution of the mildly retarded is inconsistent with the “evolving standards of decency that mark the progress of a maturing society.” Trop v. Dulles, 356 U. S. 86, 101 (1958) (plurality opinion) (Warren, C. J.). Before today, our opinions consistently emphasized that Eighth Amendment judgments regarding the existence of social “standards” “should be informed by objective factors to the maximum possible extent” and “should not be, or appear to be, merely the subjective views of individual Justices.” Coker v. Georgia, 433 U. S. 584, 592 (1977) (plurality opinion); see also Stanford v. Kentucky, 492 U. S. 361, 369 (1989); McCleskey v. Kemp, 481 U. S. 279, 300 (1987); Enmund v. Florida, 458 U. S. 782, 788 (1982). “First” among these objective factors are the “statutes passed by society’s elected representatives,” Stanford, supra, at 370; because it “will rarely if ever be the case that the Members of this Court will have a better sense of the evolution in views of the American people than do their elected representatives,” Thompson, supra, at 865 (Scalia, J., dissenting).

*342The Court pays lipservice to these precedents as it miraculously extracts a “national consensus” forbidding execution of the mentally retarded, ante, at 316, from the fact that 18 States — less than half (47%) of the 38 States that permit capital punishment (for whom the issue exists) — have very recently enacted legislation barring execution of the mentally retarded. Even that 47% figure is a distorted one. If one is to say, as the Court does today, that all executions of the mentally retarded are so morally repugnant as to violate our national “standards of decency,” surely the “consensus” it points to must be one that has set its righteous face against all such executions. Not 18 States, but only 7 — 18% of death penalty jurisdictions — have legislation of that scope. Eleven of those that the Court counts enacted statutes prohibiting execution of mentally retarded defendants convicted after, or convicted of crimes committed after, the effective date of the legislation;1 those already on death row, or consigned there before the statute’s effective date, or even (in those States using the date of the crime as the criterion of retroactivity) tried in the future for murders committed many years ago, could be put to death. That is not a statement of absolute moral repugnance, but one of current preference between two tolerable approaches. Two of these States permit execution of the mentally retarded in other situations as well: Kansas apparently permits execution of all *343except the severely mentally retarded;2 New York permits execution of the mentally retarded who commit murder in a correctional facility. N. Y. Crim. Proc. Law § 400.27.12(d) (McKinney 2001); N. Y. Penal Law § 125.27 (McKinney 2002).

But let us accept, for the sake of argument, the Court’s faulty count. That bare number of States alone — 18— should be enough to convince any reasonable person that no “national consensus” exists. How is it possible that agreement among 47% of the death penalty jurisdictions amounts to “consensus”? Our prior cases have generally required a much higher degree of agreement before finding a punishment cruel and unusual on “evolving standards” grounds. In Coker, supra, at 595-596, we proscribed the death penalty for rape of an adult woman after finding that only one jurisdiction, Georgia, authorized such a punishment. In En-mund, supra, at 789, we invalidated the death penalty for mere participation in a robbery in which an accomplice took a life, a punishment not permitted in 28 of the death penalty States (78%). In Ford, 477 U. S., at 408, we supported the common-law prohibition of execution of the insane with the observation that “[t]his ancestral legacy has not outlived its time,” since not a single State authorizes such punishment. In Solem, v. Helm, 468 U. S. 277, 300 (1983), we invalidated a life sentence without parole under a recidivist statute by which the criminal “was treated more severely than he would have been in any other State.” What the Court calls evidence of “consensus” in the present case (a fudged 47%) more closely resembles evidence that we found inadequate *344to establish consensus in earlier cases. Tison v. Arizona, 481 U. S. 137, 154, 158 (1987), upheld a state law authorizing capital punishment for major participation in a felony with reckless indifference to life where only 11 of the 37 death penalty States (30%) prohibited such punishment. Stanford, 492 U. S., at 372, upheld a state law permitting execution of defendants who committed a capital crime at age 16 where only 15 of the 36 death penalty States (42%) prohibited death for such offenders.

