UNITED STATES, Appellee, v. WENDELL RIVERA-RUPERTO, a/k/a Arsenio Rivera, Defendant, Appellant.
ORDER OF COURT
Pending before the court is a petition for rehearing or rehearing en banc in United States v. Rivera-Ruperto, No. 12-2364, 12-2367 and a petition for rehearing or rehearing en banc in United States v. Rivera-Ruperto, No. 13-2017. The petitions for rehearing having been denied by the panel of judges who decided the cases, and the petitions for rehearing en banc having been submitted to the active judges of this court and a majority of the judges not having voted that either case be heard en banc, it is ordered that the petitions for rehearing and the petitions for rehearing en banc be denied.
BARRON, Circuit Judge, concurring in the denial of rehearing en banc, joined by HOWARD, Chief Judge, and TORRUELLA, LYNCH, THOMPSON, and KAYATTA, Circuit Judges. The bulk of the 161-year and ten-month prison sentence that Wendell Rivera-Ruperto challenges -- 130 years of it to be exact -- was imposed for his six convictions under 18 U.S.C. § 924(c). United States v. Rivera-Ruperto, 852 F.3d 1, 17 (1st Cir. 2017) (Rivera-Ruperto II). Those convictions stem from a federal sting operation that targeted Puerto Rican police officers. Id. at 4. As part of that sting, Rivera participated, while armed, in a number of supposed “deals” involving large amounts of fake cocaine in which agents of the Federal Bureau of Investigation (FBI) posed as both buyers and sellers. Id. at 4-5.
But, § 924(c) did not merely permit this greater-than-life-without-parole sentence. It mandated it. It did so by requiring a minimum prison sentence of five years for the first of Rivera's § 924(c) convictions and consecutive twenty-five year prison sentences thereafter for each of his “second or subsequent” § 924(c) convictions. 18 U.S.C. § 924(c). And it did so even though all but one of those additional convictions were handed down at the same trial as the initial § 924(c) conviction that Rivera, who had no prior criminal history, received. Id. at 5.1
Thus, in consequence of Rivera's multiple convictions for his involvement in this one sting operation, Rivera was required to receive a punishment that seemingly could have been more severe only if it had required his death. And that is so even though this case is replete with factors that -- under a discretionary sentencing regime -- would surely have been relevant to a judge's individualized rather than arithmetical assessment of whether what Rivera did should not only be punished severely but also deprive him (absent a pardon or commutation) of any hope of ever enjoying freedom again.2
Despite the force of Rivera's argument that this mandatory sentence is so grossly disproportionate as to be unconstitutional under the Eighth Amendment, I am not permitted to conclude that it is. Other federal judges have expressed their dismay that our legal system could countenance extreme mandatory sentences under § 924(c) that are even shorter than this one.3 And yet, just as those judges concluded that they were required by precedent to uphold the sentences in their cases, I conclude, like the panel, Rivera-Ruperto II, 852 F.3d at 18, that I am compelled by precedent -- and, in particular, by the nearly three-decades old, three-Justice concurrence in Harmelin v. Michigan, 501 U.S. 957, 1006 (1991) (opinion of Kennedy, J.) -- to uphold Rivera's greater-than-life sentence.4
I do think it is important to say something, however, about that precedent and why I believe the Supreme Court should revisit it. And so, in what follows, I explain my reasoning.
The body of precedent that controls here concerns the meaning of the Eighth Amendment, which provides that “[e]xcessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.” U.S. Const. amend. VIII. The Amendment's text does not expressly state that prison sentences may be unconstitutional solely in consequence of their length. The Supreme Court, however, has long indicated that a sentence may, in rare cases, be so disproportionate to the seriousness of the underlying offense that it violates the Eighth Amendment. See Weems v. United States, 217 U.S. 349, 368 (1910).
In Rummel v. Estelle, 445 U.S. 263 (1980), for example, the Supreme Court, in the course of rejecting an Eighth Amendment challenge to a mandatory life sentence with the possibility of parole, explained that “the Eighth Amendment prohibits imposition of a sentence that is grossly disproportionate to the severity of the crime.”5 Id. at 271. The Court then applied this principle to invalidate a prison sentence solely in consequence of its disproportionate length in Solem v. Helm, 463 U.S. 277 (1983).
Solem specified the criteria that bear on whether the length of a prison term is impermissibly out of proportion to the seriousness of the offense (or offenses) of conviction. Solem emphasized that “no single criterion can identify when a sentence is so grossly disproportionate that it violates the Eighth Amendment,” Solem, 463 U.S. at 290 n.17, but that “a combination of objective factors can make such analysis possible.” Id. Specifically, Solem held that:
[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 imposed for commission of the same crime in other jurisdictions.
Id. at 292.
Solem appeared to contemplate a holistic analysis, in which the assessment of each of these three criteria would inform the assessment of the others. That approach, notwithstanding its inherently (and appropriately) deferential nature, had teeth. In fact, in Solem, the Court concluded on the basis of this holistic assessment that “the Eighth Amendment proscribes a life sentence without the possibility of parole for a seventh nonviolent felony,” id. at 279, in a case in which that discretionary sentence was triggered by a recidivist defendant's conviction -- after he had been punished for his prior felony convictions -- for uttering a “no account” check for $100. Id. at 303.
Thus, if Solem were the last word, I would have to assess in the following way whether Rivera's mandatory life-without-parole sentence for multiple felonies -- each of which is seemingly nonviolent, though hardly minor in nature -- comports with the Eighth Amendment.6 I would have to consider, holistically, the three criteria that Solem identifies as relevant to the proportionality determination. And, based on a consideration of those criteria, as I will next explain, I would find that Rivera's mandatory, more-than-century-long sentence was grossly disproportionate and thus in violation of the Eighth Amendment.7
The first Solem criterion requires a relatively abstract inquiry. In performing it, a reviewing court must consider the gravity of the offense “in light of the harm caused or threatened to the victim or society[ ] and the culpability of the offender.” Id. at 292. A reviewing court must then consider the harshness of the sentence in light of the gravity of the offense. Id.
Solem details how a court should go about the task of assessing a crime's severity for purposes of applying this first criterion. Of direct relevance here, Solem makes clear that drug crimes are serious, even though they do not inherently require proof of any harm having been done to any identifiable victim.
That guidance from Solem matters in this case. Section 924(c) sanctions anyone who “uses or carries, or who, in furtherance of [a predicate] crime, possesses a firearm,” 18 U.S.C. § 924(c)(1)(A), and then defines those predicate crimes to include a wide variety of federal drug offenses, id. at § 924(c)(2). The predicate drug offenses that underlie each of Rivera's § 924(c) convictions are attempting to possess with intent to distribute, and conspiring to possess with intent to distribute, at least five kilograms of a substance that contained cocaine (though the drug itself need only have been present in a “detectable” amount). 21 U.S.C. §§ 84l(a)(l), (b)(l)(A)(ii)(II); see Rivera-Ruperto II, 852 F.3d at 10.
Thus, we are undoubtedly dealing with the repeated commission of a serious crime under Solem's reasoning. We are also dealing with a type of crime that is certainly more serious than the crime of uttering a “no account” check that triggered the sentence that Solem struck down. 463 U.S. at 281.
Still, Solem did not describe the repeated commission of the crime of drug dealing (let alone inchoate versions of that crime) as, in and of itself, violent conduct, even if the drug involved were heroin.8 Nor did Solem describe drug dealing as a crime that was just as serious as many violent offenses undoubtedly are, at least for purposes of making a threshold assessment of whether a sentence's length is so grossly disproportionate to the underlying offense as to violate the Eighth Amendment. Nor, finally, does Solem suggest that possession of a firearm -- even in furtherance of a drug crime -- is itself a crime of violence.
Indeed, Solem emphasized that the fact that an offense does not actually require proof that the defendant inflicted any bodily harm against any identifiable victim generally makes that offense less serious than an offense that does. 463 U.S. at 292-93. Thus, while Solem does identify felony murder with no intent to kill as an example of the type of grave offense for which a life-without-parole sentence would be constitutional, id. at 291-92 & n.15 (citing Enmund v. Florida, 458 U.S. 782, 795-96 (1982)), it is of some significance under Solem that Rivera's crimes did not require the government to prove that he engaged in conduct that foreseeably resulted in the death of, or bodily injury to, any particular victim.
In offering guidance to judges about how they should evaluate an offense's seriousness under the first criterion, Solem also explicitly distinguished completed crimes from inchoate ones. 463 U.S. at 293. Solem did so on the ground that the latter type of offenses do not require proof that any actual harm resulted. Id.
Rivera was convicted of completed crimes in one sense, given that § 924(c) requires proof of firearm possession in furtherance of the predicate crimes. 18 U.S.C. § 924(c). But, given how § 924(c) works, Rivera's unforgiving life sentence results only from the fact that his firearm possession convictions were connected with drug offenses that were themselves inchoate: attempting to possess with intent to distribute, and conspiring to possess with intent to distribute, at least five kilograms of a substance that contained a detectable amount of cocaine. Rivera-Ruperto I, 846 F.3d at 420. Indeed, Rivera's § 924(c) convictions stem from his involvement with transactions concerning fake rather than real drugs. Rivera-Ruperto II, 852 F.3d at 14. That fact explains why, in addition to the predicate conspiracy convictions, he was charged for and convicted of (as predicate offenses) only attempted rather than actual possession with intent to distribute drugs. And, for that reason, Rivera's conduct, in its nature, could not have actually caused harm to any identifiable person. Thus, this fact, too, suggests that Rivera's § 924(c) offenses, serious though they are, are not, under Solem, of the most serious kind.
