Barrett v. United States, 607 U.S. ___ (2026)

NOTICE: This opinion is subject to formal revision before publication in the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, [email protected], of any typographical or other formal errors.

SUPREME COURT OF THE UNITED STATES

_________________

No. 24–5774

_________________

DWAYNE BARRETT, PETITIONER v. UNITED STATES

on writ of certiorari to the united states court of appeals for the second circuit

[January 14, 2026]

Justice Jackson delivered the opinion of the Court, except as to Part IV–C.[1]*

This case concerns the relationship between two provisions of 18 U. S. C. §924. The first, §924(c)(1)(A)(i), criminalizes using, carrying, or possessing a firearm in connection with a federal crime of violence or drug trafficking crime. The second, §924(j), prescribes different penalties—including, in certain circumstances, capital punishment—when “a violation of subsection (c)” causes death.

Every defendant subject to subsection (j) has also, necessarily, violated subsection (c). We consider here whether a defendant who commits a single act that violates both §924(c)(1)(A)(i) and §924(j) may be convicted only under one provision or the other, or instead may suffer two convictions.

All agree on the first step of the analysis: Subsection (c)(1)(A)(i) and subsection (j) define the same offense under the test set forth in Blockburger v. United States, 284 U.S. 299 (1932). That consensus carries us most of the way to resolving this case, as we have long presumed that Congress intends to authorize only one conviction per offense. We resolve the rest by concluding that this presumption holds true here: Congress intended subsection (j) as an alternative, not a supplement, to subsection (c)(1)(A)(i). At the very least, Congress did not clearly manifest a contrary intention, as it would have to do if it wished to authorize two convictions in these circumstances. We thus reverse in part the Second Circuit’s contrary judgment.

I

Dwayne Barrett committed a series of robberies between August 2011 and January 2012. During one, Barrett’s confederate shot and killed Gamar Dafalla.

Of the seven charges on which Barrett was convicted, three are relevant here. Count five charged Barrett with Hobbs Act robbery of Dafalla under 18 U. S. C. §1951. That count served as the predicate for two more: Count six charged Barrett with using a firearm during the commission of a crime of violence in violation of §924(c)(1)(A)(i),[2] and count seven charged him with thereby causing death in violation of §924(j)(1).[3] A jury found Barrett guilty on all seven counts, and the District Court sentenced Barrett to a total term of 90 years of imprisonment.

Barrett has since been aided by two of our decisions. After we decided United States v. Davis, 588 U.S. 445 (2019), and vacated the Second Circuit’s disposition of Barrett’s direct appeal in light of that decision, the Second Circuit vacated one of Barrett’s §924(c) convictions not at issue here. See 588 U.S. 918 (2019) (granting Barrett’s petition for a writ of certiorari, vacating the judgment, and remanding for further consideration in light of Davis); 937 F.3d 126, 127–128 (2019) (vacating the conviction). The District Court then resentenced Barrett to 50 years in prison. Twenty of those years came from concurrent sentences on three Hobbs Act robbery counts, including, as relevant here, count five. Twenty-five years came from a consecutive term on count seven (the subsection (j) conviction), into which the District Court merged count six (the subsection (c) conviction).[4]

Then we decided Lora v. United States, 599 U.S. 453 (2023), which, like this case, addressed the relationship between subsections (c) and (j). Lora held that subsection (j) does not incorporate subsection (c)’s consecutive-sentence mandate. That mandate requires a subsection (c) sentence to run consecutively to “any other term of imprisonment imposed on the person.” §924(c)(1)(D)(ii). It does not, we held in Lora, require the same of a subsection (j) sentence. Id., at 459. When the District Court sentenced Barrett in this case, it had been under the opposite impression. So Barrett appealed again, and, bound by Lora, the Second Circuit vacated Barrett’s sentence and remanded for another resentencing. 102 F. 4th 60, 85–88 (2024).

In doing so, the Court of Appeals also passed upon the question that now comes before us. The District Court, recall, had declined to sentence Barrett separately on counts six (under subsection (c)) and seven (under subsection (j)), both of which had as their predicate the same fatal Dafalla robbery charged in count five. The Court of Appeals rejected Barrett’s argument that the Double Jeopardy Clause required the District Court to stay that course when the case returned for resentencing. The Court of Appeals did acknowledge that the Government regularly concedes that subsection (c)(1) and subsection (j) overlap and may not be punished cumulatively—a proposition with which the other circuits regularly agree. Id., at 91, and n. 29 (collecting cases). And it recognized that the two provisions qualify as the same offense under the governing test laid out in Blockburger. 102 F. 4th, at 89–90. But it believed that, “[a]s construed in Lora,” the two provisions “are separate offenses for which Congress has clearly authorized cumulative punishments.” Id., at 89.