Moreover, a major factor that the Court entirely disregards is that the legislation of all 18 States it relies on is still in its infancy. The oldest of the statutes is only 14 years old;3 five were enacted last year;4 over half were enacted within the past eight years.5 Few, if any, of the States have had sufficient experience with these laws to know whether they are sensible in the long term. It is “myopic to base sweeping constitutional principles upon the narrow experience of [a few] years.” Coker, 433 U. S., at 614 (Burger, C. J., dissenting); see also Thompson, 487 U. S., at 854-855 (O’Connor, J., concurring in judgment).

The Court attempts to bolster its embarrassingly feeble evidence of “consensus” with the following: “It is not so much the number of these States that is significant, but the consistency of the direction of change.” Ante, at 315 (emphasis added). But in what other direction could we possibly see change? Given that 14 years ago all the death penalty statutes included the mentally retarded, any change (except precipitate undoing of what had just been done) was bound *345to be in the one direction the Court finds significant enough to overcome the lack of real consensus. That is to say, to be accurate the Court’s “consistency-of-the-direction-of-change” point should be recast into the following unimpressive observation: “No State has yet undone its exemption of the mentally retarded, one for as long as 14 whole years.” In any event, reliance upon “trends,” even those of much longer duration than a mere 14 years, is a perilous basis for constitutional adjudication, as Justice O’Connor eloquently explained in Thompson:

“In 1846, Michigan became the first State to abolish the death penalty.... In succeeding decades, other American States continued the trend towards abolition .... Later, and particularly after World War II, there ensued a steady and dramatic decline in executions .... In the 1950’s and 1960’s, more States abolished or radically restricted capital punishment, and executions ceased completely for several years beginning in 1968....

“In 1972, when this Court heard arguments on the constitutionality of the death penalty, such statistics might have suggested that the practice had become a relic, implicitly rejected by a new societal consensus. ... We now know that any inference of a societal consensus rejecting the death penalty would have been mistaken. But had this Court then declared the existence of such a consensus, and outlawed capital- punishment, legislatures would very likely not have been able to revive it. The mistaken premise of the decision would have been frozen into constitutional law, making it difficult to refute and even more difficult to reject.” 487 U. S., at 854-855.

Her words demonstrate, of course, not merely the peril of riding a trend, but also the peril of discerning a consensus where there is none.

*346The Court’s thrashing about for evidence of “consensus” includes reliance upon the margins by which state legislatures have enacted bans on execution of the retarded. Ante, at 316. Presumably, in applying our Eighth Amendment “evolving-standards-of-decency” jurisprudence, we will henceforth weigh not only how many States have agreed, but how many States have agreed by how much. Of course if the percentage of legislators voting for the bill is significant, surely the number of people represented by the legislators voting for the bill is also significant: the fact that 49% of the legislators in a State with a population of 60 million voted against the bill should be more impressive than the fact that 90% of the legislators in a State with a population of 2 million voted for it. (By the way, the population of the death penalty States that exclude the mentally retarded is only 44% of the population of all death penalty States. U. S. Dept, of Commerce, Bureau of Census, Statistical Abstract of the United States 21 (121st ed. 2001).) This is quite absurd. What we have looked for in the past to “evolve” the Eighth Amendment is a consensus of the same sort as the consensus that adopted the Eighth Amendment: a consensus of the sovereign States that form the Union, not a nose count of Americans for and against.

Even less compelling (if possible) is the Court’s argument, ante, at 316, that evidence of “national consensus” is to be found in the infrequency with which retarded persons are executed in States that do not bar their execution. To begin with, what the Court takes as true is in fact quite doubtful. It is not at all clear that execution of the mentally retarded is “uncommon,” ibid., as even the sources cited by the Court suggest, see ante, at 316, n. 20 (citing D. Keyes, W. Edwards, & R. Perske, People with Mental Retardation are Dying Legally, 35 Mental Retardation (Feb. 1997) (updated by Death Penalty Information Center, available at http://www.advocacyone.org/deathpenalty.html (as visited *347June 12, 2002) (showing that 12 States executed 35 allegedly mentally retarded offenders during the period 1984-2000)). See also Bonner & Rimer, Executing the Mentally Retarded Even as Laws Begin to Shift, N. Y. Times, Aug. 7,2000, p. Al (reporting that 10% of death row inmates are retarded). If however, execution of the mentally retarded is “uncommon”; and if it is not a sufficient explanation of this that the retarded constitute a tiny fraction of society (1% to 3%), Brief for American Psychological Association et al. as Amici Curiae 7; then surely the explanation is that mental retardation is a constitutionally mandated mitigating factor at sentencing, Penry, 492 U. S., at 328. For that reason, even if there were uniform national sentiment in favor of executing the retarded in appropriate cases, one would still expect execution of the mentally retarded to be “uncommon.” To adapt to the present case what the Court itself said in Stanford, 492 U. S., at 374: “[I]t is not only possible, but overwhelmingly probable, that the very considerations which induce [today’s majority] to believe that death should never be imposed on [mentally retarded] offenders . . . cause prosecutors and juries to believe that it should rarely be imposed.”

