BROWN

Appeal from the United States District Court for the District of Columbia (No. 04cr00355–05).
Robert S. Becker, appointed by the court, argued the cause and filed the briefs for appellant. A.J. Kramer, Federal Public Defender, and Rosanna M. Taormina, Assistant Federal Public Defender, were on the brief of the Federal Public Defender as amicus curiae in support of appellant.
Paul F. Enzinna and Barry J. Pollack were on the brief of amicus curiae NationalAssociation of Criminal Defense Lawyers, Inc. in support of appellant.

Stratton C. Strand, Assistant U.S. Attorney, argued the cause for appellee. With him on the briefs were Ronald C. Machen Jr., U.S. Attorney, and Roy W. McLeese III, Daniel P. Butler, and Stephanie C. Brenowitz, Assistant U.S. Attorneys.

Before: SENTELLE, Chief Judge, and HENDERSON, ROGERS, TATEL, GARLAND, BROWN, GRIFFITH and KAVANAUGH, Circuit Judges.

Opinion for the Court filed by Circuit Judge BROWN.
Concurring opinion filed by Chief Judge SENTELLE.
Concurring opinion filed by Circuit Judge HENDERSON.
Dissenting opinion filed by Circuit Judge ROGERS.
Dissenting opinion filed by Circuit Judge KAVANAUGH, with whom Circuit Judge TATEL joins.

BROWN, Circuit Judge:

Appellant Bryan Burwell was part of a crew of bank robbers that engaged in a violent crime spree across the D.C. Metro area. The crew employed decidedly old-school tactics, including subduing bystanders by brandishing AK–47s, pistol whipping a victim, and spraying a pursuing police car with bullets. After a lengthy jury trial and an appeal before a panel of this Court, only a single legal question remains: whether 18 U.S.C. § 924(c)(1)(B)(ii), which imposes a mandatory thirty-year sentence for any person who carries a machinegun while committing a crime of violence, requires the government to prove that the defendant knew the weapon he was carrying was capable of firing automatically. To resolve this question, we reexamine one of our longstanding precedents, United States v. Harris, 959 F.2d 246, 257–59 (D.C.Cir.1992), in light of intervening decisions of the Supreme Court.

I

This appeal, which focuses on a narrow question of law, requires only an abbreviated version of the essential underlying facts. Between 2003 and 2004, a gang of robbers committed six armed bank heists; Burwell, who joined the crew in the middle of the crime spree, participated in two. Before Burwell joined up, Noureddine Chtaini, the nominal leader, along with Miguel Morrow and Omar Holmes, bought four fully automatic AK–47s.

The crew carried these weapons in all their subsequent bank robberies, and on one occasion, “sprayed” bullets at a pursuing police car. Following this particular robbery, Burwell said he wanted to start robbing banks with the crew. He carried an AK–47 in both of the robberies in which he participated, though there is no evidence he fired any of the weapons.

A grand jury issued an indictment charging Burwell and his co-defendants with, inter alia, RICO conspiracy and armed bank robbery conspiracy. They also charged Burwell with one count of armed bank robbery and one count of using or carrying a firearm during a violent crime. A jury returned verdicts as to all defendants on July 15, 2005, convicting each of RICO conspiracy and conspiracy to commit armed bank robbery. In addition, it convicted Burwell of armed robbery and of using or carrying a machinegun in relation to a violent crime. The district court sentenced Burwell to concurrent prison terms of 135 months each for RICO conspiracy and armed bank robbery, 60 months for conspiracy to commit armed bank robbery, and a consecutive term of 360 months for using or carrying a machinegun during the robbery.

On appeal, Burwell argued the government presented insufficient evidence to support his conviction under 18 U.S.C. § 924(c)(1)(B)(ii) because the government failed to show he knew the AK–47 he carried was capable of firing automatically. All four weapons could function in both semiautomatic and fully automatic modes through the use of a selector switch—a lever on the side of the weapon that slides up and down to allow the user to choose between safe, semi-automatic, and fully automatic modes. But at trial, both parties' firearms experts agreed that the weapons contained no clear markings indicating that they could be put into automatic firing mode. The defense's expert, William Welch, testified that “[t]here's some letters here [near the selector switch] that I cannot identify because they're probably written in a foreign language, but they're only letters, not words.” Welch determined that the AK–47 could fire in automatic mode by noticing that the selector switch had three positions, which he “was kind of looking for ... anyway,” and by disassembling the gun.

Section 924(c)(1)(A) provides for a mandatory consecutive sentence of at least five years for any person who uses or carries a firearm “during and in relation to” a crime of violence or drug trafficking crime, or for any person who possesses a firearm “in furtherance of” such crime. The mandatory sentence skyrockets to thirty years, however, if the firearm involved was a machinegun. 18 U.S.C. § 924(c)(1)(B)(ii). A machinegun is defined as “a gun capable of firing automatically, that is, of firing several bullets with one pull on the trigger.” Harris, 959 F.2d at 257. The court upheld the jury verdict and sentence, finding that this Court's decision in Harris dictated affirmance.

In Harris, the Court concluded Congress “inten[ded] to apply strict liability” to the machinegun provision of § 924(c). 959 F.2d at 258. The Court began its analysis by recognizing the general presumption in favor of a mens rea requirement in criminal cases, but reasoned that § 924(c)(1) already requires a defendant to have “intentionally” used a firearm in committing a predicate crime, and to have done so with “knowledge that the objects used to facilitate the crime are ‘firearms.’ ” Id. at 258–59. Thus, the statute does require “[d]eliberate culpable conduct” as to “the essential elements of the crime ... before the issue of sentence enhancement for use of a machine gun arises.” Id. at 259. The structure of the statute, the Court found, supported the inference that Congress intended no additional mens rea requirement to apply to the machinegun element.

Burwell, aided by amici curiae National Association of Criminal Defense Lawyers (“NACDL”) and the Federal Public Defender (“FPD”), sought rehearing en banc, claiming the Supreme Court's decision in United States v. O'Brien, –––U.S. ––––, 130 S.Ct. 2169, 176 L.Ed.2d 979 (2010)—which held that § 924(c)(1)(B)(ii) is an element of the offense, rather than a sentencing factor—implicitly overruled Harris. Burwell also claims contrary decisions of other circuits support abandonment of Harris. Harris's interpretation of § 924(c) is, he contends, fundamentally flawed as a matter of law. We disagree on all counts, and conclude the high burden imposed on any party who urges this Court to depart from the principle of stare decisis has not been satisfied.

II

By claiming § 924(c)(1)(B)(ii) contains an additional, implicit mens rea requirement, Burwell asks us to set aside a circuit precedent that has governed our interpretation for twenty years.

“[T]he doctrine of stare decisis is of fundamental importance to the rule of law.” Welch v. Texas Dep't of Highways & Pub. Transp., 483 U.S. 468, 494, 107 S.Ct. 2941, 97 L.Ed.2d 389 (1987). “[A]ny departure from the doctrine ... demands special justification.” Arizona v. Rumsey, 467 U.S. 203, 212, 104 S.Ct. 2305, 81 L.Ed.2d 164 (1984). The burden borne by a party urging the disavowal of an established precedent is greater “where the Court is asked to overrule a point of statutory construction ... for here, unlike in the context of constitutional interpretation, ... Congress remains free to alter what we have done.” Patterson v. McLean Credit Union, 491 U.S. 164, 172–73, 109 S.Ct. 2363, 105 L.Ed.2d 132 (1989).

