Miller v. Alabama/Dissent Roberts

Miller v. Alabama, 567 U.S. 460 (2012)
Dissenting opinion by John Roberts
Court Documents
Case Syllabus
Opinion of the Court
Concurring Opinion
Dissenting Opinions


Determining the appropriate sentence for a teenager convicted of murder presents grave and challenging questions of morality and social policy. Our role, however, is to apply the law, not to answer such questions. The pertinent law here is the Eighth Amendment to the Constitution, which prohibits "cruel and unusual punishments. " Today, the Court invokes that Amendment to ban a punishment that the Court does not itself characterize as unusual, and that could not plausibly be described as such. I therefore dissent.

The parties agree that nearly 2,500 prisoners are presently serving life sentences without the possibility of parole for [p494] murders they committed before the age of 18. Brief for Petitioner in No. 10–9647, p. 62, n. 80 (Jackson Brief); Brief for Respondent in No. 10–9646, p. 30 (Alabama Brief). The Court accepts that over 2,000 of those prisoners received that sentence because it was mandated by a legislature. Ante, at 483, n. 10. And it recognizes that the Federal Government and most States impose such mandatory sentences. Ante, at 482. Put simply, if a 17-year-old is convicted of deliberately murdering an innocent victim, it is not "unusual" for the murderer to receive a mandatory sentence of life without parole. That reality should preclude finding that mandatory life imprisonment for juvenile killers violates the Eighth Amendment.

Our precedent supports this conclusion. When determining whether a punishment is cruel and unusual, this Court typically begins with "'objective indicia of society's standards, as expressed in legislative enactments and state practice.'" Graham v. Florida, 560 U.S. 48, 61 (2010); see also, e.g., Kennedy v. Louisiana, 554 U.S. 407, 422 (2008); Roper v. Simmons, 543 U.S. 551, 564 (2005). We look to these "objective indicia" to ensure that we are not simply following our own subjective values or beliefs. Gregg v. Georgia, 428 U.S. 153, 173 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.). Such tangible evidence of societal standards enables us to determine whether there is a "consensus against" a given sentencing practice. Graham, supra, at 61. If there is, the punishment may be regarded as "unusual." But when, as here, most States formally require and frequently impose the punishment in question, there is no objective basis for that conclusion.

Our Eighth Amendment cases have also said that we should take guidance from "evolving standards of decency that mark the progress of a maturing society." Ante, at 469 (quoting Estelle v. Gamble, 429 U.S. 97, 102 (1976); internal quotation marks omitted). Mercy toward the guilty can be a form of decency, and a maturing society may abandon harsh [p495] punishments that it comes to view as unnecessary or unjust. But decency is not the same as leniency. A decent society protects the innocent from violence. A mature society may determine that this requires removing those guilty of the most heinous murders from its midst, both as protection for its other members and as a concrete expression of its standards of decency. As judges we have no basis for deciding that progress toward greater decency can move only in the direction of easing sanctions on the guilty.

In this case, there is little doubt about the direction of society's evolution: For most of the 20th century, American sentencing practices emphasized rehabilitation of the offender and the availability of parole. But by the 1980's, outcry against repeat offenders, broad disaffection with the rehabilitative model, and other factors led many legislatures to reduce or eliminate the possibility of parole, imposing longer sentences in order to punish criminals and prevent them from committing more crimes. See, e.g., Alschuler, The Changing Purposes of Criminal Punishment, 70 U. Chi. L. Rev. 1, 1–13 (2003); see generally Crime and Public Policy (J. Wilson & J. Petersilia eds. 2011). Statutes establishing life without parole sentences in particular became more common in the past quarter century. See Baze v. Rees, 553 U.S. 35, 78, and n. 10 (2008) (Stevens, J., concurring in judgment). And the parties agree that most States have changed their laws relatively recently to expose teenage murderers to mandatory life without parole. Jackson Brief 54–55; Alabama Brief 4–5.

The Court attempts to avoid the import of the fact that so many jurisdictions have embraced the sentencing practice at issue by comparing these cases to the Court's prior Eighth Amendment cases. The Court notes that Graham found a punishment authorized in 39 jurisdictions unconstitutional, whereas the punishment it bans today is mandated in 10 fewer. Ante, at 483. But Graham went to considerable lengths to show that although theoretically allowed in many [p496] States, the sentence at issue in that case was "exceedingly rare" in practice. 560 U.S., at 67. The Court explained that only 123 prisoners in the entire Nation were serving life without parole for nonhomicide crimes committed as juveniles, with more than half in a single State. It contrasted that with statistics showing nearly 400,000 juveniles were arrested for serious nonhomicide offenses in a single year. Based on the sentence's rarity despite the many opportunities to impose it, Graham concluded that there was a national consensus against life without parole for juvenile nonhomicide crimes. Id., at 64–67.