Solem did recognize that the fact that an offender is a recidivist is also potentially relevant to the analysis of how serious the conduct being punished is for Eighth Amendment purposes. Solem, 463 U.S. at 296. But Solem did not equate recidivism with the mere commission of multiple offenses that then result in multiple convictions. Solem instead equated recidivism with being a “habitual offender.” Id.
That understanding of recidivism accords with the understanding relied on in Rummel. There, the Supreme Court rejected an Eighth Amendment challenge to a life sentence with the possibility of parole that had been imposed for a defendant's conviction for committing a third nonviolent felony. Rummel, 445 U.S. at 265. The defendant challenging that sentence had already served his sentences for his convictions for committing the earlier two offenses. Id. at 265-66. In upholding the defendant's life-with-parole sentence, the Supreme Court emphasized the special interest that a state has in imposing such a harsh sentence when the offender has already “demonstrate[d] that conviction and actual imprisonment [does] not deter him from returning to crime once he is released.” Id. at 278.
Rivera, by contrast, was sentenced to a prison term of more than 100 years for the § 924(c) convictions that he received at a single trial, Rivera-Ruperto II, 852 F.3d at 5, despite the fact that he had no prior criminal history, id. at 33 (Torruella, J., dissenting). And his additional sentence for his conviction for the other § 924(c) offense, for which he was tried separately, was imposed for conduct he had engaged in before he had served any time for his other § 924(c) offenses or even been charged with them. Id. at 5. As a result, his “forever” sentence was not premised, as the life sentence with the possibility of parole in Rummel was, on a state's determination that “actual imprisonment [would] prove[ ] ineffective” in dissuading the defendant from future law-breaking. Rummel, 445 U.S. at 278 n.17.
But, although Rivera's criminal conduct is not of the most serious kind, his no-hope sentence undoubtedly is. Indeed, his sentence could not have been harsher save for a sentence of death having been imposed. Yet, the Supreme Court has made clear that the Constitution does not permit a death sentence to be imposed for offenses that do not result in death. See Coker v. Georgia, 433 U.S. 584, 599 (1977) (reversing on Eighth Amendment grounds a sentence of death for a non-homicide crime).
Nor is the severity of Rivera's sentence solely a function of its length. His sentence is especially unforgiving because the sentencing judge was required to ignore any mitigating circumstances, like Rivera's lack of any criminal history prior to the sting. Rivera-Ruperto I, 846 F.3d at 420. Rivera's sentence in this respect is less forgiving than the life-without-parole sentence that Solem deemed disproportionate. That sentence was at least discretionary and therefore necessarily tailored to the defendant's particular circumstances, see Solem, 463 U.S. at 290, including most notably his prior criminal history.
So, what are we to conclude from a consideration of Solem's first criterion? Are the offenses that Rivera committed serious enough that the imposition of the most serious of prison sentences would not be grossly disproportionate?
Notably, Solem recognized the problem with calling upon judges to make this kind of abstract assessment. The range of criminal conduct that might reasonably be thought to be serious enough to warrant very severe punishment is broad. But, as one moves from consideration of crimes that involve core violent conduct to more boundary-pressing cases, judicial judgments about the relative severity of the crime necessarily risk becoming subjective.
Solem also appeared to recognize (even if it did not expressly hold) that this concern about judicial subjectivity is not properly addressed by simply requiring judges to uphold life-without-parole sentences so long as there is a rational basis to think the sentence is not grossly disproportionate. The cruelty and unusualness of punishment has long been understood to be determined, in part, by “evolving standards of decency,” which themselves become knowable in part through a consideration of the actual penal practices of comparable jurisdictions. See Miller v. Alabama, 567 U.S. 460, 469-70 (2012) (“[W]e view [Eighth Amendment proportionality] less through a historical prism than according to 'the evolving standards of decency that mark the progress of a maturing society.”' (quoting Estelle v. Gamble, 429 U.S. 97, 102 (1976))); see also Gregg v. Georgia, 428 U.S. 153, 175-83 (1976).
It is not surprising, then, that Solem appears to have proceeded on the understanding that judges need to undertake a real-world comparative inquiry, even if the more abstract threshold inquiry does not in and of itself demonstrate the sentence to be grossly disproportionate. For, at least in a case involving conduct such as is involved here, I read Solem to require courts to move beyond an abstract, threshold assessment of the “gravity of the offense and the harshness of the penalty,” Solem, 463 U.S. at 292, to a more grounded comparative assessment of how comparable crimes are actually treated both by the punishing jurisdiction and by other jurisdictions. And that is because I read Solem to require judges to undertake such a further inquiry if the question whether the sentence gives rise to an inference of gross disproportionality -- when viewed abstractly -- is at least fairly debatable.
This more holistic approach accords with the approach that is often taken in applying the Eighth Amendment. For, as I have noted, its bounds have long been understood to be drawn, at least in part, by actual legislative practices and by the norms of decency that those practices may be understood to reflect. See Graham v. Florida, 560 U.S. 48, 62 (2010). I turn next, then, to an assessment of the proportionality of this mandatory life-without-parole sentence in light of the two comparative criteria that Solem identifies. Those criteria train the focus of the inquiry on “the sentences imposed on other criminals in the same jurisdiction” and “the sentences imposed for commission of the same crime in other jurisdictions.” Solem, 463 U.S. at 292.
I begin by reviewing the sentences that the federal government imposes for other serious criminal conduct. That review suggests that, however debatable the question might be in the abstract, there is a gross disproportionality between the gravity of Rivera's offenses (serious though they are) and the severity of the punishment that he received for them.
Under federal law, “an aircraft hijacker ․ , a terrorist who detonates a bomb in a public place ․ , a racist who attacks a minority with the intent to kill and inflicts permanent or life threatening injuries ․ , a second-degree murderer, [and] a rapist,” Rivera-Ruperto II, 852 F.3d at 31 (Torruella, J., dissenting) (citation omitted), would all be subject to less harsh sentences than Rivera. Congress has not mandated that any of these offenders receive life-without-parole sentences. In fact, the recommended prison terms for each of these offenses under the United States Sentencing Guidelines are no more than one-fifth as long as the one that Rivera received for his offenses. See id. It is hard to see, though, how Rivera's conduct is five times as serious as that of a terrorist who detonates a bomb in a public building, seven times as serious as that of a person who inflicts life-threatening injuries on members of a racial minority because of their race, or eighteen times as serious as that of a rapist.
Consideration of the federal government's treatment of seemingly comparable conduct under § 924(c) itself further suggests that Rivera's sentence is grossly disproportionate. Rivera was involved in a series of putative drug transactions with, among other people, a group of FBI agents who were merely pretending to be drug traffickers. That the only person other than Rivera who was involved in each of the fake transactions was an FBI agent conducting a sting rather than an actual drug trafficker hardly makes Rivera's course of conduct more concerning than if he had been dealing with the same actual drug trafficker in each transaction. Yet, due to a quirk of conspiracy law and the way that it interacts with § 924(c), his involvement in an FBI-engineered sting rather than a true drug trafficking conspiracy dramatically increased his sentencing exposure under § 924(c).
Specifically, under our precedent, Rivera could not have been charged with participating in a single overarching conspiracy due to the way the FBI staged the sting. We have held that a conspiracy may not be between one individual and a government agent. United States v. Portela, 167 F.3d 687, 699-700, 700 n.8 (1st Cir. 1999) (“[G]overnment agents do not count as co-conspirators.” (quoting United States v. Giry, 818 F.2d 120, 125 (1st Cir. 1987)). But, in this sting, the only common participant in each transaction other than Rivera himself was an FBI agent.9 Thus, due to that quirk, the government could only charge Rivera with participating in the full course of his conspiratorial conduct by charging him with being a participant in six discrete conspiracies that corresponded to each of the six fake transactions.10
The decision to charge Rivera for his course of conduct in that manner was quite consequential. It helped to pave the way for the more-than-century-long mandatory prison sentence that he received under § 924(c). Each of his six § 924(c) convictions was predicated on one of the underlying drug conspiracy convictions that corresponded to Rivera's participation in one of the six fake drug transactions that the FBI staged.11
Notably, though, if Rivera had participated in the same type of extended conspiracy with a real drug trafficker standing in the stead of the FBI agent who was present for each of the six transactions, and if Rivera had then been charged with participating in a single, extended conspiracy for his course of conduct, he could have been sentenced under § 924(c) to a prison term of only five years for possessing a firearm in furtherance of that conspiracy. And that is because a single conspiracy conviction may not serve as the predicate for multiple § 924(c) convictions, United States v. Rodriguez, 525 F.3d 85, 111 (1st Cir. 2008) (holding that the Double Jeopardy Clause bars multiple § 924(c) offenses predicated on the defendant's conviction for participation in a single conspiracy), no matter how large or extended that predicate conspiracy happens to be.12
In this way, then, § 924(c) itself appears to treat the very same course of conspiratorial conduct in which Rivera engaged far more leniently depending on how that course of conduct happens to be charged. After all, Rivera received a mandatory sentence that is more than twenty-five times greater than the defendant in Rodriguez received. And Rivera received that sentence, even though, just like the defendant in Rodriguez, Rivera was found to have committed multiple acts of gun possession in the course of committing a predicate offense and even though these acts were as a functional matter part and parcel of a single -- somewhat extended -- criminal conspiracy. Compare Rivera-Ruperto II, 852 F.3d at 4-5, with Rodriguez, 525 F.3d at 93.