Overturning a statutory precedent is justified under a very narrow range of circumstances, such as cases in which an “intervening development of the law, through either the growth of judicial doctrine or further action taken by Congress,” necessitates a shift in the Court's position. Id. at 173, 109 S.Ct. 2363. Precedents may also be abandoned where an intervening development “ha[s] removed or weakened the conceptual underpinnings from the prior decision, or where the later law has rendered the decision irreconcilable with competing legal doctrines or policies.” Id. The Court has also overruled prior cases where the precedent “may be a positive detriment to coherence and consistency in the law, either because of inherent confusion created by an unworkable decision, or because the decision poses a direct obstacle to the realization of important objectives embodied in other laws.” Id.

A court of appeals sitting en banc may also reexamine its own interpretation of a statute “if it finds that other circuits have persuasively argued a contrary construction.” Critical Mass Energy Project v. NRC, 975 F.2d 871, 876 (D.C.Cir.1992) ( en banc ). Or an en banc court may set aside its own precedent “if, on reexamination of an earlier decision, it decides that the panel's holding on an important question of law was fundamentally flawed.” Id.

a. Effect of Subsequent Supreme Court Decisions

Burwell and NACDL rely on the Supreme Court's recent decision in United States v. O'Brien, which held that possession of a machinegun “[was] an element to be proved to the jury beyond a reasonable doubt,” not a “sentencing factor” to be proved to the judge by a preponderance of the evidence at sentencing. 130 S.Ct. at 2172. According to NACDL, the Harris Court justified departing from the traditional presumption in favor of mens rea by finding the machinegun provision is not an “essential element of the crime” but rather a “sentence enhancement.” See Harris, 959 F.2d at 258–59. If the machinegun provision is an essential element of the crime, they argue, it must also be afforded the presumption of a mens rea requirement.

It is far from clear, however, that the classification of § 924(c)(1)(B)(ii) as a “sentencing factor” was one of the “conceptual underpinnings” of Harris. Much like the Supreme Court's recent decision in Flores–Figueroa, 129 S.Ct. 1886 (2009), Harris's holding turned on the Court's interpretation of § 924(c) “using the usual tools of statutory analysis” in an attempt to determine congressional intent. 959 F.2d at 258. Specifically, the Court focused on the structure of the statute, i.e. the separation between the elements of the underlying crime and the machinegun provision, and its judgment that Congress would not have required an additional showing of mens rea beyond what the government must prove with respect to the predicate crime and the use of a firearm. See id. at 258. The Court also found other circuits' conclusions that no showing of additional mens rea was required for “aggravating elements” of similar statutes persuasive. See id. Nothing turned on whether the machinegun provision was considered an element of the offense or a sentencing factor. Moreover, the Harris Court referred to the machinegun provision as both an “element of the offense,” id. at 258, and a “sentence enhancement,” id. at 259, making it clear that its holding did not depend on which description more accurately characterized the machinegun provision. It is thus unclear how much, if at all, the rationale of Harris relies upon the Court's apparent assumption that § 924(c)(1)(B)(ii) was a “sentence enhancement” rather than an element of the offense.

But Burwell makes an even broader claim, arguing that O'Brien's holding implicitly overruled Harris because offense elements require proof of mens rea while sentencing factors may not. While Burwell can marshal some support for this argument from expansive dicta in decisions from the Supreme Court and this Circuit, close reading of these cases reveals that his argument is overstated.

First, Burwell's suggestion that the label “element of the offense,” as opposed to “sentencing factor,” is determinative of the mens rea requirement is misguided. Had the Supreme Court viewed that distinction as dispositive, it would not have explicitly declined to decide whether “a defendant who uses, carries, or possesses a firearm must be aware of the weapon's characteristics” in O'Brien. 130 S.Ct. at 2173. At the very least, the Court's reservation suggests there is more to the analysis than a simple equation.

Second, before mechanically applying a presumption—particularly a presumption as sweeping as the one put forward here—it seems prudent to revisit first principles. The Supreme Court developed the presumption in favor of mens rea for one particular reason: to avoid criminalizing otherwise lawful conduct. One of the earliest cases to adopt a presumption in favor of mens rea was United States v. U.S. Gypsum Co., 438 U.S. 422, 98 S.Ct. 2864, 57 L.Ed.2d 854 (1978), in which the Court evaluated whether intent was an element of a criminal antitrust offense. Relying primarily on Morissette v. United States, 342 U.S. 246, 250, 72 S.Ct. 240, 96 L.Ed. 288 (1952), the Court noted that its precedents “can be fairly read as establishing, at least with regard to crimes having their origin in the common law, an interpretative presumption that mens rea is required.” 438 U.S. at 437, 98 S.Ct. 2864. The Court further noted that strict liability offenses, while “not unknown to the criminal law” and not invariably violative of the Constitution, are “generally disfavored.” Id. at 437–38, 98 S.Ct. 2864. Up to this point, the case appears to support Burwell's position.

Further analysis, however, reveals that the Court inferred a mens rea requirement because “the behavior proscribed by the [Sherman] Act is often difficult to distinguishfrom the gray zone of socially acceptable and economically justifiable business conduct.” Id. at 440–41, 98 S.Ct. 2864. Not requiring the government to prove criminal intent in such a situation would risk using criminal sanctions simply to “ regulate business practices,” a result not intended by Congress. Id. at 442, 98 S.Ct. 2864. In other words, the Court inferred a mens rea requirement because criminal intent was necessary to differentiate a violation of the Act from otherwise lawful business conduct.

The Court applied the same principle in Staples v. United States, when it implied a mens rea requirement in 26 U.S.C. § 5861(d), a provision that made it “unlawful for any person ... to receive or possess a firearm which is not [federally] registered.” 511 U.S. 600, 605, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994). The Court held the statute did require the Government to prove mens rea because a contrary ruling would “criminalize a broad range of apparently innocent conduct.” Id. at 610, 114 S.Ct. 1793. In closing, the Court favorably cited a different section of our Harris opinion, in which we held that weapon-specific knowledge is required in a prosecution under § 5861(d), reasoning that “if Congress had intended to make outlaws of gun owners who were wholly ignorant of the offending characteristics of their weapons, and to subject them to lengthy prison terms, it would have spoken more clearly to that effect.” Id. at 620, 114 S.Ct. 1793;see Harris, 959 F.2d at 261 (“We believe that if Congress, against the background of widespread lawful gun ownership, wished to criminalize the mere unregistered possession of certain types of firearms[,] ... it would have spoken clearly to that effect.”).

Similar concerns prompted application of the presumption in United States v. X–Citement Video, Inc., 513 U.S. 64, 115 S.Ct. 464, 130 L.Ed.2d 372 (1994), in which the Court held that 18 U.S.C. § 2252 requires knowledge that a performer in a sexually explicit video actually is a minor. As in Staples, the Court feared that not requiring proof of mens rea “would sweep within the ambit of the statute actors who had no idea that they were even dealing with sexually explicit material,” such as a retail druggist who returns an uninspected roll of developed film to a customer. Id. at 69, 115 S.Ct. 464. Indeed, the court clarified the scope of the presumption by explaining that “Morissette, reinforced by Staples, instructs that the presumption in favor of a scienter requirement should apply to each of the statutory elements that criminalize otherwise innocent conduct.” Id. at 72, 115 S.Ct. 464 (emphasis added).

Our own recent precedent follows the same logic. In United States v. Project on Government Oversight (“POGO”), 616 F.3d 544 (D.C.Cir.2010), this Court adopted the presumption that “criminal statutes and regulations contain a mens rea element unless otherwise clearly intimated in the language or legislative history.” Id. at 549. Like Staples and X–Citement Video, however, POGO also involved a statute that criminalized otherwise lawful behavior—in this case, contributions to public officials. The Court based its application of the presumption on a similar concern about criminalizing otherwise innocent conduct. Absent an intent requirement, “a parent's monthly checks to a child who works for the government could be construed as violating § 209(a): only the parent's intent distinguishes payments to help cover the rent from payments to subsidize what the parent regards as an insufficient public-sector salary.” Id. at 550. Thus, this Court made clear the presumption in favor of mens rea was triggered by the need to avoid imposing substantial penalties—including jail sentences—on innocent citizenswho had no idea they were committing a crime.