Textually, the court pointed to §924(c)(1)(D)(ii)’s mandate that a subsection (c) sentence must run consecutively to any other sentence—including, the court reasoned, to one under subsection (j). Id., at 90–91. Practically, the court highlighted that a contrary ruling would permit defendants who commit especially serious subsection (c) offenses—ones resulting in death—to escape subsection (c)’s mandatory minimums, despite “Congress’s intent for every defendant convicted under that statute . . . to be incarcerated for no less than the stated minimum term.” Id., at 90; see also id., at 93–94. And, turning to precedent, the court perceived our statement in Lora that subsection (j) evinced a “ ‘different approach to punishment’ ” than subsection (c) to establish that the two provisions define different offenses raising no double jeopardy concerns. 102 F. 4th, at 92–93 (quoting 599 U. S., at 462). The Court of Appeals therefore instructed the District Court to impose separate convictions and sentences on counts six and seven.

Because the Second Circuit’s decision deepened a split among the Courts of Appeals, we granted certiorari. 604 U. S. ___ (2025).[5] And because the Government agrees with Barrett, we appointed an amicus curiae to defend the Second Circuit’s judgment.[6] We now reverse the Second Circuit’s judgment in relevant part.

II

A

When Congress passed 18 U. S. C. §924(c) in 1968, the new subsection “made it a discrete offense” to use or carry a firearm in connection with a federal crime of violence or drug trafficking crime, known as predicates. Abbott v. United States, 562 U.S. 8, 12 (2010).[7] The original statute left open two questions relevant to today’s case. It did not specify whether a subsection (c) conviction could coexist with, or instead must displace, a conviction for the underlying predicate. And assuming two convictions could coexist, it did not specify whether the two resulting sentences should run concurrently or consecutively.

Congress answered both questions in 1971. On the first, Congress made clear that a subsection (c) conviction must be “ ‘in addition to the punishment provided for the commission of ’ ” the predicate; that is, a violation of subsection (c) ought to result in two convictions, one for subsection (c) and one for the predicate. §13, 84Stat. 1890. On the second, Congress mandated that the two resulting sentences run consecutively, not concurrently: “ ‘[T]he term of imprisonment imposed under this subsection [shall not] run concurrently with any term of imprisonment imposed for’ ” the predicate. Ibid.

Congress later extended this second feature—the consecutive-sentence mandate—beyond the relationship between subsection (c)’s sentence and the predicate’s sentence, so that the consecutive-sentence mandate applies as between a subsection (c) sentence and “any other term of imprisonment.” §924(c)(1)(D)(ii); see United States v. Gonzales, 520 U.S. 1, 5 (1997). Congress did not, however, expand the double-conviction mandate. That mandate still instructs only that a subsection (c)(1) conviction shall be “in addition to the punishment provided for [the underlying] crime of violence or drug trafficking crime”—that is, the predicate. §924(c)(1)(A).

Subsection (c) can be violated in several ways but all trigger mandatory minimum sentences. Subsection (c)(1)(A)(i) imposes a mandatory minimum sentence of five years; other provisions within subsection (c)(1) impose mandatory minimums up to life in prison, depending on the use and type of weapon and the defendant’s recidivist history. See §924(c)(1).[8] So if the predicate offense carries a mandatory minimum sentence of its own, then a defendant will face “ ‘two consecutive mandatory minimum sentences for the single use of a single firearm.’ ” Abbott, 562 U. S., at 20.

This case asks whether a different subsection of §924, subsection (j), piles yet another layer of punishment atop—resulting in three stacked convictions and sentences—or instead provides an alternative to subsection (c)(1)(A)(i).

B

Subsection (j) arrived “decades after subsection (c).” Lora, 599 U. S., at 457. Congress added it in 1994, in a provision entitled “Death Penalty for Gun Murders During Federal Crimes of Violence and Drug Trafficking Crimes.” §60013, 108Stat. 1973. The new subsection supplied a different penalty scheme for subsection (c) violations that cause death. Unlike subsection (c), subsection (j) did not impose mandatory minimums. Instead, it authorized significant maximum sentences—including, true to the new provision’s name, the death penalty if the underlying violation is a murder:

“A person who, in the course of a violation of subsection (c), causes the death of a person through the use of a firearm, shall—

“(1) if the killing is a murder (as defined in section 1111), be punished by death or by imprisonment for any term of years or for life; and

“(2) if the killing is manslaughter (as defined in section 1112), be punished as provided in that section.” §924(j).