But the Prize for the Court’s Most Feeble Effort to fabricate “national consensus” must go to its appeal (deservedly relegated to a footnote) to the views of assorted professional and religious organizations, members of the so-called “world community,” and respondents to opinion polls. Ante, at 316-317, n. 21. I agree with The Chief Justice, ante, at 325-328 (dissenting opinion), that the views of professional and religious organizations and the results of opinion polls are irrelevant.6 Equally irrelevant are the practices of the *348“world community,” whose notions of justice are (thankfully) not always those of our people. “We must never forget that it is a Constitution for the United States of America that we are expounding. .. . [W]here there is not first a settled consensus among our own people, the views of other nations, however enlightened the Justices of this Court may think them to be, cannot be imposed upon Americans through the Constitution.” Thompson, 487 U. S., at 868-869, n. 4 (Scalia, J., dissenting).

Ill

Beyond the empty talk of a “national consensus,” the Court gives us a brief glimpse of what really underlies today’s decision: pretension to a power confined neither by the moral sentiments originally enshrined in the Eighth Amendment (its original meaning) nor even by the current moral sentiments of the American people. “ ‘[T]he Constitution,’ ” the Court says, “contemplates that in the end our own judgment will be brought to bear on the question of the acceptability of the death penalty under the Eighth Amendment.’ ” Ante, at 312 (quoting Coker, 433 U. S., at 697) (emphasis added). (The unexpressed reason for this unexpressed “contemplation” of the Constitution is presumably that really good lawyers have moral sentiments superior to those of the common herd, whether in 1791 or today.) The arrogance of this assumption of power takes one’s breath away. And it explains, of course, why the Court can be so cavalier about the evidence of consensus. It is just a game, after all. “‘[I]n the end,’” Thompson, supra, at 823, n. 8 (plurality opinion (quoting Coker, supra, at 697 (plurality opinion))), it is the feelings and intuition of a majority of the Justices that count — “the perceptions of decency, or of penology, or of mercy, entertained ... by a majority of the small and *349unrepresentative segment of our society that sits on this Court.” Thompson, supra, at 873 (Scalia, J., dissenting).

The genuinely operative portion of the opinion, then, is the Court’s statement of the reasons why it agrees with the contrived consensus it has found, that the “diminished capacities” of the mentally retarded render the death penalty excessive. Ante, at 317-321. The Court’s analysis rests on two fundamental assumptions: (1) that the Eighth Amendment prohibits excessive punishments, and (2) that sentencing juries or judges are unable to account properly for the “diminished capacities” of the retarded. The first assumption is wrong, as I explained at length in Harmelin v. Michigan, 501 U. S. 957, 966-990 (1991) (opinion of Scalia, J.). The Eighth Amendment is addressed to always-and-everywhere “cruel” punishments, such as the rack and the thumbscrew. But where the punishment is in itself permissible, “[t]he Eighth Amendment is not a ratchet, whereby a temporary consensus on leniency for a particular crime fixes a permanent constitutional maximum, disabling the States from giving effect to altered beliefs and responding to changed social conditions.” Id., at 990. The second assumption — inability of judges or juries to take proper account of mental retardation — is not only unsubstantiated, but contradicts the immemorial belief, here and in England, that they play an indispensable role in such matters:

“[I]t is very difficult to define the indivisible line that divides perfect and partial insanity; but it must rest upon circumstances duly to be weighed and considered both by the judge and jury, lest on the one side there be a kind of inhumanity towards the defects of human nature, or on the other side too great an indulgence given to great crimes ....” 1 Hale, Pleas of the Crown, at 30.