The concerns animating the presumption in favor of mens rea in Morissette,U.S. Gypsum,Staples, X–Citement Video, and POGO simply are not present here. Section 924(c)(1)(B)(ii) poses no danger of ensnaring “an altar boy [who made] an innocent mistake,” Harris, 959 F.2d at 259, because the government must first prove the defendant is guilty of either drug trafficking or a violent crime, and must further prove that the defendant intentionally used or carried a firearm, or intentionally possessed a firearm, during or in furtherance of that offense. Id. There is thus no risk of unfairness because the defendant “knows from the very outset that his planned course of conduct is wrongful.” United States v. Feola, 420 U.S. 671, 685, 95 S.Ct. 1255, 43 L.Ed.2d 541 (1975). Even NACDL's characterization of Harris as imposing “strict liability” is inaccurate, as the government is still required to establish mens rea with respect to the predicate crime and with respect to the use, carrying, or possession of the firearm.

This is not to say, as Judge Kavanaugh charges, that “the fact that the defendant is a ‘bad person’ who has done ‘bad things' ... justif[ies] dispensing with the presumption of mens rea” entirely. Dissent at 544 (Kavanaugh, J., joined by Tatel, J.). Nor can our opinion be read to mean that a defendant's guilty mind with respect to one type of offense would suffice to allow the imposition of strict liability with respect to a wholly different category of offense. The dissent's “altar boy” hypothetical is thus beside the point. The dissent claims that under our rationale, an altar boy would be guilty of both larceny and drug possession if he stole a collection bag that, unbeknownst to him, contained a stash of cocaine sewn into the lining. See id. But this is not so, because a person who does not know a bag contains drugs does not “knowingly” possess them. If the boy steals the collection bag, knowing that it contains cash, he has the requisite intent with respect to the theft offense and can therefore be found guilty of larceny. But if the bag also happens to contain cocaine, entirely without his knowledge or complicity, he lacks the knowledge statutorily required for guilt of the controlled substances offense.

Nor is it unusual to punish individuals for the unintended consequences of their unlawful acts. Perhaps the most obvious example is the felony-murder rule. In Dean v. United States, 556 U.S. 568, 129 S.Ct. 1849, 1855, 173 L.Ed.2d 785 (2009), the Supreme Court suggested the machinegun provision and the felony murder rule are analytically congruent. Moreover, courts have concluded similarly structured statutes do not require a showing of mens rea. For example, this Court has interpreted 21 U.S.C. § 841, which imposes an additional ten-year penalty for trafficking certain types or quantities of drugs, as not requiring the government to prove the defendant knew how much or what type of drug he was selling. United States v. Branham, 515 F.3d 1268, 1276 (D.C.Cir.2008). Every other circuit to have addressed the question has agreed. See id. at 1275 n. 3 (collecting cases).

Moreover, when asked to infer mens rea requirements in other criminal statutes, neither this Court nor our sister circuits have relied solely on whether a particular provision is an element of the offense or a sentencing factor. Instead, this Court and others have frequently found that certain offense elements do not require proof of an additional mens rea, so long as the offense as a whole carries a scienter requirement that separates innocent from criminal conduct. If O'Brien required the overturning of Harris, it likely would require the overturning of each of these precedents as well, because there is no obvious way to distinguish them. For example, the Drug Free School Zones Act, 21 U.S.C. § 860, provides heightened penalties for drug distribution within 1,000 feet of a school. In a prosecution under that statute, however, the government need not prove the defendant's knowledge of his proximity to a school. See United States v. Holland, 810 F.2d 1215, 1223–24 (D.C.Cir.1987); see also United States v. Jackson, 443 F.3d 293, 299 (3d Cir.2006); United States v. Dimas, 3 F.3d 1015, 1022 (7th Cir.1993). Likewise, when interpreting the Controlled Substances Act, 21 U.S.C. § 841, courts do not require proof of the defendant's knowledge of the type or quantity of the substance, despite each being an element of the offense. See Branham, 515 F.3d at 1275–76;see also, e.g., United States v. King, 345 F.3d 149, 152–53 (2d Cir.2003); United States v. Brower, 336 F.3d 274, 277 (4th Cir.2003); United States v. Villarce, 323 F.3d 435, 439 (6th Cir.2003). Certain statutes involving juveniles, where the victim's age is an element of the offense— e.g. 18 U.S.C. § 2241(c) (criminalizing the crossing of state lines for purpose of engaging in sex with a minor under the age of 12); 18 U.S.C. § 2423 (the Mann Act) (prohibiting transportation of juveniles across state lines for the purpose of prostitution); 21 U.S.C. § 861 (criminalizing use of a juvenile to commit or conceal a drug offense)—do not require proof of mens rea with respect to the juvenile's age. See, e.g., United States v. Cox, 577 F.3d 833, 836–38 (7th Cir.2009) (holding that a prosecution under the Mann Act does not require proof of the defendant's knowledge of the victim's age); United States v. Taylor, 239 F.3d 994, 996–97 (9th Cir.2001) (same); United States v. Chin, 981 F.2d 1275 (D.C.Cir.1992) (holding that a defendant need not know the juvenile's age to be convicted of violating 21 U.S.C. § 861); United States v. Williams, 922 F.2d 737, 738–39 (11th Cir.1991) (same).

Finally, statutes defining offenses by reference to the value of the property taken or damaged, such as 18 U.S.C. § 2113 (defining different bank robbery offenses based on the value of the property stolen) and 18 U.S.C. § 1361 (providing different offense levels and penalties based on the value of the property damaged) do not require the government to prove that the defendant knew the exact monetary value. Absent either a clear statement from the Supreme Court establishing a presumption of mens rea for every element of an offense or a clear demarcation in our caselaw between our treatment of elements and sentencing factors, we cannot say that the conceptual underpinnings of Harris have been weakened at all, much less weakened so much as to justify abandoning it.

Finally, the argument that the machinegun provision in § 924(c)(1)(B)(ii) must carry an implicit mens rea requirement simply because the Court has construed it as an offense element ignores the practical distinction between proving objective facts and subjective mental states. The kind of weapon used, like the type and quantity of drug, is a physical fact, readily susceptible to proof beyond a reasonable doubt. As such, Congress could well intend such factors to be offense elements without intending to include an implicit, subjective mens rea requirement. In that sense, this case is similar to Chin, in which we found that the government is not required to prove the defendant's knowledge of the victim's age in a prosecution under 21 U.S.C. § 861, because it was “implausible that Congress would have placed on the prosecution the often impossible burden of proving, beyond a reasonable doubt, that a defendant knew the youth he enticed was under eighteen.” 981 F.2d at 1280.

With their quiver of arguments almost empty, Burwell and NACDL claim that certain dicta in O'Brien reveal the Court's belief—despite its express reservation of the question discussed above—that § 924(c)(1)(B)(ii) contains an implicit mens rea requirement. In O'Brien, the Supreme Court explained that it is “not likely that Congress intended to remove the indictment and jury trial protections” from the machinegun provision of § 924(c). 130 S.Ct. at 2178. The Court cited several factors, including “[t]he immense danger posed by machineguns [and] the moral depravity in choosing the weapon.” Id. NACDL contends this sentence undermines Harris's statement that “there does not seem to be a significant difference in mens rea between a defendant who commits a drug crime using a pistol and one who commits the same crime using a machine gun.” 959 F.2d at 259. Thus, NACDL claims the O'Brien Court held the varying penalties attached to the different crimes enumerated in § 924(c) “are pegged, inter alia, to the defendant's relative moral blameworthiness, i.e., to differing levels of scienter.” NACDL Br. at 10.

