with whom Justice O’Connor and Justice Souter join, concurring in part and concurring in the judgment.
I concur in Part IV of the Court’s opinion and in the judgment. I write this separate opinion because my approach to the Eighth Amendment proportionality analysis differs from Justice Scalia’s. Regardless of whether Justice Scalia or Justice White has the best of the historical argument, compare ante, at 966-985, with post, at 1009-1011, and n. 1, stare decisis counsels our adherence to the narrow proportionality principle that has existed in our Eighth Amendment jurisprudence for 80 years. Although our proportionality decisions have not been clear or consistent in all respects, *997they can be reconciled, and they require us to uphold petitioner’s sentence.
I
A
Our decisions recognize that the Cruel and Unusual Punishments Clause encompasses a narrow proportionality principle. We first interpreted the Eighth Amendment to prohibit “‘greatly disproportioned’” sentences in Weems v. United States, 217 U. S. 349, 371 (1910), quoting O’Neil v. Vermont, 144 U. S. 323, 340 (1892) (Field, J., dissenting). Since Weems, we have applied the principle in different Eighth Amendment contexts. Its most extensive application has been in death penalty cases. In Coker v. Georgia, 433 U. S. 584, 592 (1977), we held 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.” We applied like reasoning in Enmund v. Florida, 458 U. S. 782 (1982), to strike down a capital sentence imposed for a felony-murder conviction in which the defendant had not committed the actual murder and lacked intent to kill. Cf. Tison v. Arizona, 481 U. S. 137 (1987).
The Eighth Amendment proportionality principle also applies to noncapital sentences. In Rummel v. Estelle, 445 U. S. 263 (1980), we acknowledged the existence of the proportionality rule for both capital and noncapital cases, id., at 271-274, and n. 11, but we refused to strike down a sentence of life imprisonment, with possibility of parole, for recidivism based on three underlying felonies. In Hutto v. Davis, 454 U. S. 370, 374, and n. 3 (1982), we recognized the possibility of proportionality review but held it inapplicable to a 40-year prison sentence for possession with intent to distribute nine ounces of marijuana. Our most recent decision discussing the subject is Solem v. Helm, 463 U. S. 277 (1983). There we held that a sentence of life imprisonment without possibility of parole violated the Eighth Amendment because it was *998“grossly disproportionate” to the crime of recidivism based on seven underlying nonviolent felonies. The dissent in Solem disagreed with the Court’s application of the proportionality principle but observed that in extreme cases it could apply to invalidate a punishment for a term of years. Id., at 280, n. 3. See also Hutto v. Finney, 437 U. S. 678, 685 (1978) (dicta); Ingraham v. Wright, 430 U. S. 651, 667 (1977) (dicta).
B
Though our decisions recognize a proportionality principle, its precise contours are unclear. This is so in part because we have applied the rule in few cases and even then to sentences of different types. Our most recent pronouncement on the subject in Solem, furthermore, appeared to apply a different analysis than in Rummel and Davis. Solem twice stated, however, that its decision was consistent with Rummel and thus did not overrule it. Solem, supra, at 288, n. 13, 303, n. 32. Despite these tensions, close analysis of our decisions yields some common principles that give content to the uses and limits of proportionality review.
The first of these principles is that the fixing of prison terms for specific crimes involves a substantive penological judgment that, as a general matter, is “properly within the province of legislatures, not courts.” Rummel, supra, at 275-276. Determinations about the nature and purposes of punishment for criminal acts implicate difficult and enduring questions respecting the sanctity of the individual, the nature of law, and the relation between law and the social order. “As a moral or political issue [the punishment of offenders] provokes intemperate emotions, deeply conflicting interests, and intractable disagreements.” D. Garland, Punishment and Modern Society 1 (1990). The efficacy of any sentencing system cannot be assessed absent agreement on the purposes and objectives of the penal system. And the responsibility for making these fundamental choices and implementing them lies with the legislature. See Gore v. United States, *999357 U. S. 386, 393 (1958) (“Whatever views may be entertained regarding severity of punishment, whether one believes in its efficacy or its futility, . . . these are peculiarly questions of legislative policy”). Thus, “[Reviewing courts . . . should grant substantial deference to the broad authority that legislatures necessarily possess in determining the types and limits of punishments for crimes. ” Solem, supra, at 290. See also Rummel, supra, at 274 (acknowledging “reluctance to review legislatively mandated terms of imprisonment”); Weems, supra, at 379 (“The function of the legislature is primary, its exercises fortified by presumptions of right and legality, and is not to be interfered with lightly, nor by any judicial conception of their wisdom or propriety”).