The references to §1111 and §1112 relate to the pre-existing murder and manslaughter statutes, respectively. Section 1111 sets a maximum penalty of death or life imprisonment for murder. 18 U. S. C. §1111(b). And §1112 sets a maximum penalty of 15 years for voluntary manslaughter and 8 years for involuntary manslaughter. §1112(b).

Subsection (j) gave federal prosecutors tools they lacked both in the pre-existing subsection (c) and in the pre-existing murder and manslaughter statutes. Subsection (c) does not authorize the death penalty, for example. Subsection (j) thus raised the ceiling of punishments §924 authorized, and, in doing so, offered prosecutors flexibility of another sort: time. See §3281 (“An indictment for any offense punishable by death may be found at any time without limitation”). Section 1111—the general federal murder statute—did already authorize the death penalty. But its jurisdictional reach, as well as that of the manslaughter statute (§1112), is limited to “the special maritime and territorial jurisdiction of the United States.” §§1111(b), 1112(b).[9] The new subsection (j) had no such geographical limitation; it borrowed instead subsection (c)’s broader jurisdictional hook. So subsection (j) expanded federal jurisdiction over more killings and authorized the death penalty for those amounting to murder.

III

This case asks whether subsection (j) also increased the number of convictions—rather than simply the maximum sentence—that can result from a fatal violation of §924(c)(1)(A)(i). Barrett says it did not, and for support he turns to the Fifth Amendment’s Double Jeopardy Clause. But while his argument arrives in constitutional garb, it “cannot be separated entirely from a resolution of the question of statutory construction.” Whalen v. United States, 445 U.S. 684, 688 (1980). That is because “the question whether punishments . . . are unconstitutionally multiple cannot be resolved without determining what punishments the Legislative Branch has authorized.” Ibid. It is that question of congressional authorization that we answer today.[10]

We undertake this exercise in statutory interpretation with a thumb on the scale, in the form of the Blockburger presumption. Named for Blockburger v. United States, 284 U.S. 299, the presumption instructs “that Congress ordinarily does not intend to punish the same offense under two different statutes.” Whalen, 445 U. S., at 691–692.

The first step in the inquiry, then, is to determine whether §924(c)(1)(A)(i) and §924(j) define the “same offense.” To make that determination, Blockburger requires us to compare the provisions in question and ask whether “each . . . requires proof of a fact which the other does not.” 284 U. S., at 304. If the answer is yes, then the offenses prescribed by each statute are different and the inquiry generally ends. If the answer is no, then the statutes define the same offense, and the Blockburger presumption is triggered.

But because we have treated Blockburger as “a rule of statutory construction to help determine legislative intent,” we have said that its presumption can yield to a “plainly expressed” intent to abandon it. Garrett v. United States, 471 U.S. 773, 778–779 (1985); see also Whalen, 445 U. S., at 692 (requiring “a clear indication of contrary legislative intent”). Accordingly, the second step of today’s analysis requires us to search for a clear manifestation of Congress’s intent to authorize more than one punishment.

All involved in this case share common ground as to the first step of the analysis. The Court of Appeals, Barrett, the Government, and Court-appointed amicus agree that §924(c)(1)(A)(i) and §924(j) define the same offense.[11] After all, the relationship between subsection (c)(1)(A)(i) and subsection (j) is “the classic relation of the ‘lesser included offense’ to the greater offense,” wherein “[t]he very same conduct” violates two statutes, one which is fully subsumed within the other. Garrett, 471 U. S., at 787. And all lesser included offenses are the “same” as their greater cousins under Blockburger. See Almendarez-Torres v. United States, 523 U.S. 224, 231 (1998) (“[T]he federal courts have [long] presumed that Congress does not intend for a defendant to be cumulatively punished for two crimes where one crime is a lesser included offense of the other”).

This case therefore turns on the analysis’s second step: discerning whether Congress clearly intended to authorize multiple convictions for one act that violates both §924(c)(1)(A)(i) and §924(j). To ascertain such intent, we turn to statutory text, structure, and (for those who accept its help) legislative history. See Garrett, 471 U. S., at 779; United States v. Woodward, 469 U.S. 105, 109 (1985) (per curiam) (consulting “[a]ll guides to legislative intent”). Only if those tools leave us certain that Congress intended to break from its normal practice and authorize multiple convictions for the same offense will we shed the presumption.

IV

A

We begin with the text. Textual clues are critical because “Congress [i]s aware of the Blockburger rule and legislate[s] with it in mind.” Albernaz v. United States, 450 U.S. 333, 342 (1981). So Congress typically includes Blockburger-surmounting language when it wishes to authorize dual convictions for the same offense. See Brief for Petitioner 28, n. 3 (listing examples).