To be sure, in addition to his conspiracy convictions, Rivera was also convicted of six counts of attempted drug possession with intent to distribute, and those convictions independently served as predicates for his § 924(c) convictions. But the conduct underlying those predicate attempt convictions was itself part and parcel of the conduct that could have supported charging Rivera with participating in one extended conspiracy, had an FBI agent not been the only other party to the whole of it. And it is hard to see how those predicate convictions for attempted possession with intent to distribute a substance containing a detectable amount of cocaine in and of themselves show that Rivera's course of conduct was more than twenty-five times worse than that of a § 924(c) offender who, while conspiring with actual drug traffickers in a similarly extended conspiracy to possess cocaine with the intent to distribute, served as an armed lookout for each drug transaction but (unlike Rivera) never had “the power and intent to exercise control over” the cocaine. Henderson v. United States, 135 S. Ct. 1780, 1784; see United States v. Sliwo, 620 F.3d 630, 638 (6th Cir. 2010) (“The government only showed that Defendant was involved in a scheme, and the evidence of his participating in transporting the empty van and serving as a lookout would not allow a rational jury to find beyond a reasonable doubt that Defendant conspired to possess with intent to distribute marijuana.”); United States v. Penagos, 823 F.2d 346, 351 (9th Cir. 1987) (“Even if defendant acted as security and lookout ․ these actions do not indicate that he had dominion or control over cocaine.”).13 Yet, under Rodriguez, that latter offender at most could be sentenced to a prison term of five years under § 924(c) -- rather than the 130-year prison term that Rivera received -- if the prosecutor chose to treat that offender's course of conduct as evidencing his participation in one overarching conspiracy rather many discrete conspiracies. Rodriguez, 525 F.3d at 85.
This assessment of Rivera's mandatory sentence relative to the way that the federal government treats seemingly worse or at least comparable conduct does little to allay the concerns about disproportionality -- however debatable those concerns may be in the abstract -- that a consideration of the first Solem criterion raised. This comparison in turn raises the concern that the congressional choice to mandate this level of punishment for an offender like Rivera may not have been a carefully considered one. And that fact necessarily diminishes (even though it does not negate) the legislative claim to deference that informs the whole of the Solem framework. Solem, 463 U.S. at 3009 (“Reviewing courts, of course, should grant substantial deference to the broad authority that legislatures necessarily possess in determining the types and limits of punishments for crimes.”).
The final Solem criterion requires a comparison of this sentence with “the sentences imposed for commission of the same crime in other jurisdictions.” Solem, 463 U.S. at 292. A consideration of this criterion would also appear to point in favor of Rivera's challenge.
As the government did not address this prong of the Solem inquiry, the government does not address whether there is any state that would impose for comparable conduct the same draconian punishment that § 924(c) required the District Court to impose in this case. But, my own unaided review accords with Rivera's contention that this sentence is an outlier compared to the sentencing practices elsewhere in the United States. That review indicates that virtually all “drug and weapons crimes amenable to federal mandatory minimums are actually prosecuted in state courts pursuant to state laws carrying much lower sentences.” Erik Luna and Paul Cassell, Mandatory Minimalism, 32 Cardozo L. Rev. 1, 16 (2010); see also Rivera-Ruperto II, 852 F.3d at 19 (Torruella, J., dissenting).14
In addition, it appears that no country subject to the jurisdiction of European Court of Human Rights (ECtHR) may impose this sentence for any offense, let alone for an offense that is not of the most serious kind. Cf. Graham, 560 U.S. at 80 (“The Court has looked beyond our Nation's borders for support for its independent conclusion that a particular punishment is cruel and unusual.”). The ECtHR has held that the nations subject to its jurisdiction must ensure that all sentences (even for serial murderers and terrorists) respect a right to hope under Article 3 of the European Convention on Human Rights. Hutchinson v. United Kingdom, (No. 57592/08) Eur. Ct. H.R. ¶ 20 (Feb. 3, 2015), http://hudoc.echr.coe.int/eng?i=001-150778; Vinter & Others v. United Kingdom, 2013-III Eur. Ct. H.R. 317, 349-50.15
In accord with that right, the ECtHR ruled that all people facing “whole life” sentences must be afforded a “review which allows the domestic authorities to consider whether any changes in the life prisoner are so significant, and such progress towards rehabilitation has been made in the course of the sentence, as to mean that continued detention can no longer be justified on legitimate penological grounds.” Hutchinson, Eur. Ct. H.R. ¶ 20(a). The court declined to “prescribe the form -- executive or judicial -- which that review should take, or to determine when that review should take place.” Id. ¶ 20(b). But, the court emphasized that “comparative and international law materials provide clear support for the institution of a dedicated mechanism guaranteeing a review no later than twenty-five years after the imposition of a life sentence, with further periodic reviews thereafter[.]” Id. And the court added that “[a] whole life prisoner is entitled to know, at the outset of his sentence, what he must do to be considered for release and under what conditions, including when a review of his sentence will take place or may be sought [.]” Id. ¶ 20(d). Thus, the court explained that, “where domestic law does not provide any mechanism or possibility for review of a whole life sentence,” the unlawfulness of the sentence under Article 3 of the Convention “arises at the moment of the imposition of the whole life sentence and not at the later stage of incarceration.” Id.
Accordingly, consideration of the last two Solem criteria reinforces the concern about whether Rivera's sentence is grossly disproportionate that consideration of the first Solem criterion raises. The consideration of these last two criteria reveals that Rivera's severe sentence is most unusual when compared to the sentences that have been imposed for crimes that would seem to be no less serious. And that is so whether one looks to the sentencing practices of other jurisdictions or even to the sentencing practices of the federal government itself, which appears to punish conduct that is quite similar, and even seemingly worse, far less severely.
In the end, the question whether Rivera's sentence is constitutional under Solem is not without some difficulty. His crimes are more serious than the minor one that triggered the sentence that Solem struck down. But, Rivera received the harshest of prison sentences for crimes that Solem does not treat as being of the most serious kind. Moreover, comparative analyses reveal that his sentence is an outlier. I thus conclude that, if Solem were the last word, then Rivera's sentence would be grossly disproportionate. Under the Eighth Amendment, therefore, Rivera would be entitled to have his mandatory life-without-parole sentence vacated and his case remanded for resentencing.16
Solem, however, is not the last word. I thus must address the post-Solem Supreme Court precedent that addresses the constitutionality of imposing mandatory life-without-parole sentences under the Eighth Amendment for drug offenses. And that precedent is Harmelin.17
There, a defendant brought an Eighth Amendment proportionality challenge to his mandatory life-without-parole sentence under Michigan law for the possession of what the Supreme Court described as 672.5 grams of cocaine. Harmelin, 501 U.S. at 1008 (opinion of Kennedy, J.). Notwithstanding Solem, the Supreme Court upheld that sentence. Id. at 996 (opinion of Kennedy, J.).
Harmelin did not produce a majority opinion. Rather, a fractured Court yielded a controlling opinion that took the form of a three-Justice concurrence. Id.; see Graham, 560 U.S. at 59-60 (determining that Justice Kennedy's opinion in Harmelin is controlling). But that concurrence is still in my view dispositive in this case and in a manner that disfavors Rivera's challenge.18
The first way in which the Harmelin concurrence adversely affects Rivera's proportionality challenge has to do with the concurrence's treatment of the second and third Solem criteria. The concurrence makes clear that consideration of these two criteria -- which require real-world comparative 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.”19 Harmelin, 501 U.S. at 1005 (opinion of Kennedy, J.). The concurrence further indicates that there need only be a “rational basis” for a legislature's conclusion that an offense is as serious as one that, like felony murder, may constitutionally merit a life-without-parole sentence in order for the threshold Solem inquiry to require the conclusion that no such inference is warranted and thus that the sentence must be upheld. Id. at 1004.
I agree that a sentence's outlier status does not in and of itself demonstrate that a sentence is so grossly disproportionate as to be unconstitutional. But, as the discussion above demonstrates, there are consequences if judges are too quickly barred from gaining insight into whether a sentence is grossly disproportionate through a comparative analysis of other relevant sentencing practices. Those consequences are likely to be especially significant, moreover, in cases in which the offense is, per Solem, not of the most serious kind, but the prison sentence is.