The second principle is that the Eighth Amendment does not mandate adoption of any one penological theory. “The principles which have guided criminal sentencing . . . have varied with the times.” Payne v. Tennessee, ante, at 819. The federal and state criminal systems have accorded different weights at different times to the penological goals of retribution, deterrence, incapacitation, and rehabilitation. Compare Mistretta v. United States, 488 U. S. 361, 363-366 (1989), with Williams v. New York, 337 U. S. 241, 248 (1949). And competing theories of mandatory and discretionary sentencing have been in varying degrees of ascendancy or decline since the beginning of the Republic. See United States v. Grayson, 438 U. S. 41, 45-47 (1978).
Third, marked divergences both in underlying theories of sentencing and in the length of prescribed prison terms are the inevitable, often beneficial, result of the federal structure. See Solem, supra, at 291, n. 17 (“The inherent nature of our federal system” may result in “a wide range of constitutional sentences”). “Our federal system recognizes the independent power of a State to articulate societal norms through criminal law.” McCleskey v. Zant, 499 U. S. 467, 491 (1991). State sentencing schemes may embody different penological assumptions, making interstate comparison of *1000sentences a difficult and imperfect enterprise. See Rummel, 445 U. S., at 281. See also Solem, 463 U. S., at 294-295 (comparison of different terms of years for imprisonment “troubling” but not “unique to this area”). And even assuming identical philosophies, differing attitudes and perceptions of local conditions may yield different, yet rational, conclusions regarding the appropriate length of prison terms for particular crimes. Thus, the circumstance that a State has the most severe punishment for a particular crime does not by itself render the punishment grossly disproportionate. Rummel, 445 U. S., at 281. “[O]ur Constitution ‘is made for people of fundamentally differing views.’ . . . Absent a constitutionally imposed uniformity inimical to traditional notions of federalism, some State will always bear the distinction of treating particular offenders more severely than any other State.” Id., at 282, quoting Lochner v. New York, 198 U. S. 45, 76 (1905) (Holmes, J., dissenting). See also Graham v. West Virginia, 224 U. S. 616 (1912).
The fourth principle at work in our cases is that proportionality review by federal courts should be informed by “‘objective factors to the maximum possible extent.’” Rummel, supra, at 274-275, quoting Coker, 433 U. S., at 592 (plurality opinion). See also Solem, supra, at 290. The most prominent objective factor is the type of punishment imposed. In Weems, “the Court could differentiate in an objective fashion between the highly unusual cadena temporal and more traditional forms of imprisonment imposed under the Anglo-Saxon system.” Rummel, 445 U. S., at 275. In a similar fashion, because “ ‘[t]he penalty of death differs from all other forms of criminal punishment,”’ id., at 272, quoting Furman v. Georgia, 408 U. S. 238, 306 (1972) (opinion of Stewart, J.), the objective line between capital punishment and imprisonment for a term of years finds frequent mention in our Eighth Amendment jurisprudence. See Solem, supra, at 294 (“The easiest comparison [of different sentences] is between capital punishment and noncapital punish*1001ment”). By contrast, our decisions recognize that we lack clear objective standards to distinguish between sentences for different terms of years. Rummel, supra, at 275. See also Solem, 463 U. S., at 294 (“It is clear that a 25-year sentence generally is more severe than a 15-year sentence, but in most cases it would be difficult to decide that the former violates the Eighth Amendment while the latter does not”) (footnote omitted). Although “no penalty is per se constitutional,” id., at 290, the relative lack of objective standards concerning terms of imprisonment has meant that “ ‘[o]utside the context of capital punishment, successful challenges to the proportionality of particular sentences [are] exceedingly rare.’” Id., at 289-290, quoting Rummel, supra, at 272.
All of these principles — the primacy of the legislature, the variety of legitimate penological schemes, the nature of our federal system, and the requirement that proportionality review be guided by objective factors — inform the final one: The Eighth Amendment does not require strict proportionality between crime and sentence. Rather, it forbids only extreme sentences that are “grossly disproportionate” to the crime. Solem, supra, at 288, 303. See also Weems, 217 U. S., at 371 (Eighth Amendment prohibits “greatly dispro-portioned” sentences); Coker, supra, at 592 (Eighth Amendment prohibits “grossly disproportionate” sentences); Rum-mel, supra, at 271 (same).