An argument that relies on one sentence of dicta from a Supreme Court opinion is necessarily tenuous, and this one is especially so. Burwell's attempt to pluck this clause out of O'Brien, strip it of all context, and use it as justification for overturning established precedent is unconvincing. In O'Brien, the critical question was whether Congress's decision to amend § 924(c) to provide mandatory minimum sentences

should alter the Court's characterization of the machinegun provision as an element of the offense or a sentencing factor.

O'Brien, 130 S.Ct. at 2178. The Court's statement, in context, is one consideration in the evaluation of one of the five factors bearing on the decision's holding.

As such, it hardly rises to the level of an “intervening [legal] development,” Patterson, 491 U.S. at 173, 109 S.Ct. 2363, let alone one that could fundamentally undermine longstanding precedent.

Moreover, another recent Supreme Court opinion strongly suggests that—contrary to Burwell's argument—the penalties in § 924(c) are not precisely calibrated to the level of mens rea. The Court in Dean concluded the discharge provision of § 924(c)(1)(A)(iii) requires no separate proof of intent. See129 S.Ct. at 1856. In so holding, the Court noted that a § 924(c) defendant whose firearm discharges “is already guilty of unlawful conduct twice over: a violent or drug trafficking offense and the use, carrying, or possession of a firearm in the course of that offense.” Id. at 1855. Accordingly, “[t]he fact that the actual discharge of a gun ... may be accidental does not mean that the defendant is blameless,” and the sentencing enhancement in § 924(c)(1)(A)(iii) properly “accounts for the risk of harm resulting from the manner in which the crime is carried out, for which the defendant is responsible.” Id. To be sure, Dean is not dispositive here, because the Court found that the discharge provision is a sentencing factor rather than a separate element of the offense. Id. at 1854. Still, the Court's analysis suggests that a mens rea requirement is not, as Burwell and his amici suggest, mechanically linked to the various provisions of the statute in accordance with the relative severity of the penalty.

Finally, the FPD argues this Court implicitly overruled Harris in United States v. Brown, which noted that Harris had been “somewhat undermined” by the Supreme Court's statement in Castillo that the difference between carrying a pistol and a machinegun is “great, both in degree and kind.” 449 F.3d 154, 158 (D.C.Cir.2006) (quoting Castillo, 530 U.S. at 126, 120 S.Ct. 2090). But Brown's reading of Harris is more accurately characterized as equivocal, as the Court proceeded to acknowledge that Harris's reading of the machinegun provision “might be reasonable” given “the hazard of the weapon itself (which in almost all instances would likely be obvious to the defendant).” Id. at 158.

Questioning the continuing viability of a precedent is a far cry from implicitly overruling it, particularly where, as in Brown, the precedent is not directly relevant to the issue before the Court.

In sum, we conclude that Burwell and his amici have failed to establish that any intervening legal development has weakened, much less removed, the conceptual underpinnings of Harris. Patterson, 491 U.S. at 173, 109 S.Ct. 2363.

b. Decisions of Other Circuits

This Court may also overrule its established interpretation of a statute “if it finds that other circuits have persuasively argued a contrary construction.” Critical Mass Energy Project, 975 F.2d at 876. Burwell urges us to overturn Harris becauseeach circuit that has decided the machinegun provision is an element of a § 924(c) offense has also assumed that knowledge of the type of weapon is required. See United States v. Franklin, 321 F.3d 1231, 1240 (9th Cir.2003) (finding that evidence was sufficient to show the defendant “knew the weapon was capable of being fired in an automatic setting”); United States v. Rodriguez, 54 Fed.Appx. 739, 747 (3d Cir.2002) (assuming without deciding that knowledge of the type of weapon is an element of the offense); United States v. Dixon, 273 F.3d 636, 640–41 (5th Cir.2001) (assuming without deciding that Castillo makes defendant's knowledge of the weapon's automatic firing capability an element of the offense).

As Burwell seems to concede, however, no circuit has rejected Harris. In fact, none of the cases Burwell cites even consider the analysis in Harris. In Franklin, for example, the court concluded that the evidence presented was sufficient for a rational jury to find that the defendant knew the weapon was capable of being fired in an automatic setting, citing neither § 924(c)(1)(B)(ii) nor Harris's analysis of that provision. 321 F.3d at 1240. Similarly, in both Rodriguez and Dixon, the courts concluded—without wrestling with the larger question of whether knowledge was required—that any failure to submit the element of knowledge to the jury was harmless error. See Rodriguez, 54 Fed.Appx. at 747;Dixon, 273 F.3d at 641. None of these cases can be reasonably construed as rejecting our analysis in Harris.

Moreover, an equal number of circuits have held that the machinegun provision does not contain an implied knowledge requirement, although some have done so based on their conclusion that the provision is a sentencing enhancement. See United States v. Ciszkowski, 492 F.3d 1264, 1268–69 (11th Cir.2007) (basing holding on language of the statute and inapplicability of Staples );United States v. Gamboa, 439 F.3d 796, 812 (8th Cir.2006) (basing holding on the view that the machinegun provision was a sentencing factor); United States v. Eads, 191 F.3d 1206, 1212–14 (10th Cir.1999) (same). All of these cases—including the ones cited by Burwell and his amici—were decided prior to O'Brien, however, so it is unclear how these other circuits will evaluate the impact of that decision on their respective analyses of § 924(c)'s mens rea requirement. At best, Burwell can plausibly claim that three other circuits have implied—though not actually decided—that they would reject Harris if forced to decide the question. One other circuit has implied that it would continue to follow Harris's logic. The possibility of a future circuit split hardly constitutes the “tide of recent judicial developments” necessary to justify overruling an established precedent. Critical Mass Energy Project, 975 F.2d at 876.

c. Fundamental Flaws in Harris's Analysis

The final basis on which an en banc court may set aside its own precedent is “if, on reexamination of an earlier decision, it decides that a panel's holding on an important question of law was fundamentally flawed.” Id. Despite appellant's arguments to the contrary, we remain convinced that Harris interpreted § 924(c) correctly.

Section 924(c)(1)(B)(ii) is silent regarding a mens rea requirement, and the Supreme Court has “ordinarily resist[ed] reading words or elements into a statute that do not appear on its face.” Dean, 129 S.Ct. at 1853. The text of § 924(c)(1)(B)(ii) provides that the machinegun penalty is triggered if the firearm“is” a machinegun—which refers to a state of being that exists “without respect to a specific actor, and therefore without respect to any actor's intent or culpability.” Id. In other words, Congress's grammatical choice telegraphs its intent to eliminate an additional mens rea requirement for that particular provision. After all, a firearm “is” a machinegun, whether the defendant knows it or not.

The structure of the statute and the context of § 924(c)(1)(B)(ii) also suggest that Congress did not intend it to include a mens rea requirement. Dean noted Congress had defined the “brandishing” provision, 18 U.S.C. § 924(c)(1)(A)(ii), to include a mens rea requirement because to “brandish” means “to display ... in order to intimidate.” Id. § 924(c)(4). Congress did not, however, include such a requirement for any of the other provisions in § 924(c). “[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.” Dean, 129 S.Ct. at 1854. Here, it cannot be said that Congress simply forgot about mens rea when it drafted § 924(c), as the drafters of the statute quite clearly chose to require a showing of intent for one particular provision but not for the others.