In fact, all four dissenting Justices in Harmelin challenged the concurrence on this point. Id. at 1018-19 (White, J., dissenting); id. at 1027 (Marshall, J., dissenting); id. at 1028 (Stevens, J., dissenting). The dissenters explained that a virtue of the second and third criteria is that they help to inform the analysis of the first criterion. Id. at 1020-21 (White, J., dissenting). Under the Harmelin concurrence's approach, the dissenters worried, courts addressing the first Solem criterion would have “no basis for [a] determination that a sentence was -- or was not -- disproportionate, other than the 'subjective views of individual [judges],' which is the very sort of analysis our Eighth Amendment jurisprudence has shunned.” Id. at 1020 (quoting Coker, 433 U.S. at 592).
The dissenters also expressed the concern that the concurrence's approach to the first criterion -- by making it so difficult to make a showing that would justify undertaking a real-world comparative analysis -- threatened to render any objective Eighth Amendment proportionality analysis “futile.” Id. at 1020. Justice White even went so far as to contend that the concurrence's gloss on the first criterion was inconsistent with Solem because it reduced Solem to “an empty shell.” Id. at 1018.
Nevertheless, the dissenters did not prevail. I thus must, like the panel, Rivera-Ruperto II, 852 F.3d at 18, make the kind of critical threshold determination that the Harmelin concurrence requires. And that means that I must decide whether the severity of Rivera's § 924(c) sentence is so grossly disproportionate to the gravity of his underlying § 924(c) offenses that, as an abstract matter, it would not be rational for a legislature to conclude that such a sentence is at least as permissible as the same sentence would be for the offense of felony murder without the intent to kill. For, under the Harmelin concurrence, I am permitted to assess, in real-world terms, whether Rivera's sentence is an outlier -- and then to incorporate a determination that it is into the overall assessment of the sentence's proportionality -- only after first finding that this threshold Solem inquiry favors Rivera.
The second way in which the Harmelin concurrence adversely affects Rivera's Eighth Amendment challenge concerns the way in which the Harmelin concurrence actually performed Solem's threshold inquiry with respect to the criminal conduct at issue in that case. Specifically, the concurrence determined that the drug possession crime in that case was of a sufficiently “serious nature” that no inference of gross disproportionality was warranted by the imposition of a mandatory life-without-parole sentence. Harmelin, 501 U.S. at 1004 (opinion of Kennedy, J.). Accordingly, the concurrence concluded, the judicial inquiry into the sentence's proportionality need not reach the second or third Solem criteria. Id.
In making this critical judgment, the concurrence reasoned that the “[p]ossession, use, and distribution of illegal drugs represents 'one of the greatest problems affecting the health and welfare of our population.”' Id. at 1002 (quoting Treasury Emps. v. Von Raab, 489 U.S. 656, 668 (1989)). For that reason, the concurrence explained, Harmelin's crime “falls in a different category from the relatively minor, nonviolent crime at issue in Solem.” Id.
The concurrence stressed in this regard that the suggestion that the “crime was nonviolent and victimless ․ is false to the point of absurdity.” Id. The concurrence emphasized that 650 grams of cocaine contained “between 32,500 and 65,000 doses.” Id.
The concurrence further explained that the fact that the offense involved drug possession was important because “quite apart from the pernicious effects on the individual who consumes illegal drugs, such drugs relate to crime [.]” Id. For example, the concurrence reasoned, drug users may themselves commit crimes because of the effect of drugs on their cognitive state or to “obtain money to buy more drugs.” Id. In addition, “a violent crime may occur as part of the drug business or culture.” Id. Thus, the concurrence concluded, there was a basis for finding “a direct nexus between illegal drugs and crimes of violence.” Id. at 1003.20
The concurrence then concluded, without the benefit of any comparative inquiry into the practices of other jurisdictions, that whether or not Michigan's penalty scheme was “correct or the most just in the abstract sense,” 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.” Id. at 1003 (emphasis added). The Harmelin concurrence justified this conclusion by explaining that “a rational basis exists for Michigan to conclude that petitioner's crime is as serious and violent as felony murder without specific intent to kill,” which, the concurrence noted, is a crime that Solem had stated was one “for which 'no sentence of imprisonment would be disproportionate.”' Id. at 1004 (quoting Solem, 463 U.S. at 290 n.15) (emphasis added). Thus, the Harmelin concurrence held that, “[i]n 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.” Id. at 1005.
This reasoning, in my view, is dispositive here. Rivera's convictions are not for offenses that are identical to Harmelin's. Indeed, he was not convicted of actually possessing any drugs. Still, I do not see how a lower court may say that the Michigan legislature had reason to conclude that a conviction for possession of a large quantity of cocaine and no guns warranted a mandatory life-without-parole sentence, but that Congress could not have had a rational basis for concluding that such a sentence was warranted for multiple convictions for possession of a firearm in furtherance of conspiring or attempting to possess with intent to distribute a “detectable amount” of cocaine packaged in five-kilogram-sized substances. See 21 U.S.C. § 841(b)(1)(A)(ii)(I). Accordingly, I agree with the panel that, no matter how much of an outlier Rivera's sentence may be, we must affirm this sentence in light of Harmelin, and we must do so at the threshold of the Solem inquiry. See Rivera-Ruperto II, 852 F.3d at 18; cf. Hutto, 454 U.S. at 374-75 (“By affirming the District Court decision [deeming a sentence to be a violation of the Eighth Amendment] after our decision in Rummel, the Court of Appeals sanctioned an intrusion into the basic line-drawing process that is properly within the province of legislatures, not courts ․ [and] ignored ․ the hierarchy of the federal court system created by the Constitution and Congress.” (citation omitted)).21
Although I am convinced that the Harmelin concurrence controls the outcome here, and that it does so by limiting our inquiry to a consideration of only Solem's first criterion, I am also convinced that the Court should revisit the logic of the Harmelin concurrence, at least insofar as it applies to mandatory greater-than-life-without-parole sentences under § 924(c) in cases involving predicate drug offenses.22 That is so for three reasons.
First, given the range of possible ways that a defendant may commit multiple § 924(c) offenses, it is not realistic to posit that the Congress that enacted § 924(c) made a focused judgment that defendants like Rivera should receive a mandatory life-without-parole sentence for their drug-related criminal conduct. There was, by contrast, far more reason to believe in Harmelin that the legislature had made a focused penal judgment to mandate a life-without-parole sentence for the particular criminal conduct in which the defendant there had engaged.
Accordingly, the Harmelin concurrence's concern that “set[ting] aside [Harmelin's] mandatory sentence would require rejection not of the judgment of a single jurist ․ but rather the collective wisdom of the ․ Legislature and, as a consequence, the ․ citizenry,” Harmelin, 501 U.S. at 1006 (opinion of Kennedy, J.), is in my view less salient here. And, for that reason, it is less clear to me that simply because it might be rational for a legislature to think that Rivera's conduct warranted punishment as severe as the punishment that Harmelin received, a court should not proceed to assess how much of an outlier such a sentence is before determining whether that sentence violates the Eighth Amendment.
As the Harmelin concurrence noted, the life-without-parole sentence in that case was mandated pursuant to a carefully calibrated and graduated penalty scheme in which the Michigan legislature specially singled out only a subset of precisely defined large-quantity drug possession crimes for such harsh punishment. Michigan's penalty scheme, the concurrence explained, “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 address a most serious contemporary social problem.” Id. at 1007-08 (emphasis added). Thus, although the concurrence did acknowledge that it was not untroubled by the result, or certain “that Michigan's bold experiment [would] succeed,” the concurrence concluded that it could not “say the law before us has no chance of success and is on that account so disproportionate as to be cruel and unusual punishment.” Id. at 1008.23
Perhaps, in the face of the exercise of such legislative care to address a new social problem in a new way, there is a case to be made for according the kind of deference to the penal judgment at issue in Harmelin that the concurrence in Harmelin thought proper. And thus, perhaps, in such a circumstance, there is less need to check the judicial intuition about the proportionality of a mandatory life-without-parole sentence for a large-quantity drug possession offense against actual legislative practice than the dissenters in Harmelin thought there was.
But even if, in light of the legislative care taken in Michigan, the sentence at issue in Harmelin warranted such deferential review, uninfluenced by real-world sentencing practices, I cannot see what the case would be for applying the same limited form of review here. In contrast to the focused sentencing scheme considered in Harmelin, which targeted only carefully specified large-quantity drug possession crimes, § 924(c) criminalizes much conduct that -- given that statute's famously ambiguous scope -- is in its nature not similarly precisely knowable to legislators.24
Moreover, § 924(c) imposes a sentence as harsh as the one that Rivera received only because the statute requires the stacking of various individual § 924(c) sentences. As explained above, under § 924(c), a first conviction leads to a mandatory sentence of five years, and each “second or subsequent conviction” mandates an additional twenty-five year prison term that must be served consecutively. 18 U.S.C. §§ 924(c)(1)(A)(i), 924(c)(1)(C)(i).
In consequence, life-without-parole sentences may be required under § 924(c) for an astoundingly wide array of possible offense combinations, including mixes potentially of both state and federal offenses and various combinations of predicate drug offenses, whether or not paired with “crime[s] of violence.” See id. § 924(c)(1)(A). For this reason, too, it is pure fiction to imagine that Congress, in requiring a sentence of imprisonment for more than 100 years with no chance of parole, was focused on the type of drug-related conduct at issue in this case in the way that the Harmelin concurrence understood the Michigan legislature to have been focused on the much more precisely defined type of drug-related conduct singled out for harsh punishment in that case.