I — I h — 1
With these considerations stated, it is necessary to examine the challenged aspects of petitioner’s sentence: its severe length and its mandatory operation.
A
Petitioner’s life sentence without parole is the second most severe penalty permitted by law. It is the same sentence received by the petitioner in Solem. Petitioner’s crime, however, was far more grave than the crime at issue in Solem.
*1002The crime of uttering a no account check at issue in Solem was “‘one of the most passive felonies a person could commit.”’ Solem, 463 U. S., at 296 (citation omitted). It “involved neither violence nor threat of violence to any person,” and was “viewed by society as among the less serious offenses.” Ibid. The felonies underlying the defendant’s recidivism conviction, moreover, were “all relatively minor.” Id., at 296-297. The Solem Court contrasted these “minor” offenses with “very serious offenses” such as “a third offense of heroin dealing,” and stated that “[n]o one suggests that [a statute providing for life imprisonment without parole] may not be applied constitutionally to fourth-time heroin dealers or other violent criminals.” Id., at 299, and n. 26.
Petitioner was convicted of possession of more than 650 grams (over 1.5 pounds) of cocaine. This amount of pure cocaine has a potential yield of between 32,500 and 65,000 doses. A. Washton, Cocaine Addiction: Treatment, Recovery, and Relapse Prevention 18 (1989). From any standpoint, this crime falls in a different category from the relatively minor, nonviolent crime at issue in Solem. Possession, use, and distribution of illegal drugs represent “one of the greatest problems affecting the health and welfare of our population.” Treasury Employees v. Von Raab, 489 U. S. 656, 668 (1989). Petitioner’s suggestion that his crime was nonviolent and victimless, echoed by the dissent, see post, at 1022-1023, is false to the point of absurdity. To the contrary, petitioner’s crime threatened to cause grave harm to society.
Quite apart from the pernicious effects on the individual who consumes illegal drugs, such drugs relate to crime in at least three ways: (1) A drug user may commit crime because of drug-induced changes in physiological functions, cognitive ability, and mood; (2) A drug user may commit crime in order to obtain money to buy drugs; and (3) A violent crime may occur as part of the drug business or culture. See Goldstein, Drugs and Violent Crime, in Pathways to Criminal Violence *100316, 24-36 (N. Weiner & M. Wolfgang eds. 1989). Studies bear out these possibilities and demonstrate a direct nexus between illegal drugs and crimes of violence. See generally id., at 16-48. To mention but a few examples, 57 percent of a national sample of males arrested in 1989 for homicide tested positive for illegal drugs. National Institute of Justice, 1989 Drug Use Forecasting Annual Report 9 (June 1990). The comparable statistics for assault, robbery, and weapons arrests were 55, 73, and 63 percent, respectively. Ibid. In Detroit, Michigan, in 1988, 68 percent of a sample of male arrestees and 81 percent of a sample of female arrest-ees tested positive for illegal drugs. National Institute of Justice, 1988 Drug Use Forecasting Annual Report 4 (Mar. 1990). Fifty-one percent of males and seventy-one percent of females tested positive for cocaine. Id., at 7. And last year an estimated 60 percent of the homicides in Detroit were drug related, primarily cocaine related. U. S. Department of Health and Human Services, Epidemiologic Trends in Drug Abuse 107 (Dec. 1990).
These and other facts and reports detailing the pernicious effects of the drug epidemic in this country do not establish that Michigan’s penalty scheme is correct or the most just in any abstract sense. But they do demonstrate that the Michigan Legislature could with reason conclude that the threat posed to the individual and society by possession of this large an amount of cocaine — in terms of violence, crime, and social displacement — is momentous enough to warrant the deterrence and retribution of a life sentence without parole. See United States v. Mendenhall, 446 U. S. 544, 561 (1980) (Powell, J., concurring in part and concurring in judgment) (“Few problems affecting the health and welfare of our population, particularly our young, cause greater concern than the escalating use of controlled substances”); Florida v. Royer, 460 U. S. 491, 513 (1983) (Blackmun, J., dissenting) (same). See also Terrebonne v. Butler, 848 F. 2d 500, 504 (CA5 1988) (en banc).