1

The text of §924 suggests strongly, perhaps conclusively, that Congress did not disavow Blockburger here. Not for lack of know-how: Congress twice wrote Blockburger-surmounting language into subsection (c) itself. Congress mandated that a §924(c)(1) conviction must be “in addition to the punishment provided for” the predicate, and it also mandated that a §924(c)(5) conviction—for using or carrying “armor piercing ammunition”—must be “in addition to the punishment provided for” the predicate “or conviction under” §924. See §§924(c)(1)(A), (c)(5). We have elsewhere called such “in addition to” language “crystal clear” evidence of a legislature’s intent to overcome Blockburger. Missouri v. Hunter, 459 U.S. 359, 362, 368 (1983).[12]

In short, “[w]hen Congress has the will” to authorize dual convictions for the same offense, Congress “has no difficulty in expressing it.” Bell v. United States, 349 U.S. 81, 83 (1955). But Congress used no similar language with respect to the interplay between subsection (c)(1) and subsection (j). Its silence on the topic speaks volumes. See Albernaz, 450 U. S., at 341–342.

2

Like the Court of Appeals, see 102 F. 4th, at 90–93, amicus challenges the premise of this assessment: He insists that Congress has textually authorized dual convictions under subsection (c)(1) and subsection (j). For proof, he points to subsection (c)’s consecutive-sentence mandate. It reads:

“Notwithstanding any other provision of law . . . no term of imprisonment imposed on a person under this subsection shall run concurrently with any other term of imprisonment imposed on the person, including any term of imprisonment imposed for the crime of violence or drug trafficking crime during which the firearm was used, carried, or possessed.” §924(c)(1)(D)(ii).

As amicus would have it, this instruction permits—actually, requires—a subsection (c)(1) conviction to add on to any other conviction, including one under subsection (j).

This deployment of the consecutive-sentence mandate misunderstands today’s inquiry. The argument attempts to make Blockburger purely about sentences—so that an instruction to run sentences consecutively rather than concurrently makes clear Congress’s intent to overcome the presumption. But Blockburger addresses the permissibility of multiple convictions, not just multiple sentences. “The assumption underlying the Blockburger rule is that Congress ordinarily does not intend to punish the same offense under two different statutes,” where “punishment” means “a criminal conviction and not simply the imposition of sentence.” Ball v. United States, 470 U.S. 856, 861 (1985). It is for this reason that running sentences concurrently does not avoid the problem; an unauthorized “ ‘second conviction, even if it results in no greater sentence, is an impermissible punishment.’ ” Rutledge v. United States, 517 U.S. 292, 302 (1996) (quoting Ball, 470 U. S., at 865).

Accordingly, §924(c)’s consecutive-sentence mandate simply speaks past the question in this case: whether one act may result in two convictions. To that question it is no answer to say, as the consecutive-sentence mandate does, that a “term of imprisonment imposed” for a conviction shall run consecutively to “any other term of imprisonment imposed.” §924(c)(1)(D)(ii). Before the consecutive-sentence mandate gains any relevance, a court must first determine whether two punishments (convictions) may be imposed at all. Only if two convictions may coexist does a court consult the consecutive-sentence mandate, to arrange properly the resulting sentences.

This description is not merely the law as it exists; it is the law as Congress understands it. When Congress amended subsection (c) in 1971, it added two distinct instructions, as recited above. The first—that a subsection (c)(1) conviction must be “in addition to the punishment provided for” the predicate—answers the threshold question: May one act spawn two convictions, one for subsection (c)(1) and one for the predicate? The second—the consecutive-sentence mandate—answers the follow-on question: Presuming the existence of two convictions, should their resulting sentences run concurrently or consecutively? Cf. Dean v. United States, 581 U.S. 62, 69–70 (2017) (describing these provisions as “two” distinct “limitations”). Amicus would have the second instruction answer the first question. But Congress has already addressed it: A subsection (c)(1) conviction adds to the predicate conviction. See §924(c)(1)(A). Nowhere does the statute prescribe the same in relation to a subsection (j) conviction.[13]

3

Amicus’s second textual argument is equally unavailing. Turning away from the concrete and toward the abstract, amicus tells us that subsection (c)(1) and subsection (j) have different focuses. Subsection (c)(1), he points out, calibrates its punishments to “the use of the firearm, the type of firearm, and recidivism.” Brief for Court-Appointed Amicus Curiae 10. By contrast, subsection (j) focuses on something “irrelevant” to subsection (c)(1): the harm (namely, death) inflicted. Ibid. From this amicus urges us to conclude that the two subsections “target different wrongs” and thus may be punished cumulatively. Ibid.; see also id., at 17–18.