Nor is there anything in § 924(c)'s legislative history to indicate that Congress, in enacting § 924(c), gave the kind of focused consideration to potential sentencing implications in a case of this sort that the concurrence in Harmelin plainly thought that the Michigan legislature had given to the type of case presented there. Section 924(c) in its original form -- before the statute was amended to add “drug trafficking crime[s]” as predicate offenses -- was introduced as a floor amendment. Simpson v. United States, 435 U.S. 6, 13 (1978); 114 Cong. Rec. 22231 (1968). No mention was made in the ensuing floor “debate” of the feature of this statute that results in the imposition of mandatory life-without-parole sentences for conduct at all like Rivera's -- namely, the “second or subsequent” provision at issue here. 114 Cong. Rec. 22231 (1968). Nor was any mention made of the draconian results that could follow from the “stacking” of § 924(c) sentences, let alone of the Eighth Amendment implications of doing so when multiple § 924(c) convictions are handed down at a single trial or across a pair of trials and thus before the defendant has served time for any of them and demonstrated that punishment will not deter him from future criminal conduct.25
The concern that Congress did not give focused consideration is not allayed by the text of § 924(c). Its use of the curious “second or subsequent” phrase hardly reveals that Congress must have foreseen a result such as this one in amending the statute to encompass defendants who were involved not in committing “crimes of violence” but only in inchoate drug offenses. Anti-Drug Abuse Act of 1986, Pub. L. No. 99–570, 100 Stat. 3207 (codified at 21 U.S.C. § 801). In fact, for more than a decade after that amendment, due to the oddness of that statute's original text (“second or subsequent”), it was a matter of great uncertainty in the lower courts as to whether § 924(c) even allowed the stacking of sentences when multiple convictions were handed down at one trial. See, e.g., United States v. Deal, 954 F.2d 262, 263 (5th Cir. 1992) (citing cases), aff'd, 508 U.S. 129 (1993).
The Supreme Court in Deal did finally reject the view of some lower courts -- and the four dissenters in that case, 508 U.S. at 137 (Stevens, J., dissenting) -- that Congress intended only to impose such harsh sentences on true recidivists, such that a “defendant who commits a second § 924(c) offense before trial on the first would not be eligible for sentence enhancement[.]” Id. at 145 (Stevens, J., dissenting). But, Deal hardly reveals that Congress must have had in mind the notion that greater-than-life sentences would be mandatorily imposed for offenses committed in circumstances remotely like those involved here. Deal's only functional explanation for why it would make sense to read the “second or subsequent” language to encompass even an offender who is charged cumulatively for seemingly-related conduct, rather than only a true recidivist, took the form of the example of an offender who, through stealth, manages to evade detection in repeatedly committing unrelated crimes of violence -- namely, bank robberies. Id. at 137. And there is nothing to indicate that Congress has subsequently ratified that previously sharply-contested conclusion about what “second or subsequent” means in any way that would suggest that Congress did so while focused on the type of conduct that is at issue here, given that Rivera's acts were part and parcel of a single, albeit extended, course of conspiratorial conduct.
Moreover, unlike the scheme at issue in Harmelin, § 924(c) subjects offenders to mandatory life-without-parole sentences even for predicate drug offenses that -- like ones for conspiracy -- are inchoate. Prosecutorial decisions about whether to treat a series of events as part of one conspiracy or as multiple discrete offenses, however, can lead to wildly different sentencing outcomes under § 924(c), even though comparable conduct has occurred and is being punished. Cf. Deal, 508 U.S. at 134 n.2 (emphasizing the distinction between a prosecutor's “universally available and unavoidable power to charge or not to charge an offense” and the possibility of an “extraordinary new power to determine the punishment for a charged offense by simply modifying the manner of charging.”).
Thus, this sentencing regime is very different from the one at issue in Harmelin, in which the state legislature “mandated the penalty” for a discrete drug possession crime. 501 U.S. at 1006 (opinion of Kennedy, J.). Here, there is a real possibility that, in upholding a more-than-century-long sentence based on multiple related § 924(c) offenses, we uphold not so much a legislative determination to punish the relevant conduct this severely as a prosecutorial one to divvy that same conduct up into a series of discrete charges that, if proved, will require the stacking of a series of stiff sentences that cumulatively will exceed 100 years.26
I am troubled that the “forever” sentence that results from such charges must be upheld on the basis of only the abstract and highly deferential threshold inquiry that Harmelin limits us to undertaking. And yet, under that constricted inquiry, judges have no choice but to approve mandatory “forever” sentences under § 924(c) so long as they can hypothesize a rational reason for the legislature to have thought that the underlying criminal conduct was as serious as the large-quantity drug possession at issue in Harmelin.
Simply put, it is one thing to uphold such a sentence for the drug-related conduct at issue here on the basis of a limited and abstract threshold inquiry when that sentence has been legislatively “calibrated with care, clarity, and much deliberation to address a most serious contemporary social problem.” Harmelin, 501 U.S. at 1007-08 (opinion of Kennedy, J.). It seems to me quite another to do so when that sentence does not appear to have been the product of such serious and careful legislative thought and may in fact have been the result of an exercise of a prosecutor's decision to break one course of conduct into many discrete offenses. For, in that event, the judge in rejecting a challenge to the sentence's proportionality is deferring to a hypothesized legislative choice, notwithstanding that there in fact may be no legislature -- not even the one imposing the sentence -- that has both thoughtfully focused on the need for such a sentence for such conduct and then carefully chosen to mandate it as a proportionate response.
There is a second reason for my concern about applying the constricted form of the analysis that the Harmelin concurrence requires in this case. Harmelin was decided at a time at which, on the concurrence's own account, a state was trying out a new means of responding to a serious crime problem that was causing great concern. Id. In that circumstance, the concurrence expressed its understandable wariness about the federal Constitution's proportionality requirement being construed in a manner that would invalidate one state's “bold experiment” and thereby stifle the kind of innovation that our federal system invites. Id. at 1008. The concurrence thus declined to permit Michigan's outlier status to be held against it, as doing so hardly seemed consistent with the notion instinct in our system of federalism -- that states are laboratories of democracy. See New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting); see also Harmelin, 501 U.S. at 1009.
But, here, we are considering a federal statutory sentencing mandate. And that mandate bears none of the hallmarks of considered experimentation, undertaken as a means of fashioning a bold, if untried, response to a new and vexing problem. In fact, this mandate's dramatic sentencing consequences result in significant part from a judicial construction of a much debated statutory phrase -- “second or subsequent” -- that was the subject of seemingly little discussion in Congress.
Moreover, we are reviewing that mandate's proportionality at a time when decades have passed since the Supreme Court first considered Michigan's arguably similar approach to combating the drug scourge through the imposition of mandatory life-without-parole sentences. Yet, during those intervening years, virtually no other jurisdiction has seen fit to follow suit. Indeed, if anything, the trend lines are moving in just the opposite direction. See Bullock, 485 N.W.2d at 877; cf. Graham, 560 U.S. at 109 (looking to “legislative trends” in determining whether a sentencing practice violated the Eighth Amendment).
Thus, for this reason, too, the concerns that appear to have animated the Harmelin concurrence's conclusion that a real-world comparative inquiry was not properly undertaken in that case do not appear to me to be present here. Rather, in a case like this, it seems to me that there is good reason for courts to undertake the holistic review that the dissenters in Harmelin understood Solem to require but that the Harmelin concurrence determined was not needed to review a mandatory life sentence that a state's legislature was thought to have required as a “bold experiment” to address the drug problem. Harmelin, 501 U.S. at 1008 (opinion of Kennedy, J.). By doing so, courts may factor the sentence's evident outlier status into the ultimate assessment of its gross disproportionality.
These two concerns about applying the Harmelin concurrence's gloss on the Solem inquiry to this context are reinforced, in my view, by two lines of Supreme Court precedent that have developed since Harmelin was decided. I briefly describe each in turn.
First, in Alleyne v. United States, 570 U.S. 99 (2013), the Court held that the Sixth Amendment requires that “any fact that increases the mandatory minimum is an 'element' that must be submitted to the jury.” Id. at 103. Thus, under Alleyne, the minimum sentence that a defendant can receive must be based on the minimum conduct criminalized by a statute, that is, the elements of that crime. Id. at 116. It would thus seem that, in evaluating the proportionality of a particular mandatory sentence, we must likewise look to the least of the conduct criminalized by the elements of the offense. Consideration of anything further -- like conduct alleged in the indictment or found at sentencing to have occurred -- would impermissibly permit the assessment of the sentence's proportionality to be based on conduct that had not been found by a factfinder beyond a reasonable doubt, even though under Alleyne the mandatory sentence may not be imposed based on such conduct.
The concurrence in Harmelin did not have the benefit of Alleyne. But, insofar as Alleyne indicates that the focus must be on the least of the conduct criminalized in evaluating a sentence's proportionality, the potential consequences of following the Harmelin concurrence's extremely deferential approach in the context of § 924(c) become even more concerning.