*1004The severity of petitioner’s crime brings his sentence within the constitutional boundaries established by our prior decisions. In Hutto v. Davis, 454 U. S. 370 (1982), we upheld against proportionality attack a sentence of 40 years’ imprisonment for possession with intent to distribute nine ounces of marijuana. Here, Michigan could with good reason conclude that petitioner’s crime is more serious than the crime in Davis. Similarly, a rational basis exists for Michigan to conclude that petitioner’s crime is as serious and violent as the crime of felony murder without specific intent to kill, a crime for which “no sentence of imprisonment would be disproportionate,” Solem, 463 U. S., at 290, n. 15. Cf. Rummel, 445 U. S., at 296, n. 12 (Powell, J., dissenting) (“A professional seller of addictive drugs may inflict greater bodily harm upon members of society than the person who commits a single assault”).
Petitioner and amici contend that our proportionality decisions require a comparative analysis between petitioner’s sentence and sentences imposed for other crimes in Michigan and sentences imposed for the same crime in other jurisdictions. Given the serious nature of petitioner’s crime, no such comparative analysis is necessary. Although Solem considered these comparative factors after analyzing “the gravity of the offense and the harshness of the penalty,” 463 U. S., at 290-291, it did not announce a rigid three-part test. In fact, Solem stated that in determining unconstitutional disproportionality, “no one factor will be dispositive in a given case.” Id., at 291, n. 17. See also ibid. (“[N]o single criterion can identify when a sentence is so grossly disproportionate that it violates the Eighth Amendment”).
On the other hand, one factor may be sufficient to determine the constitutionality of a particular sentence. Consistent with its admonition that “a reviewing court rarely will be required to engage in extended analysis to determine that a sentence is not constitutionally disproportionate,” id., at 290, n. 16, Solem is best understood as holding that comparative *1005analysis within and between jurisdictions is not always relevant to proportionality review. The Court stated that “it may be helpful to compare sentences imposed on other criminals in the same jurisdiction,” and that “courts may find it useful to compare the sentences imposed for commission of the same crime in other jurisdictions.” Id., at 291-292 (emphasis added). It did not mandate such inquiries.
A better reading of our cases leads to the conclusion that intrajurisdictional and interjurisdictional analyses are appropriate only in the rare case in which a threshold comparison of the crime committed and the sentence imposed leads to an inference of gross disproportionality. In Solem and Weems, decisions in which the Court invalidated sentences as disproportionate, we performed a comparative analysis of sentences after determining that the sentence imposed was grossly excessive punishment for the crime committed. Solem, supra, at 298-300; Weems, 217 U. S., at 377-381. By contrast, Rummel and Davis, decisions in which the Court upheld sentences against proportionality attacks, did not credit such comparative analyses. In rejecting this form of argument, Rummel noted that “[e]ven were we to assume that the statute employed against Rummel was the most stringent found in the 50 States, that severity hardly would render Rummers punishment ‘grossly disproportionate’ to his offenses.” Rummel, supra, at 281.
The proper role for comparative analysis of sentences, then, is to validate an initial judgment that a sentence is grossly disproportionate to a crime. This conclusion neither “eviscerate[s]” Solem, nor “abandon[s]” its second and third factors, as the dissent charges, post, at 1018, 1020, and it takes full account of Rummel and Davis, cases ignored by the dissent. In light of the gravity of petitioner’s offense, a comparison of his crime with his sentence does not give rise to an inference of gross disproportionality, and comparative analysis of his sentence with others in Michigan and across the Nation need not be performed.
*1006B
Petitioner also attacks his sentence because of its mandatory nature. Petitioner would have us hold that any severe penalty scheme requires individualized sentencing so that a judicial official may consider mitigating circumstances. Our precedents do not support this proposition, and petitioner presents no convincing reason to fashion an exception or adopt a new rule in the case before us. The Court demonstrates that our Eighth Amendment capital decisions reject any requirement of individualized sentencing in noncapital cases. Ante, at 994 — 996.