At oral argument, Burwell nonetheless contended that the implicit scienter requirement in § 924(c) must be applied to the machinegun element in § 924(c)(1)(B)(ii), citing the Supreme Court's decision in Flores–Figueroa v. United States, 556 U.S. 646, 129 S.Ct. 1886, 173 L.Ed.2d 853 (2009), and our own opinion in United States v. Villanueva–Sotelo, 515 F.3d 1234 (D.C.Cir.2008), for the proposition that mens rea must apply to every element of the offense, unless Congress clearly indicates otherwise. See, e.g., Oral Arg. Recording 9:20—9:45. But neither Flores–Figueroa nor Villanueva–Sotelo stands for that sweeping proposition. Rather, the holdings in those cases rested on circumstances not present here: namely, rules of “ordinary English grammar” indicated that the word “knowingly” in 18 U.S.C. § 1028A(a)(1)

be read to apply to “all the subsequently listed elements of the crime,” Flores–Figueroa, 129 S.Ct. at 1890;see generally Villanueva–Sotelo, 515 F.3d at 1239–41, and Congress's use of the statutory title “aggravated identity theft ” suggested that the offender must “know that what he has taken identifies a different real person,” Flores–Figueroa, 129 S.Ct. at 1893;see Villanueva–Sotelo, 515 F.3d at 1246. The question before the Court was thus primarily one of statutory construction—in other words, the Court had to determine how far down the sentence the word “knowingly” traveled. Here, by contrast, § 924(c)(1)(B)(ii) contains no “phrase ... that introduces the elements of [the] crime with the word ‘knowingly.’ ” Flores–Figueroa, 129 S.Ct. at 1891. And the structure and legislative history of § 924(c) contain no clues that Congress intended for an implied scienter requirement to modify the weapon-specific sub-provisions in § 924(c)(1)(B).

Finally, the purpose of § 924(c) is quite clear: “to persuade the man who is tempted to commit a Federal felony to leave his gun at home.” Muscarello v. United States, 524 U.S. 125, 132, 118 S.Ct. 1911, 141 L.Ed.2d 111 (1998). The higher penalties attached to the use of the most dangerous kinds of firearms reflect Congress's desire to create a deterrent commensurate with the increased danger posed by these weapons. See id. (citing legislative history of § 924(c)). Burwell argues, however, that a mens rea requirement is necessary to effectuate Congress's deterrent purpose, as deterrence “assumes that potential violators can anticipate what punishment they might receive.” United States v. Cavera, 550 F.3d 180, 223 (2d Cir.2008) (Sotomayor, J., dissenting in part). Sentencing defendants for using a machinegun without first ensuring that they knew the weapon's characteristics “converts a deterrent statute to one that is mainly punitive.” Reply Brief at 11–12.

In so arguing, appellants adopt an unduly crabbed definition of “deterrence.” To be sure, a statute might aim to deter each individual offender from committing a particular crime (or in this case, choosing a particular weapon to commit a crime), which implicitly requires that the offender make a conscious choice to engage (or not) in a particular course of conduct. But a statute might also attempt to deter offenders more generally through the imposition of a particularly severe penalty for a certain offense. In the case of § 924(c)(1)(B)(ii), for example, Congress likely attached such a steep penalty to the use of a machinegun in an attempt to deter all offenders from using such weapons. This broader understanding of “deterrence” does not require that each individual offender convicted under the statute have mens rea with respect to the machinegun, because the deterrent value of the statute arises out of its capacity to deter future offenders. As the Court stated in Dean, in the course of holding that the discharge provision in § 924(c)(1)(A)(iii) applies even if the defendant did not intend to fire the weapon, “[t]hose criminals wishing to avoid the penalty for an inadvertent discharge can lock or unload the firearm [or] handle it with care during the underlying violent or drug trafficking crime.” 129 S.Ct. at 1856. Similarly, a defendant wishing to avoid the 30–year mandatory minimum for using, carrying, or possessing a machinegun can carefully inspect his weapon, “leave the gun at home, or—best yet—avoid committing the felony in the first place.” Id.

NACDL argues the severity of the penalty for violating § 924(c)(1)(B)(ii)—a mandatory minimum sentence of 30 years' imprisonment—heightens the intuition that Congress would not eliminate the mens rea requirement. NACDL Brief at 19–20. NACDL notes strict liability “public welfare” statutes are “disfavored” and generally involve only light penalties, such as fines or short jail sentences. Staples, 511 U.S. at 607, 616, 114 S.Ct. 1793. They further contend that because the Supreme Court deemed the 10–year sentences in Staples and X–Citement Video to be “harsh,” the machinegun provision's mandatory penalty of a consecutive 30 years to life “can only be described as draconian.” NACDL Brief at 21.

Amicus' attempt to broaden the reach of existing precedent is unconvincing. In Staples, as explained above, the Court's consideration of the severity of the penalty was decidedly narrow. The Court expressly declined to adopt a sweeping rule of construction that would endorse consideration of the severity of the penalty as an element in determining whether mens rea is required. Rather, the Court “note[d] only that where, as here, dispensing with mens rea would require the defendant to have knowledge only of traditionally lawful conduct, a severe penalty is a further factor tending to suggest that Congress did not intend to eliminate a mens rea requirement.” Staples, 511 U.S. at 618, 114 S.Ct. 1793;see also X–Citement Video, 513 U.S. at 72, 115 S.Ct. 464 (evincing the same concern about imposing harsh penalties on actors who had no idea they were violating the law). This is obviously not the case with respect to § 924(c)(1)(B)(ii), as any defendant faced with the machinegun provision's mandatory 30–year penalty must already have been found guilty of a predicate violent crime or drug trafficking offense, and must have intentionally used, carried, or possessed a firearm during and in relation to, or in furtherance of, that offense.

Burwell's comparison of § 924(c) to “strict liability crimes” is inapposite for the same reason. In contrast to traditional “public welfare” offenses, under which the government need not prove mens rea at all, the government is required under this statute to first establish mens rea with respect to the predicate offense, and then to prove that the defendant intentionally used, carried, or possessed a firearm in the course of that crime. Similarly, Burwell's comparison of penalties fails to recognize the fundamentally different contexts in which those penalties are imposed. In both Staples and X–Citement Video, the Court declined to impose 10 years' imprisonment on defendants who would otherwise not be convicted of any crime. Here, by contrast, the defendants already face substantial sentences for committing a violent crime. While an additional 30 years obviously represents a substantial multiple of their sentence, its severity pales in comparison to imposing a lengthy jail sentence on a person who would otherwise be free. Moreover, as noted above, several federal statutes expressly impose severe penalties without requiring mens rea for every element of the offense. See, e.g., Branham, 515 F.3d at 1275–76 (imposing penalty of 10 years to life for possession with intent to distribute certain types and quantities of drugs without requiring proof of additional mens rea under 21 U.S.C. § 841).

Finally, Burwell and the Federal Public Defender argue Harris was fundamentally flawed because it imposes unjust penalties on co-conspirators. They note that without a separate knowledge requirement, “mere” co-conspirators in low-level drug conspiracies might be subjected to 30–year sentences for violent or drug-trafficking crimes committed with machineguns in furtherance of the conspiracy, as long as it is reasonably foreseeable that such crimes would involve guns of any kind. This would be unjust, the FPD argues, if the particular defendant “had no reason to foresee, let alone know, that some member of the conspiracy—any member—would use or possess a machinegun (as opposed to a generic firearm).” FPD Brief at 14. But the premise of this argument is not necessarily correct. The Supreme Court has not extended vicarious liability to situations in which “the substantive offense ... could not be reasonably foreseen as a necessary or natural consequence of the unlawful agreement.” Pinkerton v. United States, 328 U.S. 640, 647–48, 66 S.Ct. 1180, 90 L.Ed. 1489 (1946). Because the machinegun provision is an element of the substantive § 924(c) offense, it is not clear (and we express no opinion as to) whether liability would attach to co-conspirators who could not reasonably foresee the use of the machinegun.