Consider in this regard that, seemingly contrary to Alleyne's logic, the Harmelin concurrence reasoned that the sentence there at issue was not disproportionate because Harmelin “possessed” 672.5 grams of “undiluted cocaine” as well as assorted drug paraphernalia, 501 U.S. at 1008 (opinion of Kennedy, J.). The concurrence emphasized that fact in spinning out how many “doses” of the drug could have been dispensed by the defendant. See id. And, the concurrence did so in order to describe the seriousness of the tangible harm caused by the defendant's conduct and thus the reasonableness of the legislative sentencing judgment. See id.
The offense in that case, however, actually held Harmelin criminally liable merely because he “possessed” 650 grams of a “mixture containing” cocaine, Mich. Comp. Laws Ann. § 333.7403(2)(a)(i). But, of course, such a mixture could contain a much smaller amount of the actual drug. Id.; People v. Puertas, 332 N.W.2d 399, 400 (Mich. App. 1983). And thus the least of the conduct criminalized there could not have caused the same harm that the concurrence attributed to the defendant's actions.
Still, there is no doubt that the Michigan legislature did intend to mandate a life-without-parole sentence for even that “mixture” crime, given how clearly the statute at issue set forth that penalty scheme. By contrast, it is less clear to me that Congress would have been fully aware of just how minimal the conduct could be that would result in a “forever” sentence under § 924(c). As I have explained, § 924(c)'s scope is notoriously ambiguous, the statute encompasses even inchoate crimes, and it requires the “stacking” of mandatory sentences even for related conduct that results in multiple convictions at a single trial due to a prosecutorial choice to divvy up the conduct. Thus, in light of how Alleyne suggests proportionality review must now proceed, there is additional reason to doubt that Congress, in enacting this sentencing regime, contemplated the full implications of its mandate, even if that mandate does encompass a range of cases involving more serious conduct that Congress no doubt had in view.
The second line of post-Harmelin cases that I have in mind further gives me pause about applying the Harmelin concurrence's more limited form of Solem review here. This line of precedent has resulted in the invalidation under the Eighth Amendment of life-without-parole sentences for juveniles. Miller, 567 U.S. at 474; Graham, 560 U.S. at 69.
Those cases, of course, are by no means controlling here. But, in them, the Court has emphasized in a way that it had not previously -- and thus in a way that it had not when Congress enacted § 924(c) -- that life sentences without the possibility of parole raise special constitutional concerns.
In particular, the Court has explained that such sentences constitute some of the “most severe punishments” that society imposes. Miller, 567 U.S. at 474; Graham, 560 U.S. at 69. And, the Court has added, such sentences:
[S]hare some characteristics with death sentences that are shared by no other sentences ․ [T]he sentence alters the offender's life by a forfeiture that is irrevocable. It deprives the convict of the most basic liberties without giving hope of restoration ․ [A] life without parole sentence ․ means denial of hope; it means that good behavior and character improvement are immaterial; it means that whatever the future might hold in store for the mind and spirit of [the convict], he will remain in prison for the rest of his days.
Graham, 560 U.S. at 69–70 (citations and quotations omitted). The Court has also recently stressed, in connection with reviewing the proportionality of such sentences, that “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.” Id. at 69.
It may be that, even despite these strong statements, the Eighth Amendment is still best understood to permit Congress to mandate, even for conduct like Rivera's that resulted in no bodily harm, that “whatever the future might hold” for him, he must “remain in prison for the rest of his days.” Id. at 70. He was, after all, an adult, not a child, when he committed his crimes. And judges are not entitled to second-guess the wisdom of the penal judgments of legislatures. Instead, judges are supposed to accord them deference.
But, at least in a case involving a sentence this harsh for crimes of this type, one would think that such deference would stem from confidence that the legislature has in fact made a considered penal judgment to impose such an unforgiving sentence and from careful consideration of the way in which offenders more generally are punished for comparable or even worse conduct. For such confidence and consideration would ensure that judges in deferring to a legislative judgment are recognized to be engaged in an understandable, rather than an unforgivable, means of carrying out their duty to say what the constitutional prohibition against “cruel and unusual” punishment is.
Thus, in light of the concerns that the Court has recently expressed about the imposition of life-without-parole sentences, I do not see how the kind of abstract review that is contemplated under the first Solem criterion -- and that the Harmelin concurrence requires us to treat as dispositive here -- can suffice to permit us to determine whether Rivera's sentence is grossly disproportionate under the Eighth Amendment. In my view, a comparative assessment, grounded in actual legislative practice, should be required to inform the judge's assessment of proportionality in such a case.
Such a requirement would prevent judges from simply substituting their own preferences for legislative ones in evaluating whether a mandatory life-without-parole sentence is cruel and unusual. Such a requirement would also ensure that the judicial assessment of a mandatory life-without-parole sentence for drug-related offenses of the sort at issue here does not unduly discount the defendant's Eighth Amendment right to be protected from grossly disproportionate punishment.
Rivera faces the longest and most unforgiving possible prison sentence for conduct that, though serious, is not of the most serious kind. He does so not because the legislature had authorized its imposition and a judge had then considered all of the aggravating and mitigating circumstances and determined that this sentence was appropriate. He does so only because Congress has been deemed to have made a blanket judgment that even an offender like Rivera -- who has no prior criminal record and whose series of related crimes resulted in no harm to an identifiable victim -- should have no hope of ever living free. And he does so even though virtually every comparable jurisdiction punishes comparable criminal conduct less harshly, and even though the federal government itself punishes nearly the same or seemingly worse conduct more leniently.
Almost three decades have now passed since the concurring Justices in Harmelin concluded, without reference to real-world comparative benchmarks, that the Eighth Amendment afforded the Michigan legislature the scope to try out what at the time was viewed as a permissible sentencing experiment to address a newly concerning crime problem. In those intervening decades, virtually no jurisdiction has been willing to replicate that state's experiment. In fact, even the state that the Harmelin concurrence permitted to try it has abandoned it. And yet the Harmelin concurrence still controls.
In my view, a consequence as grave as the one that Harmelin requires in a case like this should have the imprimatur of more than only a nearly three-decade old, three-Justice concurrence. I thus urge the Supreme Court to consider whether the Eighth Amendment permits, at least in a case such as this, the mandatory stacking of sentences under § 924(c) that -- due to their cumulative length -- necessarily results in the imposition of a mandatory sentence of life without parole.
LIPEZ, Circuit Judge, statement regarding the denial of rehearing. In voting to deny panel rehearing, I express my agreement with the concurring statement issued by my colleagues in denying appellant's petition for en banc review.
By the Court:
Margaret Carter, Clerk
1. Rivera was indicted for six counts of violating 18 U.S.C. § 924(c) along with a variety of other charges also stemming from those fake drug transactions. Rivera-Ruperto II, 852 F.3d at 5. Those other charges consisted of six counts each of conspiracy to possess with intent to distribute five kilograms or more of a controlled substance, 21 U.S.C. §§ 841(a)(1) & (b)(l)(A)(ii)(II), 846; six counts of attempt to possess with intent to distribute five kilograms or more of a controlled substance, id. §§ 841(a)(1) & (b)(l)(A)(ii)(II), 846; and one count of possession of a firearm with an obliterated serial number, 18 U.S.C. § 922(k). For both the conspiracy and attempt charges, the controlled substance was “5 kilograms or more of a mixture or substance containing a detectable amount of ․ cocaine.” 21 U.S.C. § 841(b)(l)(A)(ii)(II). The charges were then divided between two separate trials.The first trial concerned the charges arising out of five of the six fake drug transactions, including five of the § 924(c) counts. Id.; United States v. Rivera-Ruperto, 846 F.3d 417, 420-21, 423 (1st Cir. 2017) (Rivera-Ruperto I). The first trial resulted in Rivera being sentenced to a term of imprisonment of 126 years and ten months, of which 105 years were for his convictions under § 924(c). Rivera-Ruperto I, 846 F.3d at 420-21. The second trial concerned the charges arising out of the other fake drug transaction and resulted in Rivera being sentenced to a term of imprisonment of thirty-five years -- of which twenty-five years were for his sixth § 924(c) conviction. Id. at 420. This sentence was to be served consecutively with his first term of incarceration. Id.Rivera appealed both of his convictions. Id.; Rivera-Ruperto II, 852 F.3d at 5. Because of the order in which this Court decided his two appeals, “Rivera-Ruperto I“ refers to his appeal from his second trial, which resulted in a sentence of thirty-five years' imprisonment, and “Rivera-Ruperto II“ refers to his appeal from his first trial, which resulted in a sentence of 126 years and ten months imprisonment.
2. Among the individualized factors that the sentencing judge could have considered if he had the discretion to do so are the fact that, although Rivera was caught up in a sting designed to catch corrupt police officers, he was himself not a police officer, Rivera-Ruperto II, 852 F.3d at 4 n.3, that Rivera had no prior criminal history, id. at 20 (Torruella, J., dissenting), that he caused no physical harm to any identifiable victim, id. at 35, and that he was never involved with any real drugs, id. at 19.
3. United States v. Abbott, 30 F.3d 71, 72 n.1 (7th Cir. 1994) (recognizing the district court's concern that the defendant's sentence of twenty-six years for, in part, a § 924(c) conviction was an “illustration of the lack of wisdom in mandatory minimum sentences, but I cannot take it upon myself to change the law that Congress has written because I think it is an inappropriate disposition.”); see also United States v. Angelos, 345 F. Supp. 2d 1227, 1247 (D. Utah 2004), aff'd, 433 F.3d 738 (10th Cir. 2006) (affirming a sentence of sixty-one years under § 924(c) despite its conclusion that § 924(c)'s sentencing requirements were “irrational”).