Proceeding from these faulty assumptions, the Court gives two reasons why the death penalty is an excessive punishment for all mentally retarded offenders. First, the “dimin*350ished capacities” of the mentally retarded raise a “serious question” whether their execution contributes to the “social purposes” of the death penalty, viz., retribution and deterrence. Ante, at 318-319. (The Court conveniently ignores a third “social purpose” of the death penalty — “incapacitation of dangerous criminals and the consequent prevention of crimes that they may otherwise commit in the future,” Gregg v. Georgia, 428 U. S. 153, 183, n. 28 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.). But never mind; its discussion of even the other two does not bear analysis.) Retribution is not advanced, the argument goes, because the mentally retarded are no more culpable than the average murderer, whom we have already held lacks sufficient culpability to warrant the death penalty, see Godfrey v. Georgia, 446 U. S. 420, 433 (1980) (plurality opinion). Ante, at 319. Who says so? Is there an established correlation between mental acuity and the ability to conform one’s conduct to the law in such a rudimentary matter as murder? Are the mentally retarded really more disposed (and hence more likely) to commit willfully cruel and serious crime than others? In my experience, the opposite is true: being childlike generally suggests innocence rather than brutality.

Assuming, however, that there is a direct connection between diminished intelligence and the inability to refrain from murder, what scientific analysis can possibly show that a mildly retarded individual who commits an exquisite torture-killing is “no more culpable” than the “average” murderer in a holdup-gone-wrong or a domestic dispute? Or a moderately retarded individual who commits a series of 20 exquisite torture-killings? Surely culpability, and deservedness of the most severe retribution, depends not merely (if at all) upon the mental capacity of the criminal (above the level where he is able to distinguish right from wrong) but also upon the depravity of the crime — which is precisely why this sort of question has traditionally been thought answerable not by a categorical rule of the sort the Court today *351imposes upon all trials, but rather by the sentencer’s weighing of the circumstances (both degree of retardation and depravity of crime) in the particular case. The fact that juries continue to sentence mentally retarded offenders to death for extreme crimes shows that society’s moral outrage sometimes demands execution of retarded offenders. By what principle of law, science, or logic can the Court pronounce that this is wrong? There is none. Once the Court admits (as it does) that mental retardation does not render the offender morally blameless, ante, at 318, there is no basis for saying that the death penalty is never appropriate retribution, no matter how heinous the crime. As long as a mentally retarded offender knows “the difference between right and wrong,” ibid., only the sentencer can assess whether his retardation reduces his culpability enough to exempt him from the death penalty for the particular murder in question.

As for the other social purpose of the death penalty that the Court discusses, deterrence: That is not advanced, the Court tells us, because the mentally retarded are “less likely” than their nonretarded counterparts to “process the information of the possibility of execution as a penalty and ... control their conduct based upon that information.” Ante, at 320. Of course this leads to the same conclusion discussed earlier — that the mentally retarded (because they are less deterred) are more likely to kill — which neither I nor the society at large believes. In any event, even the Court does not say that all mentally retarded individuals cannot “process the information of the possibility of execution as a penalty and . . . control their conduct based upon that information”; it merely asserts that they are “less likely” to be able to do so. But surely the deterrent effect of a penalty is adequately vindicated if it successfully deters many, but not all, of the target class. Virginia’s death penalty, for example, does not fail of its deterrent effect simply because some criminals are unaware that Virginia has the death penalty. In other words, the supposed fact that some *352retarded criminals cannot fully appreciate the death penalty has nothing to do with the deterrence rationale, but is simply an echo of the arguments denying a retribution rationale, discussed and rejected above. I am not sure that a murderer is somehow less blameworthy if (though he knew his act was wrong) he did not fully appreciate that he could die for it; but if so, we should treat a mentally retarded murderer the way we treat an offender who may be “less likely” to respond to the death penalty because he was abused as a child. We do not hold him immune from capital punishment, but require his background to be considered by the sentencer as a mitigating factor. Eddings v. Oklahoma, 455 U. S. 104, 113-117 (1982).