In our search for congressional intent, we might in some circumstances think that when two provisions focus on entirely “different interests,” that suggests Congress intended them to accumulate rather than swap out. See United States v. Dixon, 509 U.S. 688, 724 (1993) (White, J., concurring in judgment in part and dissenting in part); see also Albernaz, 450 U. S., at 343 (observing that the Court’s conclusion about the permissibility of dual convictions was “reinforced by the fact that the two . . . statutes are directed to separate evils”). Whatever those circumstances, this is not one. Amicus stretches to find a common thread running throughout subsection (c)(1)—he bundles firearm use, firearm type, and recidivism and places them in opposition to harm. But it is not obvious that those three features form a theme of any significance such that they can create the contrast amicus wants to glean from them.

In any event, the conduct (or result) that differentiates a greater offense from its lesser included offenses will often introduce some new “focus.” That reality cannot do much to overcome the Blockburger presumption if the presumption is to retain its force. See, e.g., Whalen, 445 U. S., at 691, n. 6, 693–694 (Blockburger not overcome, despite some contrary statutory language, for convictions on charges of (1) rape and (2) killing in the course of rape); Illinois v. Vitale, 447 U.S. 410, 420–421 (1980) (Blockburger not overcome as between felony and felony murder); Harris v. Oklahoma, 433 U.S. 682, 682–683 (1977) (per curiam) (same).

In sum, if Congress expressed any will to overcome Blockburger in this scenario, it must be found somewhere other than the statutory text.

B

Failing textual support, amicus turns to the statute’s operation and structure. But this ground proves only slightly more fertile. Primarily, amicus fears that defendants convicted of and sentenced under subsection (j) will be rewarded with more lenient sentences than those convicted of the less serious subsection (c)(1) offense. He points out that subsection (c)(1) imposes mandatory minimums while subsection (j) speaks in terms of maximums. And he insists that, rather than let subsection (j) offenders out from under subsection (c)(1)’s mandatory minimums, Congress must have intended to authorize subsection (c)(1)’s mandatory minimums plus any punishment doled out under subsection (j). Cf. Abbott, 562 U. S., at 21 (rejecting construction of §924(c) under which “the worst offenders would often secure the shortest sentences”).

Lora answers this concern without breaking a sweat. There, we explained that subsection (j)—along with several other provisions enacted simultaneously—“eschews mandatory penalties in favor of sentencing flexibility.” 599 U. S., at 462. It was not lost on us in Lora, nor is it now, that subsection (j) defines an especially serious offense. Subsection (j) simply “reflects th[at] seriousness . . . using a different approach than subsection (c)’s mandatory penalties”: by authorizing the death penalty for murder and “the same harsh punishment that the Federal Criminal Code prescribes for other manslaughters.” Id., at 463. It is for this reason, among others, that Lora—anticipating the question this case presents—explained that our conclusion in that case “aligns with” the mutually exclusive relationship between subsection (c)(1)(A)(i) and subsection (j) that we endorse today. Id., at 461.

Just as subsection (j)’s seriousness is not lost on us, we are confident—as Congress apparently was—that it will not be lost on sentencing judges. Bound by 18 U. S. C. §3553(a) to craft a sentence that, among other considerations, reflects “the seriousness of the offense” and avoids “unwarranted sentence disparities” among similarly situated defendants, sentencing judges will doubtless recognize the relevance of a victim’s death when sentencing under subsection (j). Amicus’s worry thus strikes us as more theoretical than realistic. But if it threatens to manifest, it is not without remedy: If, in a given case, prosecutors fear that a subsection (j) sentence will dip below what subsection (c)(1)(A)(i) would otherwise guarantee, they are free to choose subsection (c)(1)(A)(i)’s low-end rigidity over subsection (j)’s high-end flexibility. See Ball, 470 U. S., at 860–861, nn. 7–8.[14]

There is a textual answer to this concern, too, and here (again) §924(c)(5) plays the foil. That provision, which pertains to the use of armor-piercing ammunition, states that, “in addition to the punishment provided for” the predicate “or conviction under this section,” a defendant shall face a mandatory minimum sentence of 15 years “and,” if death results, shall be sentenced in accordance with a scheme that mirrors subsection (j)’s. (Emphasis added.) This illustrates that, when Congress wished to preserve a mandatory minimum it feared might otherwise disappear, it so specified. But subsection (j), we explained in Lora, is “cast from a different mold.” 599 U. S., at 461.

Next, amicus highlights the physical separation in the U. S. Code between subsections (c)(1) and (j). And he underscores their independent operation, emphasizing that subsection (j)’s sentencing scheme operates without reference to subsection (c)’s.