For these reasons, we remain unpersuaded that Harris was fundamentally flawed as a matter of law, and therefore conclude appellant failed to demonstrate any of the considerations that would justify overruling Harris. Accordingly, we need not reach appellant's claim that if § 924(c) contains a knowledge requirement, the evidence was insufficient to sustain his conviction.

III

Appellant also claims that in light of O'Brien, the rule of lenity requires the Court to vacate his conviction under § 924(c)(i)(B)(ii). Appellant's Br. at 28. The rule of lenity prevents the interpretation of a federal criminal statute “so as to increase the penalty that it places on an individual when such an interpretation can be based on no more than a guess as to what Congress intended.” United States v. Villanueva–Sotelo, 515 F.3d 1234, 1246 (D.C.Cir.2008). But to invoke the rule of lenity, a court must conclude that “there is a grievous ambiguity or uncertainty in the statute.” Muscarello, 524 U.S. at 139, 118 S.Ct. 1911 (emphasis added). “The simple existence of some statutory ambiguity ... is not sufficient to warrant application of that rule, for most statutes are ambiguous to some degree.” Id. at 138, 118 S.Ct. 1911. There is no grievous ambiguity here because, as this Court held twenty years ago in Harris, the structure, statutory context, and purpose of § 924(c)(1)(B)(ii) all make clear that the provision does not contain a separate mens rea requirement. Indeed, as the Supreme Court held in Dean when declining to apply the rule of lenity to § 924(c)(1)(A)(iii), “the statutory text and structure convince us that” Congress did not require proof of any additional mens rea, and Burwell's “contrary arguments are not enough to render the statute grievously ambiguous.” 129 S.Ct. at 1856.

IV

Because the principle of stare decisis is “of fundamental importance to the rule of law,” Welch, 483 U.S. at 494, 107 S.Ct. 2941, this Court imposes a substantial burden on a party advocating the abandonment of an established precedent. We will not overturn our prior decision simply because a reading of Supreme Court dicta might support some inference that the Court might, in some future case, come to question our approach. Our job is simply to apply the law as it currently exists.

Judge Kavanaugh's dissent makes two dubious claims. First, he argues that the Supreme Court's precedents definitively establish that neither statutory silence, nor Congress's decision to include mens rea in certain parts of a statute but omit it in others, serves to defeat the presumption of mens rea. Second, he contends that the Supreme Court has established and applied a rule of statutory interpretation for federal crimes wherein the Court imputes the presumption of mens rea to each element of an offense unless plainly— i.e., explicitly—indicated otherwise. Flores–Figueroa is the lever with which the dissent proposes to upend decades of precedent and establish the bona fides of these otherwise unmoored assertions. But that case is not up to the task.

Judge Kavanaugh's attempt to analogize our position to that of the government in Flores–Figueroa is inapt, because that case—as he acknowledges—involved a statute containing an explicit mens rea requirement. See18 U.S.C. § 1028A (punishing someone who “knowingly transfers, possesses, or uses ... a means of identification of another person” while committing an enumerated predicate crime); 129 S.Ct. at 1888–89.Section 924(c)(1)(B)(ii) contains no such language. Judge Kavanaugh's argument boils down to the assertion that we should rewrite the text of § 924(c) and imply a mens rea requirement where none was meant to exist, in the service of an assertedly strong “traditional presumption of mens rea ” applicable to every element of an offense.

Historically, the altar boy archetype, i.e., innocent conduct, justified imposition of an extratextual gloss on statutes that lacked an explicit scienter requirement. In Flores–Figueroa, the Court rejected the government's argument that the absence of innocence should circumscribe the reach of an explicit mens rea requirement. Judge Kavanaugh insists this portends a major shift in the Court's jurisprudence. Perhaps. But ignoring the lack of an innocence rationale where a statute contains an express requirement does not mean innocence is irrelevant where the statute is silent. Indeed, the Court's strongly textual approach in Flores–Figueroa counsels against judicial creation of a mens rea requirement for every element in the face of statutory silence.

Judge Kavanaugh, applying rules of his own creation, ignores the role of innocence and concludes the mens rea presumption—writ large—still justifies an extratextual (or even countertextual) reading of the statute, because that background presumption applies to every element of a criminal offense unless Congress expressly disclaims it. Nothing in Flores–Figueroa or any other Supreme Court precedent supports that result.

Judge Rogers' approach is even more unbounded. Her solution—a balancing test completely unmoored from circuit or Supreme Court precedent—is substantially broader than anything we have proposed. See id. at 526 (“I would take my lead from Staples and simply hold that the thirty-year term of imprisonment ... is so severe in length that it outweighs the fact that the conduct prohibited is not otherwise innocent[.]”). Nowhere does Staples, or any other case, suggest these two considerations should be balanced in order to determine whether a court should imply a mens rea requirement when faced with statutory silence. If anything, such an open-ended test creates a much greater potential for spillover into a “host of other applications” than does our solution, which is firmly rooted in the text and structure of § 924(c)(1)(B)(ii).

Finding nothing in the Supreme Court's holdings, our own caselaw, or “deeply rooted principles of law and justice,” Kavanaugh Dissent at 527, that would justify overturning our decision in Harris, we affirm our previous conclusion that § 924(c)(1)(B)(ii) does not require the government to prove that a defendant knew that the weapon he used, carried, or possessed was a machinegun. Accordingly, we reinstate the panel opinion and affirm appellant's conviction.

SENTELLE, Chief Judge, concurring:

Although I confess to having come late to the wisdom expressed by Judge Henderson, I join her view that the procedural background of this case and the applicable burden of proof make it apparent in the end that we have improvidently granted en banc review. Upon examining the merits of the case, I recall the conclusion previously expressed by one of our colleagues in an earlier concurrence in an en banc decision: “Because I believe that in a close en banc case prior precedent is entitled to some respect, I join the majority opinion.” United States v. Mills, 964 F.2d 1186, 1194 (D.C.Cir.1992) (Silberman, J., concurring).

As the majority notes, this is a statutory question, albeit with constitutional implications. Errors in statutory interpretation are reparable by congressional action. Errors in our interpretation are reparable by the action of the Supreme Court. Therefore, when the question is a close one—and this one I think is exceedingly close—I will accept the weight of precedent and vote with the majority to leave undisturbed this circuit's controlling interpretation of 18 U.S.C. § 924(c)(1)(B)(ii). Nonetheless, in doing so, I agree with the dissent that the “altar boy argument” of the majority is unconvincing and perhaps stands at the head of some slippery slope. I do think it not good enough to posit that someone should know conduct is illegal and therefore to avoid it where the illegality of the conflict may be measured in degrees. That is, the dissent poses a fair question, quoted from the Supreme Court when it asked: “Would we apply a statute that makes it unlawful ‘ knowingly to possess drugs' to a person who steals a passenger's bag without knowing that the bag has drugs inside?” Dissent at 536 (quoting Flores–Figueroa v. United States, 556 U.S. 646, 650, 129 S.Ct. 1886, 173 L.Ed.2d 853 (2009)).

I do not suggest that the defendant in the present case has been inflicted any great injustice. Judge Henderson's separate concurrence points out the futility of making such a suggestion. Nonetheless, both the majority and the dissent express strong and well-reasoned arguments for their positions, and, as I stated above, I concur with the majority because of the stability principle inherent in our doctrine of stare decisis.

KAREN LeCRAFT HENDERSON, Circuit Judge, concurring:

I wholeheartedly join the majority opinion but am at a loss why its writing became necessary in the first place. En banc review serves two purposes: to ensure the consistency of our caselaw and to resolve issues of exceptional importance. Fed. R.App. P. 35(a). Because this appeal serves neither purpose, I believe the rehearing petition should have been summarily denied and the panel decision left intact.