Here the number of mandatory life without parole sentences for juvenile murderers, relative to the number of juveniles arrested for murder, is over 5,000 times higher than the corresponding number in Graham. There is thus nothing in these cases like the evidence of national consensus in Graham.[1]

The Court disregards these numbers, claiming that the prevalence of the sentence in question results from the number of statutes requiring its imposition. Ante, at 484, n. 10. True enough. The sentence at issue is statutorily mandated life without parole. Such a sentence can only result from statutes requiring its imposition. In Graham the Court relied on the low number of actual sentences to explain why the high number of statutes allowing such sentences was not dispositive. Here, the Court excuses the high number of actual sentences by citing the high number of statutes impos- [p497] ing it. To say that a sentence may be considered unusual because so many legislatures approve it stands precedent on its head.[2]

The Court also advances another reason for discounting the laws enacted by Congress and most state legislatures. Some of the jurisdictions that impose mandatory life without parole on juvenile murderers do so as a result of two statutes: one providing that juveniles charged with serious crimes may be tried as adults, and another generally mandating that those convicted of murder be imprisoned for life. According to the Court, our cases suggest that where the sentence results from the interaction of two such statutes, the legislature can be considered to have imposed the resulting sentences "inadvertent[ly]." Ante, at 485–487. The Court relies on Graham and Thompson v. Oklahoma, 487 U.S. 815, 826, n. 24 (1988) (plurality opinion), for the proposition that these laws are therefore not valid evidence of society's views on the punishment at issue.

It is a fair question whether this Court should ever assume a legislature is so ignorant of its own laws that it does not understand that two of them interact with each other, especially on an issue of such importance as the one before us. But in Graham and Thompson it was at least plausible as a practical matter. In Graham, the extreme rarity with [p498] which the sentence in question was imposed could suggest that legislatures did not really intend the inevitable result of the laws they passed. See 560 U.S., at 66–67. In Thompson, the sentencing practice was even rarer—only 20 defendants had received it in the last century. 487 U.S., at 832 (plurality opinion). Perhaps under those facts it could be argued that the legislature was not fully aware that a teenager could receive the particular sentence in question. But here the widespread and recent imposition of the sentence makes it implausible to characterize this sentencing practice as a collateral consequence of legislative ignorance.[3]

Nor do we display our usual respect for elected officials by asserting that legislators have accidentally required 2,000 teenagers to spend the rest of their lives in jail. This is particularly true given that our well-publicized decision in Graham alerted legislatures to the possibility that teenagers were subject to life with parole only because of legislative inadvertence. I am aware of no effort in the wake of Graham to correct any supposed legislative oversight. Indeed, in amending its laws in response to Graham one legislature made especially clear that it does intend juveniles who commit first-degree murder to receive mandatory life without parole. See Iowa Code Ann. § 902.1 (West Cum. Supp. 2012).

In the end, the Court does not actually conclude that mandatory life sentences for juvenile murderers are unusual. It instead claims that precedent "leads to" today's decision, primarily relying on Graham and Roper. Ante, at 470. Petitioners argue that the reasoning of those cases "compels" finding in their favor. Jackson Brief 34. The Court is apparently unwilling to go so far, asserting only that precedent points in that direction. But today's decision invalidates the laws of dozens of legislatures and Congress. This Court is [p499] not easily led to such a result. See, e.g., United States v. Harris, 106 U.S. 629, 635 (1883) (courts must presume an Act of Congress is constitutional "unless the lack of constitutional authority . . . is clearly demonstrated"). Because the Court does not rely on the Eighth Amendment's text or objective evidence of society's standards, its analysis of precedent alone must bear the "heavy burden [that] rests on those who would attack the judgment of the representatives of the people." Gregg, 428 U.S., at 175. If the Court is unwilling to say that precedent compels today's decision, perhaps it should reconsider that decision.

In any event, the Court's holding does not follow from Roper and Graham. Those cases undoubtedly stand for the proposition that teenagers are less mature, less responsible, and less fixed in their ways than adults—not that a Supreme Court case was needed to establish that. What they do not stand for, and do not even suggest, is that legislators—who also know that teenagers are different from adults—may not require life without parole for juveniles who commit the worst types of murder.

That Graham does not imply today's result could not be clearer. In barring life without parole for juvenile nonhomicide offenders, Graham stated that "[t]here is a line 'between homicide and other serious violent offenses against the individual.'" 560 U.S., at 69 (quoting Kennedy, 554 U.S., at 438). The whole point of drawing a line between one issue and another is to say that they are different and should be treated differently. In other words, the two are in different categories. Which Graham also said: "defendants who do not kill, intend to kill, or foresee that life will be taken are categorically less deserving of the most serious forms of punishment than are murderers." 560 U.S., at 69 (emphasis added). Of course, to be especially clear that what is said about one issue does not apply to another, one could say that the two issues cannot be compared. Graham said that too: "Serious nonhomicide crimes . . . cannot be compared to mur- [p500] der. " Ibid. (internal quotation marks omitted). A case that expressly puts an issue in a different category from its own subject, draws a line between the two, and states that the two should not be compared, cannot fairly be said to control that issue.