We have not bought such arguments before. Consider, for instance, the statutes at issue in Ball—statutes whose elements overlapped but whose sentencing schemes diverged. Those provisions, which were then located at 18 U. S. C. §§922(h) and 1202(a), prohibited certain categories of people from, respectively, receiving and possessing certain firearms. The provisions occupied parts of the Code much farther apart than the subsections of §924 at issue here. And in United States v. Batchelder, 442 U.S. 114 (1979), we explained that each statute, “in conjunction with its own sentencing provision, operate[d] independently of the other,” such that the statutes were “each fully enforceable on [their] own terms.” Id., at 118–121. We therefore concluded that the provisions did not incorporate each other’s penalties. Id., at 119. Yet, six years later (in Ball), we concluded that the two independent provisions were also mutually exclusive under Blockburger. See Ball, 470 U. S., at 861–864.

What Batchelder is to Ball, Lora is to this case. Subsection (j) shares subsection (c)’s elements but not its sentencing scheme. See Lora, 599 U. S., at 458–459. “Instead, subsection (j) supplies its own comprehensive set of penalties that apply instead of subsection (c)’s.” Id., at 460. And as in Ball, this dynamic magnifies (rather than eliminates) the likelihood that the Blockburger presumption holds true. After all, if offenses that share elements—as they must to satisfy Blockburger—have penalties that operate on their own rather than by reference to each other, see Batchelder, 442 U. S., at 118–121; Lora, 599 U. S., at 458–459, that suggests Congress intended to place in front of prosecutors a menu, not a buffet. Accord, Jeffers v. United States, 432 U.S. 137, 156–157 (1977) (plurality opinion) (one provision’s “comprehensive penalty structure” indicated that it was not intended to supplement a lesser included provision).

Finally, amicus analogizes this case to Garrett, in which we found Blockburger overcome even without express statutory language disclaiming the presumption. Garrett examined the relationship between the continuing criminal enterprise (CCE) statute and its constituent offenses. See 471 U. S., at 779–781.

The distinctions from Garrett are legion, because the justifications we offered for that decision are legion. But, most straightforwardly, Garrett distinguished “the classic relation of the ‘lesser included offense’ to the greater offense,” wherein “[t]he very same conduct” violates two statutes. Id., at 787 (contrasting Brown v. Ohio, 432 U.S. 161 (1977), which involved joyriding and auto theft—both of which were violated by driving a stolen car). In Garrett, by contrast, “the continuing criminal enterprise was alleged to have spanned more than five years.” 471 U. S., at 788. The Court did not believe that Congress wished to require the Government to choose between prosecuting the defendant for early predicates and forfeiting a later CCE charge, on the one hand, and allowing him to go on breaking the law to preserve the possibility of a CCE charge, on the other. Id., at 788–790. The continuing nature of the offense played a decisive role in the case: “One who insists that the music stop and the piper be paid at a particular point,” we reasoned, “must at least have stopped dancing himself before he may seek such an accounting.” Id., at 790.[15]

Today’s circumstances are markedly different. Here, we have nothing more than “the classic relation of the ‘lesser included offense’ to the greater offense.” Id., at 787.

C

Amicus turns finally to legislative history. We have said that legislative history may “fortif[y]” a conclusion that Congress intends to overcome Blockburger. See Garrett, 471 U. S., at 782. But the legislative history amicus offers pertains to double convicting for subsection (c) and its predicates—a practice Congress expressly authorized in the text. See Brief for Court-Appointed Amicus Curiae 26–28. For obvious reasons, this is unhelpful data for interpreting the relationship between subsections (c) and (j). If anything, it illuminates the sort of discussion that characterizes congressional deliberation over legislation that does intend to authorize cumulative punishment—and thus highlights the notable absence of any similar deliberation over subsection (j). See Brief for National Association of Criminal Defense Lawyers as Amicus Curiae 5–10.

Rather, the only legislative history that is relevant to today’s inquiry accords with Barrett’s understanding of subsection (j). That legislative history suggests that, rather than stacking punishment or creating a brand “[n]ew . . . offens[e],” subsection (j) (at the time styled subsection (i)) “made” a “capital offens[e]” out of a “[p]re-existing crim[e]”—the one at subsection (c). Cong. Research Serv., C. Doyle, Crime Control Act of 1994: Capital Punishment Provisions Summarized, pp. 3–4 (94–721 S, 1994); see also 140 Cong. Rec. 11158 (1994) (Sen. Biden remarking that the new provision “says if you are guilty of committing a crime that results in the death of an individual through the use of a gun, you are eligible for the death penalty, assuming it is a Federal crime”).

Like its silence in the statutory text, Congress’s debate-floor silence says a lot. The Blockburger presumption is “long-settled,” Gamble v. United States, 587 U.S. 678, 710 (2019), muscular, and fundamental to our law. We think it highly unlikely that the “ ‘lawyer’s body’ ” that is Congress, Albernaz, 450 U. S., at 341 (quoting Callanan v. United States, 364 U.S. 587, 594 (1961)), discarded Blockburger without comment here.