First, en banc consideration is not “necessary to secure or maintain uniformity of the court's decisions.” Fed. R.App. P. 35(a)(1). The sole issue on en banc review is whether 18 U.S.C. § 924(c)(1)(B)(ii) “requires the government to prove that the defendant knew the weapon he was carrying was capable of firing automatically.” Maj. Op. at 502. Circuit law on this issue has been clear and consistent for twenty years. In United States v. Harris, 959 F.2d 246 (D.C.Cir.1992), we rejected the defendant's argument that “the government must show the defendant knew the precise nature of the weapon,” concluding instead that it need show only that the defendant “knowingly or intentionally possessed a firearm, and that [he] did so intentionally to facilitate” one of the two types of predicate offenses identified in section 924(c)(1)(B)(ii)—either a crime of violence or a drug trafficking crime. Harris, 959 F.2d at 259. None of our precedents—either before or since Harris—calls its holding into question. Nor, as the majority opinion so deftly demonstrates, did the United States Supreme Court's decision in United States v. O'Brien, which, acknowledging the “case d[id] not require the Court to consider any contention that a defendant who uses, carries, or possesses a firearm must be aware of the weapon's characteristics,” “expresse[d] no views on the point.” ––– U.S. ––––, 130 S.Ct. 2169, 2173, 176 L.Ed.2d 979 (2010).

The issue on rehearing likewise falls short on the second ground for en banc review, that is, that it is plainly not “a question of exceptional importance,” Fed. R.App. P. 35(a)(2). Because Burwell's trial counsel failed to request an instruction that the jury be required to find Burwell knew of the gun's automatic capability,

we treat the court's failure to give such an instruction less rigorously and review it for plain error only. See United States v. Laureys, 653 F.3d 27, 32 (D.C.Cir.2011).

Under the plain error standard, Burwell “would have to establish (1) a legal error that was (2) plain (a term that is synonymous with clear or obvious), and that (3) affected his substantial rights.” Id. (quotation marks and alterations omitted). Even if the failure to charge mens rea were error so as to satisfy the first prong of the test (a possibility the majority opinion definitively refutes), it would not satisfy the second and third prongs. It could not be “plain” because it “contradicted no precedents of this Court or the Supreme Court.” Id. And it could not “affect [Burwell's] substantial rights” because it did not “unfairly prejudice” him given the overwhelming evidence he was aware the firearm was an automatic weapon. United States v. Mahdi, 598 F.3d 883, 891 n. 8 (D.C.Cir.2010). The trial testimony reflects that the gang members decided to use military weapons such as AK–47s instead of handguns because they believed the metropolitan police “wouldn't respond” if they “robb[ed] banks with assault weapons.” Trial Tr. at 3950. And they used their AK–47s repeatedly and with abandon.

Moreover, when Burwell carried the AK–47 to perform “crowd control” during one robbery, a circular 75–round drum magazine was affixed to the gun. With that attachment, the gun's automatic potential was eye-popping—particularly to a gun afficionado like Burwell.

See Appellee's Opp'n at 2–3 & n.3. Even the manager of the target bank, who “d[id]n't really know about guns,” testified it “look[ed] like a machine gun type of gun.” Trial Tr.1916. Given the overwhelming evidence Burwell knew the gun's capability, if the court had charged the jury on mens rea as Burwell now tardily presses, it could not have avoided finding that he knew the weapon was a machine gun and convicted him in any event.

Accordingly, the failure to so instruct the jury did not unfairly prejudice Burwell.

Because there is no conflicting Supreme Court or Circuit precedent and the issue was contested so late in the proceeding, this appeal is an egregiously inappropriate candidate for en banc review. That en banc review was nonetheless granted only prolonged Burwell's meritless appeal and squandered the court's and the parties' resources.

ROGERS, Circuit Judge, dissenting:

For many of the reasons stated by Judge Kavanaugh, I would vacate the judgment of conviction under 18 U.S.C. § 924(c)(1)(B)(ii). The majority concludes, in applying the doctrine of stare decisis, that “Burwell and his amici have failed to establish that any intervening legal development has weakened, much less removed, the conceptual underpinnings of Harris.” Ante at 510 (Brown, J., majority op.) (emphasis added). But the Supreme Court has twice stated that carrying a machinegun involves heightened culpability. See United States v. O'Brien, ––– U.S. ––––, 130 S.Ct. 2169, 2178, 176 L.Ed.2d 979 (2010) (stating that “choosing” a machinegun involves “moral depravity”); Castillo v. United States, 530 U.S. 120, 126, 120 S.Ct. 2090, 147 L.Ed.2d 94 (2000) (“[T]he difference between carrying, say, a pistol and carrying a machinegun ... is great, both in degree and kind.”). Undeniably, our holding in United States v. Harris, 959 F.2d 246, 259 (D.C.Cir.1992), that conviction under 18 U.S.C. § 924(c)(1)(B)(ii) did not require proof of mens rea, was premised on the opposite view: “[T]here does not seem to be a significant difference in mens rea between a defendant who commits a drug crime using a pistol and one who commits the same crime using a machine gun; the act is different, but the mental state is equally blameworthy,” id. Dicta or not, see ante at 542 (Brown, J., majority op.), the Supreme Court has twice rejected a key basis underlying Harris's conclusion that there is no mens rea requirement in § 924(c)(1)(B)(ii).

See Patterson v. McLean Credit Union, 491 U.S. 164, 173, 109 S.Ct. 2363, 105 L.Ed.2d 132 (1989); see also Montejo v. Louisiana, 556 U.S. 778, 792–93, 129 S.Ct. 2079, 173 L.Ed.2d 955 (2009); D.C. CIR. R. 35(a)(2).

To date, the Supreme Court's precedent regarding whether a criminal statute should be interpreted as requiring proof of mens rea has involved: (1) statutes that contained an explicit mens rea, but were unclear as to how far the mens rea requirement should “travel”; (2) statutes that were silent but, absent a showing of mens rea, risked criminalizing otherwise innocent conduct; or (3) statutes with minor punishments deemed “public welfare” offenses, where the Court was less concerned with dispensing with a mens rea requirement. Section 924(c)(1)(B)(ii) does not risk criminalizing otherwise innocent conduct in this sense, but the mandated thirty-year minimum, consecutive term of imprisonment means the public welfare exception is inapposite. In the absence of controlling precedent, and given that Harris's scienter premise has been rejected, I would hold, based on the following analysis, that the severe additional punishment mandated by section 924(c)(1)(B)(ii) (magnitudes greater than any of the other statutes considered in the relevant Supreme Court precedent) requires for conviction proof of the defendant's knowledge that the firearm was a machinegun. Accordingly, I respectfully dissent.

I.

In concluding that the fact that the firearm was a machinegun under section 924(c)(1)(B)(ii) is an element of the offense, rather than a sentencing factor, the Supreme Court acknowledged in O'Brien that the structure of section 924(c) demands escalating terms of imprisonment for increasingly culpable conduct. See130 S.Ct. at 2178. But the Court also stated that it “expresse[d] no views” on whether the provision required proof of mens rea. Id. at 2173. The issue was not before the Court; indeed, to decide it would have required quite a diversion: section 924(c)(1)(B)(ii) does not fit well with Supreme Court precedent determining whether a statute requires proof of mens rea.