Roper provides even less support for the Court's holding. In that case, the Court held that the death penalty could not be imposed for offenses committed by juveniles, no matter how serious their crimes. In doing so, Roper also set itself in a different category than these cases, by expressly invoking "special" Eighth Amendment analysis for death penalty cases. 543 U.S., at 568–569. But more importantly, Roper reasoned that the death penalty was not needed to deter juvenile murderers in part because "life imprisonment without the possibility of parole" was available. Id., at 572. In a classic bait and switch, the Court now tells state legislatures that—Roper's promise notwithstanding—they do not have power to guarantee that once someone commits a heinous murder, he will never do so again. It would be enough if today's decision proved JUSTICE SCALIA's prescience in writing that Roper's "reassurance . . . gives little comfort." Id., at 623 (dissenting opinion). To claim that Roper actually "leads to" revoking its own reassurance surely goes too far.

Today's decision does not offer Roper and Graham's false promises of restraint. Indeed, the Court's opinion suggests that it is merely a way station on the path to further judicial displacement of the legislative role in prescribing appropriate punishment for crime. The Court's analysis focuses on the mandatory nature of the sentences in these cases. See ante, at 474–480. But then—although doing so is entirely unnecessary to the rule it announces—the Court states that even when a life without parole sentence is not mandatory, "we think appropriate occasions for sentencing juveniles to this harshest possible penalty will be uncommon." Ante, at 479. Today's holding may be limited to mandatory sentences, but the Court has already announced that discre- [p501] tionary life without parole for juveniles should be "uncommon"—or, to use a common synonym, "unusual."

Indeed, the Court's gratuitous prediction appears to be nothing other than an invitation to overturn life without parole sentences imposed by juries and trial judges. If that invitation is widely accepted and such sentences for juvenile offenders do in fact become "uncommon," the Court will have bootstrapped its way to declaring that the Eighth Amendment absolutely prohibits them.

This process has no discernible end point—or at least none consistent with our Nation's legal traditions. Roper and Graham attempted to limit their reasoning to the circumstances they addressed—Roper to the death penalty, and Graham to nonhomicide crimes. Having cast aside those limits, the Court cannot now offer a credible substitute, and does not even try. After all, the Court tells us, "none of what [Graham] said about children . . . is crime-specific." Ante, at 473. The principle behind today's decision seems to be only that because juveniles are different from adults, they must be sentenced differently. See ante, at 476–480. There is no clear reason that principle would not bar all mandatory sentences for juveniles, or any juvenile sentence as harsh as what a similarly situated adult would receive. Unless confined, the only stopping point for the Court's analysis would be never permitting juvenile offenders to be tried as adults. Learning that an Amendment that bars only "unusual" punishments requires the abolition of this uniformly established practice would be startling indeed.


It is a great tragedy when a juvenile commits murder—most of all for the innocent victims. But also for the murderer, whose life has gone so wrong so early. And for society as well, which has lost one or more of its members to deliberate violence, and must harshly punish another. In recent years, our society has moved toward requiring that the [p502] murderer, his age notwithstanding, be imprisoned for the remainder of his life. Members of this Court may disagree with that choice. Perhaps science and policy suggest society should show greater mercy to young killers, giving them a greater chance to reform themselves at the risk that they will kill again. See ante, at 471–474. But that is not our decision to make. Neither the text of the Constitution nor our precedent prohibits legislatures from requiring that juvenile murderers be sentenced to life without parole. I respectfully dissent.

  1. Graham stated that 123 prisoners were serving life without parole for nonhomicide offenses committed as juveniles, while in 2007 alone 380,480 juveniles were arrested for serious nonhomicide crimes. 560 U.S., at 64–65. I use 2,000 as the number of prisoners serving mandatory life without parole sentences for murders committed as juveniles, because all seem to accept that the number is at least that high. And the same source Graham used reports that 1,170 juveniles were arrested for murder and non-negligent homicide in 2009. Dept. of Justice, Office of Juvenile Justice and Delinquency Prevention, C. Puzzanchera & B. Adams, Juvenile Arrests 2009, p. 465 (Dec. 2011).
  2. The Court's reference to discretionary sentencing practices is a distraction. See ante, at 483–484, n. 10. The premise of the Court's decision is that mandatory sentences are categorically different from discretionary ones. So under the Court's own logic, whether discretionary sentences are common or uncommon has nothing to do with whether mandatory sentences are unusual. In any event, if analysis of discretionary sentences were relevant, it would not provide objective support for today's decision. The Court states that "about 15% of all juvenile life-without-parole sentences"—meaning nearly 400 sentences—were imposed at the discretion of a judge or jury. Ante, at 484, n. 10. Thus the number of discretionary life without parole sentences for juvenile murderers, relative to the number of juveniles arrested for murder, is about 1,000 times higher than the corresponding number in Graham.
  3. The Court claims that I "take issue with some or all of these precedents" and "seek to relitigate" them. Ante, at 470–471, n. 4. Not so: Applying this Court's cases exactly as they stand, I do not believe they support the Court's decision in these cases.