*  *  *

Congress has not authorized convictions under both 18 U. S. C. §§924(c)(1)(A)(i) and (j) for one act that violates both provisions. The part of the judgment of the Court of Appeals that held otherwise is reversed, and the case is remanded for further proceedings consistent with this opinion.

It is so ordered.

SUPREME COURT OF THE UNITED STATES

_________________

No. 24–5774

_________________

DWAYNE BARRETT, PETITIONER v. UNITED STATES

on writ of certiorari to the united states court of appeals for the second circuit

[January 14, 2026]

Justice Gorsuch, concurring in part.

On the morning of December 12, 2011, Dwayne Barrett participated in an armed robbery that left a man dead. For this, Mr. Barrett was convicted twice in federal court. Once for using a firearm during a crime of violence under 18 U. S. C. §924(c)(1)(A). And once for causing a death during a §924(c) violation in violation of §924(j). 102 F. 4th 60, 67 (CA2 2024). For his two convictions, the Second Circuit ordered two separate punishments: a prison term for his §924(c)(1)(A) violation, plus another, consecutive prison term for his §924(j) violation. All of which raises the question: If Mr. Barrett is convicted and punished for both crimes, has he been “twice put in jeopardy” for “the same offence” in violation of the Double Jeopardy Clause? U. S. Const., Amdt. 5.

There is no doubt how we would answer that question if the government had prosecuted Mr. Barrett under §924(c)(1)(A) and §924(j) in two successive proceedings. The analysis would begin and end with Blockburger v. United States, 284 U.S. 299 (1932). Under Blockburger, two provisions create the same offense for double jeopardy purposes unless each provision has an element the other lacks. Id., at 304. Here, all agree that §924(c)(1)(A) has no element that §924(j) lacks. So the two provisions create the same offense, and the Double Jeopardy Clause would bar the second prosecution. United States v. Dixon, 509 U.S. 688, 696 (1993).

What happens, though, where (as here) the government brings two charges for the same offense concurrently in the same proceeding rather than successively in two separate ones? That question does not appear to have arisen much in our Nation’s early years. Maybe it didn’t in part because criminal codes were considerably thinner then, affording prosecutors fewer opportunities to bring overlapping charges. Maybe it didn’t in part, too, because of the traditional maxim that “an indictment should not include more than one felony,” which left multiple-count indictments vulnerable to being quashed by trial judges. Pointer v. United States, 151 U.S. 396, 403 (1894). But if the question didn’t arise much in the past, it is of obvious relevance in our times. These days, federal and state criminal codes have exploded, with scores of repetitive offenses on the books. Frequently, also, today’s prosecutors bring as many overlapping felony charges as they can in a single case to see what will stick, and courts often tolerate the practice.

The litigants before us proceed on the unexamined premise that Blockburger works differently in concurrent prosecutions than it does in successive ones. In concurrent prosecutions, they assume, Blockburger operates as a mere “presumption” for ascertaining congressional “inten[t].” E.g., Brief for Petitioner 2 (internal quotation marks omitted). Because §924(c)(1)(A) and §924(j) create the same offense under Blockburger, they say, we should presume that just one of Mr. Barrett’s convictions can stand unless Congress has clearly directed otherwise. The only question that divides the litigants is whether, in fact, Congress included clear directions in §924 authorizing both convictions. See Brief for Petitioner 18–26; Brief for United States 15–23; Brief for Court-Appointed Amicus Curiae 15–28.

Addressing the case as the litigants have framed it, the Court today holds that nothing in §924 overcomes the “Blockburger presumption,” so only one of Mr. Barrett’s convictions can stand. Ante, at 1–2, 9. I agree with that as far as it goes. But it’s also hard not to wonder where the litigants’ presumptive version of Blockburger comes from. If the Constitution always prohibits the government from securing two convictions for the same offense in successive prosecutions, why would it sometimes tolerate a different result in concurrent prosecutions?