4. In both Rivera-Ruperto II and Rivera-Ruperto I, Rivera challenged the cumulative length of his sentence -- 161 years and ten months (of which 130 years stem from his § 924(c) convictions) -- as disproportionate under the Eighth Amendment. Rivera-Ruperto I, 846 F.3d at 425; Rivera-Ruperto II, 852 F.3d at 13. During both appeals, the parties and the panel considered his sentence cumulatively in addressing his Eighth Amendment challenge. See Rivera-Ruperto I, 846 F.3d at 425; Rivera-Ruperto II, 852 F.3d at 13. In his petition for rehearing in each case, Rivera likewise challenges the proportionality of his sentence cumulatively. I note that, even considering only the sentence that Rivera received in Rivera-Ruperto II, he still faces an inherently greater-than-life sentence -- 105 years -- solely on the basis of his § 924(c) convictions. Rivera-Ruperto II, 852 F.3d at 13. Moreover, because that more-than-100 year sentence was imposed before Rivera was sentenced in Rivera I for his sixth § 924(c) conviction, his additional twenty-five year sentence for that sixth § 924(c) conviction was necessarily a sentence that would extend his already-imposed 100-year-plus § 924(c) sentence another twenty-five years. Thus, it makes sense to consider those two sentences cumulatively for purposes of addressing his Eighth Amendment challenge.
5. Prior to Rummel, in Weems, “the Court had struck down as cruel and unusual punishment a sentence of cadena temporal imposed by a Philippine Court. This bizarre penalty, which was unknown to Anglo-Saxon law, entailed a minimum of 12 years' imprisonment chained day and night at the wrists and ankles, hard and painful labor while so chained, and a number of 'accessories' including lifetime civil disabilities.” Solem v. Helm, 463 U.S. 277, 306-07 (1983). In Solem, as in Rummel, the Court determined that the holding in Weems could not “be wrenched from the facts of that case.” Id. (quoting Rummel, 445 U.S. at 273).
6. As a result of the length of Rivera's sentence, I do not address how long a sentence imposed on an adult defendant must be in order for that sentence to be one that necessarily constitutes a life sentence. There can be no question, after all, that a sentence of more than one hundred years is properly considered to be a life sentence for any adult defendant, no matter that defendant's age at sentencing.
7. Prior to Solem, the Court had arguably adopted a much stricter test for determining whether a sentence might run afoul of the Eighth Amendment due to its length alone. In upholding the life sentence with parole in Rummel, the Court explained that there were serious concerns about judicial line-drawing presented by challenges to the proportionality of sentences based on the number of years a defendant had been sentenced to imprisonment for a particular offense. Rummel, 445 U.S. at 275. The Court thus expressed the view that successful proportionality challenges of that type would be “exceedingly rare,” id. at 273, and the dissent identified as an example the extreme case of a sentence of “[e]ven one day in prison ․ for the 'crime' of having a common cold.” Id. at 291 (Powell, J., dissenting) (quoting Robinson v. California, 370 U.S. 660, 667 (1962)). The Court then relied on Rummel's seemingly strict test in Hutto v. Davis, 454 U.S. 370 (1982), to summarily reverse a court of appeals decision that had invalidated on Eighth Amendment grounds a forty-year sentence for possessing nine grams of marijuana. Id. at 375. But, as Solem post-dates both Rummel and Hutto, Solem would set forth the controlling test -- especially in a case involving a proportionality challenge to a mandatory life-without-parole sentence -- absent some intervening precedent, such as, as I will explain, Harmelin v. Michigan, 501 U.S. 957 (1991), represents.
8. The Court had no occasion to do so, as none of the defendant's offenses were drug related and the statute at issue in Solem imposed a mandatory life sentence on those convicted of multiple felonies -- including drug-related felonies -- if “one or more of the prior felony convictions was for a crime of violence.” S.D. Codified Laws § 22-7-8; Solem, 463 U.S. at 299 n.26. Moreover, South Dakota's definition of a “crime of violence” did not include heroin dealing, or, for that matter, any drug offenses. S.D. Codified Laws § 22-1-2(9).
9. To be precise, while two of Rivera's co-defendants participated in two transactions with Rivera, no defendant, other than Rivera himself, was present at all six transactions in which Rivera participated.
10. Rivera was not the only defendant caught up in this FBI sting to have been exposed to a much longer sentence due to how this quirk of conspiracy law interacts with § 924(c). See e.g., United States v. Diaz-Castro, 752 F.3d 101, 107 (1st Cir. 2014); United States v. Gonzalez-Perez, 778 F.3d 3, 10 (1st Cir. 2015).
11. I note that, not long after Rivera was arraigned, the government made him a plea offer of fourteen years' imprisonment for all of his charged offenses. After negotiations, the government then agreed to reduce the offer to twelve years. Rivera rejected that offer, however, and proposed a counteroffer of eight years instead, which the government declined to accept. Later, the government renewed its twelve-year offer, but Rivera again rejected it. Rivera-Ruperto II, 852 F.3d at 6. The government then gave Rivera one final offer of eighteen years' imprisonment, which he rejected. Id. at 7.
12. The large majority of circuits apply this same rule. Rodriguez, 525 F.3d at 111; UnitedStates v. Lindsay, 985 F.2d 666, 674 (2d Cir. 1993); United States v. Diaz, 592 F.3d 467, 471-75 (3d Cir. 2010); United States v. Baptiste, 309 F.3d 274, 279 (5th Cir. 2002); United States v. Taylor, 13 F.3d 986, 994 (6th Cir. 1994); United States v. Cappas, 29 F.3d 1187, 1191 (7th Cir. 1994); United States v. Fontanilla, 849 F.2d 1257, 1258–59 (9th Cir. 1988); United States v. Moore, 958 F.2d 310, 312 (10th Cir. 1992); United States v. Hamilton, 953 F.2d 1344, 1346 (11th Cir. 1992); United States v. Anderson, 59 F.3d 1323, 1327 (D.C. Cir. 1995) (en banc); but see United States v. Camps, 32 F.3d 102, 108–09 (4th Cir. 1994); United States v. Lucas, 932 F.2d 1210, 1223 (8th Cir. 1991).
13. In Rivera's case, an FBI agent handed him the bag that held the sham cocaine for him to weigh, thus ensuring that “[i]n every transaction ․ he held the [sham cocaine] in his hands.”
14. The only present exceptions of which I am aware appear to be Alabama, see Ala. Code § 13A-12-231(2)(d) (a person convicted of possession of over ten kilograms of controlled substance shall be imprisoned for life without parole); see also Alabama Bd. of Pardons & Paroles v. Smith, 25 So. 3d 1198, 1201 (Ala. Crim. App. 2009) (applying Alabama statute to conspiracy convictions), and Mississippi, see Miss. Code. Ann. § 41-29-139(f) (imposing a punishment of ten to forty years without parole on drug trafficking offenses); Arnold v. State, 225 So. 3d 561, 564 (Miss. Ct. App. 2017) (applying Mississippi statute to conspiracy convictions); McDonald v. State, 921 So. 2d 353, 356 (Miss. Ct. App. 2005) (permitting “stacking” of sentences under Mississippi law). Most states that impose mandatory life sentences without the possibility of parole do so only for those convicted of violent crimes, or for true recidivists. See Rummel, 445 U.S. at 278 & n.17; American Civil Liberties Union, A Living Death: Life without Parole for Nonviolent Offenses 98 (2013). In fact, even Michigan, which passed the law mandating a life sentence for drug possession offenses that was affirmed by the United States Supreme Court in Harmelin, see 501 U.S. at 1007-08 (opinion of Kennedy, J.), no longer has such a law. The year after the Supreme Court decided Harmelin, the Michigan Supreme Court struck down the very sentencing provision that Harmelin had upheld as unconstitutional under the Michigan Constitution's Eighth Amendment analog. People v. Bullock, 485 N.W.2d 866, 877 (Mich. 1992). The Supreme Court of Michigan concluded “largely for the reasons stated by Justice White in his dissenting opinion in Harmelin that the penalty at issue here is so grossly disproportionate as to be 'cruel or unusual.”' Id. at 875-76 (quoting Mich. Const. Art. I § 16).
15. Article 3 of the European Convention on Human Rights states: “No one shall be subject to torture or to inhuman or degrading treatment or punishment.” We note that, even prior to Vinter's ruling regarding Article 3, nine nations within the jurisdiction of the ECtHR allowed no life sentences at all. Vinter, 2013-III Eur. Ct. H.R. at 338. Of those that did, the majority had mandatory mechanisms to review such sentences after a fixed number of years, and only five had any provision at all for life sentences without the possibility of release. Id. We note, too, that Hutchinson and Vinter addressed sentences in the context of defendants who had been convicted of the most serious of crimes: murder. Id. at 327, 328, 329; Hutchinson, Eur. Ct. H.R. ¶ 6.