The Court throws one last factor into its grab bag of reasons why execution of the retarded is “excessive” in all cases: Mentally retarded offenders “face a special risk of wrongful execution” because they are less able “to make a persuasive showing of mitigation,” “to give meaningful assistance to their counsel,” and to be effective witnesses. Ante, at 320-321. “Special risk” is pretty flabby language (even flabbier than “less likely”) — and I suppose a similar “special risk” could be said to exist for just plain stupid people, inarticulate people, even ugly people. If this unsupported claim has any substance to it (which I doubt), it might support a due process claim in all criminal prosecutions of the mentally retarded; but it is hard to see how it has anything to do with an Eighth Amendment claim that execution of the mentally retarded is cruel and unusual. We have never before held it to be cruel and unusual punishment to impose a sentence in violation of some other constitutional imperative.

* * *

Today’s opinion adds one more to the long list of substantive and procedural requirements impeding imposition of the death penalty imposed under this Court’s assumed power to invent a death-is-different jurisprudence. None of those *353requirements existed when the Eighth Amendment was adopted, and some of them were not even supported by current moral consensus. They include prohibition of the death penalty for “ordinary” murder, Godfrey, 446 U. S., at 433, for rape of an adult woman, Coker, 433 U. S., at 592, and for felony murder absent a showing that the defendant possessed a sufficiently culpable state of mind, Enmund, 458 U. S., at 801; prohibition of the death penalty for any person under the age of 16 at the time of the crime, Thompson, 487 U. S., at 838 (plurality opinion); prohibition of the death penalty as the mandatory punishment for any crime, Woodson v. North Carolina, 428 U. S. 280, 305 (1976) (plurality opinion), Sumner v. Shuman, 483 U. S. 66, 77-78 (1987); a requirement that the sentencer not be given unguided discretion, Furman v. Georgia, 408 U. S. 238 (1972) (per cu-riam), a requirement that the sentencer be empowered to take into account all mitigating circumstances, Lockett v. Ohio, 438 U. S. 586, 604 (1978) (plurality opinion), Eddings v. Oklahoma, supra, at 110; and a requirement that the accused receive a judicial evaluation of his claim of insanity before the sentence can be executed, Ford, 477 U. S., at 410-411 (plurality opinion). There is something to be said for popular abolition of the death penalty; there is nothing to be said for its incremental abolition by this Court.

This newest invention promises to be more effective than any of the others in turning the process of capital trial into a game. One need only read the definitions of mental retardation adopted by the American Association on Mental Retardation and the American Psychiatric Association (set forth in the Court’s opinion, ante, at 308, n. 3) to realize that the symptoms of this condition can readily be feigned. And whereas the capital defendant who feigns insanity risks commitment to a mental institution until he can be cured (and then tried and executed), Jones v. United States, 463 U. S. 354, 370, and n. 20 (1983), the capital defendant who feigns mental retardation risks nothing at all. The mere pendency *354of the present case has brought us petitions by death row inmates claiming for the first time, after multiple habeas petitions, that they are retarded. See, e. g., Moore v. Texas, 535 U. S. 1044 (2002) (Scalia, J., dissenting from grant of applications for stay of execution).

Perhaps these practical difficulties will not be experienced by the minority of capital-punishment States that have very recently changed mental retardation from a mitigating factor (to be accepted or rejected by the sentencer) to an absolute immunity. Time will tell — and the brief time those States have had the new disposition in place (an average of 6.8 years) is surely not enough. But if the practical difficulties do not appear, and if the other States share the Court’s perceived moral consensus that all mental retardation renders the death penalty inappropriate for all crimes, then that majority will presumably follow suit. But there is no justification for this Court’s pushing them into the experiment— and turning the experiment into a permanent practice — on constitutional pretext. Nothing has changed the accuracy of Matthew Hale’s endorsement of the common law’s traditional method for taking account of guilt-reducing factors, written over three centuries ago:

“[Determination of a person’s incapacity] is a matter of great difficulty, partly from the easiness of counterfeiting this disability ... and partly from the variety of the degrees of this infirmity, whereof some are sufficient, and some are insufficient to excuse persons in capital offenses... .

“Yet the law of England hath afforded the best method of trial, that is possible, of this and all other matters of fact, namely, by a jury of twelve men all concurring in the same judgment, by the testimony of witnesses . . . , and by the inspection and direction of the judge.” 1 Pleas of the Crown, at 32-33.

I respectfully dissent.