One thing here is certain. The litigants framed this case the way they did because our cases speak confusingly about the role of the Double Jeopardy Clause in concurrent prosecutions. This Court has sometimes said that the Clause “ ‘protects against multiple punishments for the same offense.’ ” Schiro v. Farley, 510 U.S. 222, 229 (1994) (quoting North Carolina v. Pearce, 395 U.S. 711, 717 (1969)). And we have held that multiple convictions for the same offense—even when secured in a single proceeding—count as multiple punishments. Ball v. United States, 470 U.S. 856, 865 (1985). From this, it would seem to follow that Congress cannot authorize multiple convictions for the same offense in concurrent prosecutions. But this Court has also sometimes said that, in the concurrent-prosecution context, the Clause merely directs courts to ascertain statutory meaning accurately. “ ‘The question of what punishments are constitutionally permissible,’ ” we once wrote, “ ‘is no different from the question of what punishments the Legislative Branch intended to be imposed.’ ” Missouri v. Hunter, 459 U.S. 359, 368 (1983) (quoting Albernaz v. United States, 450 U.S. 333, 344 (1981); emphasis added; alteration omitted); see also Garrett v. United States, 471 U.S. 773, 778 (1985).

Someday, we will need to resolve the tension in our case law. And when we do, I see two likely solutions. One would be to say out loud what some of our cases imply: that the constitutional phrase “the same offence,” Amdt. 5, means different things in different contexts. In the successive-prosecution context, two charges amount to the same offense if they fail the Blockburger test. But in the concurrent-prosecution context, two charges amount to the same offense only if they fail the Blockburger test and Congress has not clearly intended punishment under both. That solution, though, would be a curious one indeed. Not only would it allow Congress to permit in the concurrent-prosecution context what we have long held the Constitution forbids in the successive-prosecution context. Really, it is a little “embarrassing to assert that the single term ‘same offence’ ” in the Double Jeopardy Clause “has two different meanings.” Dixon, 509 U. S., at 704.

A second solution would be to recognize, as others of our cases imply, that two charges amount to the same offense under the Double Jeopardy Clause if they fail the Blockburger test—full stop. So a defendant cannot be prosecuted for both charges in successive proceedings. Nor may he be convicted twice (much less sentenced twice) for both charges just because they happen to be brought concurrently. True, this approach would require us to admit that Blockburger is not, after all, a mere presumption in the concurrent-prosecution context. But saying that much would give the constitutional phrase “the same offence” a consistent meaning and treat like cases alike.

Nor would taking that step represent some giant leap. Our cases discussing Blockburger as a mere presumption in the concurrent-prosecution context were decided during a relatively brief period when this Court didn’t take Blockburger very seriously in any context, even when it came to successive prosecutions. See Grady v. Corbin, 495 U.S. 508, 510 (1990) (finding a double jeopardy violation based on a same-transaction test, divorced from Blockburger). Since then, however, we have expressly renounced that approach, reaffirming that Blockburger’s “long-settled rule” controls what counts as “an ‘offence’ for double jeopardy purposes” in successive prosecutions. Gamble v. United States, 587 U.S. 678, 710 (2019); see also Dixon, 509 U. S., at 696, 704 (overruling Grady). Tellingly, too, in all the years since Dixon, we have not found a single case in which the “Blockburger presumption” against concurrent prosecutions for the same offense was, in fact, overcome by a clear congressional command. Cf. Rutledge v. United States, 517 U.S. 292, 303 (1996) (rejecting the government’s argument that “the presumption against allowing multiple punishments for the same crime” was “overcome”). In short, the analytical foundations of the presumptive version of Blockburger have been gravely undermined by subsequent developments.

Perhaps some might worry about the consequences of this second solution. Today, after all, prosecutors routinely bring concurrent charges for greater offenses and their lesser included variants, giving juries several degrees of culpability to choose from. One might wonder whether affording the Double Jeopardy Clause the same meaning in the concurrent-prosecution context that it already enjoys in the successive-prosecution context would render this practice illegal. It would not. A jury could reach a guilty verdict on two charges that constitute the same offense, but no double jeopardy problem would arise so long as the court does not enter judgments of conviction on both. It may not be clear at a trial’s outset which charge (if any) will yield a conviction. But it will be certain all along that, when judgment day comes, the defendant will be convicted and punished only once for any given offense. See Ball, 470 U. S., at 865.

Today, to be sure, the Court has no occasion to tangle with any of this. The parties have not asked us to address the tension in our case law. Nor does anything here turn on its resolution, given the Court’s holding that one of Mr. Barrett’s convictions must go even under the merely presumptive version of Blockburger. All this, the Court rightly takes care to acknowledge. See ante, at 8, n. 10 (“Because we conclude that Congress did not authorize two convictions in this context, we need not revisit whether Congress could do so consistent with the Double Jeopardy Clause”).

On that understanding, I join all but Part IV–C of Justice Jackson’s opinion. But while today’s decision is correct as far as it goes, sooner or later we will have to clear up the confusion—and to my eyes, this case serves as a poster child for how that confusion should be resolved. Mr. Barrett really was charged twice for one offense. He really was convicted twice. Before our intervention, he really was set to be criminally punished twice. And whatever Congress might or might not intend, that is double jeopardy.