The mandatory nature of this sentence comports with our noncapital proportionality decisions as well. The statute at issue in Solem made the offender liable to a maximum, not a mandatory, sentence of life imprisonment without parole. Solem, 463 U. S., at 281-282, n. 6. Because a “lesser sentence . . . could have been entirely consistent with both the statute and the Eighth Amendment,” the Court’s decision “d[id] not question the legislature’s judgment,” but rather challenged the sentencing court’s selection of a penalty at the top of the authorized sentencing range. Id., at 299, n. 26. Here, by contrast, the Michigan Legislature has mandated the penalty and has given the state judge no discretion in implementing it. It is beyond question that the legislature “has the power to define criminal punishments without giving the courts any sentencing discretion,” Chapman v. United States, 500 U. S. 453, 467 (1991). Since the beginning of the Republic, Congress and the States have enacted mandatory sentencing schemes. See Mistretta v. United States, 488 U. S., at 363; United States v. Grayson, 438 U. S., at 45-46; Ex parte United States, 242 U. S. 27 (1916). To set aside petitioner’s mandatory sentence would require rejection not of the judgment of a single jurist, as in Solem, but rather the collective wisdom of the Michigan Legislature and, as a consequence, the Michigan citizenry. We have never invalidated a penalty mandated by a legislature based only on the *1007length of sentence, and, especially with a crime as severe as this one, we should do so only in the most extreme circumstance. Cf. Rummel, 445 U. S., at 274.
In asserting the constitutionality of this mandatory sentence, I offer no judgment on its wisdom. Mandatory sentencing schemes can be criticized for depriving judges of the power to exercise individual discretion when remorse and acknowledgment of guilt, or other extenuating facts, present what might seem a compelling case for departure from the maximum. On the other hand, broad and unreviewed discretion exercised by sentencing judges leads to the perception that no clear standards are being applied, and that the rule of law is imperiled by sentences imposed for no discernible reason other than the subjective reactions of the sentencing judge. The debate illustrates that, as noted at the outset, arguments for and against particular sentencing schemes are for legislatures to resolve.
Michigan’s sentencing scheme establishes graduated punishment for offenses involving varying amounts of mixtures containing controlled substances. Possession of controlled substances in schedule 1 or 2 in an amount less than 50 grams results in a sentence of up to 20 years’ imprisonment; possession of more than 50 but less than 225 grams results in a mandatory minimum prison sentence of 10 years with a maximum sentence of 20 years; possession of more than 225 but less than 650 grams results in a mandatory minimum prison sentence of 20 years with a maximum sentence of 30 years; and possession of 650 grams or more results in a mandatory life sentence. Mich. Comp. Laws Ann. § 333.7401 (West Supp. 1990-1991). Sentencing courts may depart from the minimum terms specified for all amounts, except those exceeding 650 grams, “if the court finds on the record that there are substantial and compelling reasons to do so.” §§ 333.7401(4), 333.7403(3). This system is not an ancient one revived in a sudden or surprising way; it is, rather, a recent enactment calibrated with care, clarity, and much deliberation to ad*1008dress a most serious contemporary social problem. The scheme provides clear notice of the severe consequences that attach to possession of drugs in wholesale amounts, thereby giving force to one of the first purposes of criminal law — deterrence. In this sense, the Michigan scheme may be as fair, if not more so, than other sentencing systems in which the sentencer’s discretion or the complexity of the scheme obscures the possible sanction for a crime, resulting in a shock to the offender who learns the severity of his sentence only after he commits the crime.
The Michigan scheme does possess mechanisms for consideration of individual circumstances. Prosecutorial discretion before sentence and executive or legislative clemency after-wards provide means for the State to avert or correct unjust sentences. Here the prosecutor may have chosen to seek the maximum penalty because petitioner possessed 672.5 grams of undiluted cocaine and several other trappings of a drug trafficker, including marijuana cigarettes, four brass cocaine straws, a cocaine spoon, 12 Percodan tablets, 25 tablets of Phendimetrazine Tartrate, a Motorola beeper, plastic bags containing cocaine, a coded address book, and $3,500 in cash.
HH 1 — I
A penalty as severe and unforgiving as the one imposed here would make this a most difficult and troubling case for any judicial officer. Reasonable minds may differ about the efficacy of Michigan’s sentencing scheme, and it is far from certain that Michigan’s bold experiment will succeed. The accounts of pickpockets at Tyburn hangings are a reminder of the limits of the law’s deterrent force, but we cannot say the law before us has no chance of success and is on that account so disproportionate as to be cruel and unusual punishment. The dangers flowing from drug offenses and the circumstances of the crime committed here demonstrate that the Michigan penalty scheme does not surpass constitutional *1009bounds. Michigan may use its criminal law to address the issue of drug possession in wholesale amounts in the manner that it has in this sentencing scheme. See New State Ice Co. v. Liebmann, 285 U. S. 262, 311 (1932) (Brandeis, J., dissenting). For the foregoing reasons, I conclude that petitioner’s sentence of life imprisonment without parole for his crime of possession of more than 650 grams of cocaine does not violate the Eighth Amendment.