16. I acknowledge that crafting a remedy in this case would not be without difficulty. Rivera's more-than-a-century-long sentence was imposed pursuant to § 924(c)'s consecutive sentences requirement, which raises challenges about how it could be rendered constitutional without producing arbitrary results. And these challenges are aggravated by the fact that Rivera was sentenced at one trial based on five of the § 924(c) convictions (when he was thirty-nine years old) and sentenced at a separate trial for the sixth conviction. Nevertheless, as Justice White's dissent in Harmelin made clear, post-Solem, in the rare cases in which a sentence violates the Eighth Amendment due to its length, courts can vacate the sentence and remand for resentencing as a remedy. See Harmelin, 501 U.S. at 1016 n.2 (White, J., dissenting); see also Miller, 567 U.S. at 489 (reversing and remanding for further proceedings a sentence found unconstitutional); Graham, 560 U.S. at 82 (same). In Bullock, moreover, in which the Supreme Court of Michigan struck down under the Michigan Constitution the mandatory life-without-parole sentence that was at issue in Harmelin, the remedy was simply to remove the prohibition against parole eligibility. Bullock, 485 N.W.2d at 877-78. And, of course, the problem of crafting a remedy arises in any case in which mandatory consecutive sentencing results in a disproportionate sentence relative to the underlying crime, yet both Solem and Harmelin appear to contemplate that a remedy may be needed for consecutive sentences if the resulting disproportionality is severe enough. See Harmelin, 501 U.S. at 1001 (opinion of Kennedy, J.); Solem, 463 U.S. at 297.
17. Other post-Solem Supreme Court cases do address the proportionality of life sentences. However, only Harmelin has ever upheld the mandatory imposition of such a sentence for a comparable crime where there was no possibility of parole.
18. A majority of the Supreme Court has not in any clear way embraced the reasoning of the Harmelin concurrence. Certainly Graham had no occasion to do so. Graham invalidated the life-without-parole sentence imposed on a juvenile under a special variant of the Eighth Amendment's proportionality test that applies when a sentencing practice, rather than a sentence in a particular case, is being challenged as disproportionate in all cases. 560 U.S. at 90-91. Similarly, Ewing upheld a mandatory twenty-five-years-to-life sentence under California's three strikes law. 538 U.S. at 20. Ewing explained that the concurrence in Harmelin “guide [s] our application of the Eighth Amendment in the new context that we are called upon to consider.” Id. at 23-24. Notably, however, Ewing did not hold that the Court was adopting the concurrence's approach. Id.
19. The Harmelin concurrence read Rummel as helpful to its position. Harmelin, 501 U.S. at 1005 (opinion of Kennedy, J.). Rummel noted that interjurisdictional analyses raise “complexities” and do not, alone, suffice to demonstrate that a sentence is disproportionate, given that “some State will always bear the distinction of treating particular offenders more severely than any other State.” Rummel, 445 U.S. at 282.
20. Harmelin noted that in Solem the Court contrasted the “minor” offenses for which the defendant had been convicted 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.” Harmelin, 501 U.S. at 1002 (quoting Solem, 463 U.S. at 299 & n.26). But, as I read this passage, Solem was not holding that a mandatory life-without-parole sentence for such a crime would be proportionate, whether or not the defendant was a true recidivist, as Solem was simply explaining that no argument had been made on that point. Solem, 463 U.S. at 299 n.26. Nor was Solem, in referencing “other violent criminals,” id., impliedly indicating that heroin dealers are themselves properly considered “violent” under Solem's rubric. In referring to “other violent criminals,” Solem appears to have been merely describing how the sentencing regime at issue in Solem operated, as under that regime a defendant had to have engaged in some violent conduct that led to a previous conviction, in addition to having been convicted of a fourth felony offense, even if that fourth offense was a nonviolent one. See S.D. Codified Laws § 22-7-8.
21. I note that I read the Harmelin concurrence to equate its conclusion that Michigan had a rational basis for deciding the defendant's drug-related conduct was as serious as the offense of felony murder with no intent to kill with its ultimate conclusion that the sentence that Michigan imposed for that conduct does not even give rise to an inference that such a sentence was grossly disproportionate. See Harmelin, 501 U.S. at 1005 (Kennedy, J., concurring). The concurrence's equation of those two conclusions compels, in my view, the conclusion, as a matter of precedent, that no such inference can be drawn as to this sentence either and thus that a comparative inquiry under Solem's second and third criteria is prohibited here just as the Harmelin concurrence concluded that it was prohibited there. Nevertheless, it is not clear to me that it is warranted as a general matter to equate the conclusion that there is a rational basis to deem a sentence proportionate with the conclusion that the sentence does not even give rise to an inference that it is disproportionate. After all, in challenges to the sufficiency of the evidence, courts routinely determine that the challenge ultimately fails because a jury could rationally find the evidence to have been sufficient to prove the defendant guilty beyond a reasonable doubt, even though a rational jury could also have drawn a reasonable inference from the evidence that would have resulted in an acquittal. See Musacchio v. United States, 136 S. Ct. 709, 715 (2016).
22. The Court's post-Harmelin affirmances in Ewing and Lockyer of mandatory twenty-five-year-to-life sentences with the possibility of parole under California's three strikes law offer little guidance here. Unlike in Ewing or Lockyer, we are dealing in this case with an offender with no prior criminal history. Ewing, 538 U.S. at 20; cf. Lockyer, 538 U.S. at 66. Additionally, the defendants in both Ewing and Lockyer retained the right to parole, and thus did not face sentences of equal severity to the one imposed here or in Harmelin. Ewing, 538 U.S. at 16; Lockyer, 538 U.S. at 74. Thus, in my view, Harmelin alone is our guide here.
23. I note that Ewing also emphasized the focused and deliberative nature of the judgment that the California legislature had made in imposing the severe sentence required under that regime for recidivist offenders, by determining that “individuals who have repeatedly engaged in serious or violent criminal behavior, and whose conduct has not been deterred by more conventional approaches to punishment, must be isolated from society in order to protect the public safety.” Ewing, 538 U.S. at 24.
24. There is a large body of case law interpreting the uncertainties inherent in § 924(c). For example, even as it relates solely to drug crimes, courts of appeals have grappled with what constitutes a “drug trafficking crime,” see Cazarez-Gutierrez v. Ashcroft, 382 F.3d 905, 910-16 (9th Cir. 2004), what counts as “in furtherance of” a crime, United States v. Timmons, 283 F.3d 1246, 1252 (11th Cir. 2002), and what conduct constitutes “possess[ing] a firearm,” United States v. Sparrow, 371 F.3d 851, 852-53 (3d Cir. 2004), as amended (Aug. 3, 2004).
25. In a post-Harmelin case in which the Court has addressed the proportionality of a life sentence, albeit one with the possibility of parole, the Court again emphasized that state legislatures have “broad discretion” over sentence length. Lockyer, 538 U.S. at 76. But, that case reached the Court on habeas corpus review, and thus the Court did not address the merits of whether the sentence would violate the Eighth Amendment if reviewed on direct appeal. Id. at 71, 76. As a result, Lockyer did not have occasion to engage directly with Justice Souter's conclusion in his dissent that the sentence under California's “three strikes” law was “on all fours” with Solem and was therefore unconstitutionally disproportionate. Id. at 82 (Souter, J., dissenting). The dissent reasoned that when sentences based on related courses of conduct are “stacked,” the incapacitation rationale for such lengthy sentences falls away, because a defendant in such a case does “not somehow become twice as dangerous to society when he [commits the second crime]; his dangerousness may justify treating one minor felony as serious and warranting long incapacitation, but a second such felony does not disclose greater danger warranting substantially longer incapacitation.” Id. Such sentences, the dissent urged, may well be grossly disproportionate. Id. And the one at issue in Lockyer, the dissent determined, was in fact disproportionate. Id. at 83.
26. As discussed above, see infra at 9-10, part of the reason Rivera was exposed to this “forever” sentence is that, due to a quirk of his case, his course of conspiratorial conduct could only be encompassed fully by charging him with many discrete conspiracy offenses rather than by charging him with having been a participant in a single overarching and extended conspiracy. See Portela, 167 F.3d at 699-700, 700 n.8. And yet, in consequence of these discrete conspiracy charges, he was exposed to the more-than-century long mandatory sentence under § 924(c) that he received. See Rodriguez, 525 F.3d 85, 111. That being the case, I find it hard to credit that Congress made a considered judgment that such a course of conspiratorial conduct merits a sentence of this extreme length, unless the defendant conspires with real drug traffickers and the prosecutor chooses to treat the entire course of conduct as a single extended conspiracy, in which case a prison sentence of five years will do just fine.Nor is my concern about how considered the congressional judgment was for conduct like that at issue in this case diminished by the fact that Rivera's § 924(c) convictions were predicated not only on his underlying conspiracy convictions but also on his underlying convictions for attempted possession with the intent to distribute a substance containing a detectable amount of cocaine. As I have explained, see infra at 9-11, it is hard to credit the notion that Congress made a considered judgment that an armed participant in an extended drug conspiracy who touches the cocaine-bearing substance in each transaction must be imprisoned for the rest of his life, while an armed member of the conspiracy who serves as a lookout as to each transaction only warrants a five-year prison sentence under § 924(c), as such a conspirator would receive if he were charged and convicted of being a participant in that single, extended conspiracy.