Two background principles underlie consideration of Burwell's statutory challenge. The first is that “determining the mental state required for commission of a federal crime requires ‘construction of the statute and ... inference of the intent of Congress.’ ” Staples v. United States, 511 U.S. 600, 604–05, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994) (quoting United States v. Balint, 258 U.S. 250, 253, 42 S.Ct. 301, 66 L.Ed. 604 (1922)). The second is that “silence on this point by itself does not necessarily suggest that Congress intended to dispense with a conventional mens rea element” because courts must “construe the statute in light of the background rules of the common law, in which the requirement of some mens rea for a crime is firmly embedded.” Id. at 605, 114 S.Ct. 1793 (citing United States v. United States Gypsum Co., 438 U.S. 422, 436–37, 98 S.Ct. 2864, 57 L.Ed.2d 854 (1978)). The Supreme Court has explained:

There can be no doubt that this established concept has influenced our interpretation of criminal statutes. Indeed, we have noted that the common-law rule requiring mens rea has been “followed in regard to statutory crimes even where the statutory definition did not in terms include it.” Relying on the strength of the traditional rule, we have stated that offenses that require no mens rea generally are disfavored, and have suggested that some indication of congressional intent, express or implied, is required to dispense with mens rea as an element of a crime.
Id. at 605–06, 114 S.Ct. 1793 (quoting Balint, 258 U.S. at 251–52, 42 S.Ct. 301) (internal citations omitted).

In addressing these background principles, the Court has adopted three general rules of interpretation:

First, “courts ordinarily read a phrase in a criminal statute that introduces the elements of a crime with the word ‘knowingly’ as applying that word to each element.” Flores–Figueroa v. United States, 556 U.S. 646, 652, 129 S.Ct. 1886, 173 L.Ed.2d 853 (2009) (citing United States v. X–Citement Video, Inc., 513 U.S. 64, 79, 115 S.Ct. 464, 130 L.Ed.2d 372 (1994) (Stevens, J., concurring) (“[T]he normal, commonsense reading of a subsection of a criminal statuteintroduced by the word ‘knowingly’ is to treat that adverb as modifying each of the elements of the offense identified in the remainder of the subsection.”)). This is a rule fundamentally about grammar—the mens rea is read to “travel” through to the end of the sentence (or statutory subsection).

Second, where the statute is silent as to mens rea, so the first rule provides no source of a mens rea requirement, the Supreme Court has applied a presumption of mens rea to avoid criminalizing otherwise innocent conduct. “[T]he presumption in favor of a scienter requirement should apply to each of the statutory elements that criminalize otherwise innocent conduct.” United States v. X–Citement Video, Inc., 513 U.S. 64, 72, 115 S.Ct. 464, 130 L.Ed.2d 372 (1994); see also, e.g., Staples, 511 U.S. at 614–15, 114 S.Ct. 1793;Liparota v. United States, 471 U.S. 419, 426, 105 S.Ct. 2084, 85 L.Ed.2d 434 (1985).

Third, the Court has recognized an exception to the background principle disfavoring strict liability crimes for “public welfare” or “regulatory” offenses. “In construing such statutes, [the Court] ha[s] inferred from silence that Congress did not intend to require proof of mens rea to establish an offense.” Staples, 511 U.S. at 606, 114 S.Ct. 1793. Public welfare offenses have been recognized in “ ‘limited circumstances.’ Typically, [ ] cases recognizing such offenses involve statutes that regulate potentially harmful or injurious items.” Id. (quoting Gypsum, 438 U.S. at 437, 98 S.Ct. 2864). But public welfare offenses have “almost uniformly involved statutes that provided for only light penalties such as fines or short jail sentences.” Id. at 616, 114 S.Ct. 1793. The Court in Staples rejected labeling a statute imposing a ten year prison term as a public welfare offense, noting that “[h]istorically, the penalty imposed under a statute has been a significant consideration in determining whether the statute should be construed as dispensing with mens rea.” Id.

In sum, these traditional rules of statutory interpretation apply, except that silence is not evidence of an absence of a mens rea requirement (at least where the conduct at issue is otherwise innocent). Silence does, however, provide such evidence where the public welfare offense exception applies. Together these three rules suggest that if neither of the first two rules are able to supply a mens rea, then the statute will fall within the third category of public welfare offenses.

Except when, as here, it does not.

A.

To date, the three interpretative rules have worked in tandem in Supreme Court precedent to counsel for or against requiring proof of mens rea. For example, in construing whether the criminal provisions of the Sherman Antitrust Act required proof of mens rea, the Court in Gypsum noted that, although the Act contained no “mention of intent or state of mind,” 438 U.S. at 438, 98 S.Ct. 2864, “the behavior proscribed by the Act is often difficult to distinguish from the gray zone of socially acceptable and economically justifiable business conduct,” id. at 440–41, 98 S.Ct. 2864. The Court thus concluded proof of mens rea was required, in light of the “generally inhospitable attitude to non- mens rea offenses.” Id. at 438, 98 S.Ct. 2864.

Likewise, in Liparota, the Court considered a statute imposing up to five years imprisonment for unlawfully acquiring and possessing food stamps, 7 U.S.C. § 2024(b)(1), see Liparota, 471 U.S. at 420–21, 105 S.Ct. 2084. The statute in Liparota included an explicit mens rea with respect “to some element of the crime,” but was unclear how far through the statute the word “knowingly” traveled. Id. at 424, 105 S.Ct. 2084. The Court concluded proof of mens rea was required because “to interpret the statute otherwise would be to criminalize a broad range of apparently innocent conduct.” Id. at 426, 105 S.Ct. 2084. Likewise, the Court rejected the contention that the statute was a public welfare offense, because food stamps were far from “a type of conduct that a reasonable person should know is subject to stringent public regulation and may seriously threaten the community's health or safety.” Id. at 433, 105 S.Ct. 2084.

The Court in Staples was presented with a statute criminalizing possession of unregistered firearms, including machineguns, with punishment of up to ten years' imprisonment, 26 U.S.C. § 5861(d); id. § 5845(a)(6), see Staples, 511 U.S. at 602–03, 114 S.Ct. 1793. Although the statute contained no explicit mens rea to “travel” through the subsection, the Court rejected the Government's argument that the provision was a public welfare offense, noting the “long tradition of widespread lawful gun ownership by private individuals in this country.”

Of course, we might surely classify certain categories of guns—no doubt including the machineguns, sawed-off shotguns, and artillery pieces that Congress has subjected to regulation—as items the ownership of which would have the same quasi-suspect character we [have previously attributed to public welfare offenses.] But precisely because guns falling outside those categories traditionally have been widely accepted as lawful possessions, their destructive potential[ ] ... cannot be said to put gun owners sufficiently on notice of the likelihood of regulation....
Id. at 611–12, 114 S.Ct. 1793.

The Court also observed that because “any semiautomatic weapon may be converted, either by internal modification or, in some cases, simply by wear and tear, into a machinegun within the meaning of the Act,” id. at 615, 114 S.Ct. 1793, imposing strict liability “potentially would impose criminal sanctions on a class of persons whose mental state—ignorance of the characteristics of weapons in their possession—makes their actions entirely innocent,” id. at 614–15, 114 S.Ct. 1793. Finally, the Court concluded that the “harsh” ten years' imprisonment imposed by the statute, combined with the otherwise innocent nature of the conduct, meant the statute was not a public welfare offense and that it required proof of mens rea. Id. at 616, 618–19, 114 S.Ct. 1793.

Similarly, in X–Citement Video, the Court considered a statute prohibiting “the interstate transportation, shipping, receipt, distribution, or reproduction of visual depictions of minors engaged in sexually explicit conduct,” 513 U.S. at 65–66, 115 S.Ct. 464 (citing 18 U.S.C. § 2252). The statute contained the mens rea “knowingly,” but it was unclear whether it should “travel” through the entire subsection to require proving the defendant knew the age of the person depicted in the video. Id. at 68, 115 S.Ct. 464. The Court concluded that the mens rea traveled through the entire subsection, noting that the age of the person depicted in the video was “the crucial element separating legal innocence from wrongful conduct.” Id. at 73, 115 S.Ct. 464. Furthermore, the Court concluded the statute was not a public welfare offensebecause “persons do not harbor settled expectations that the contents of magazines and film are generally subject to stringent public regulation” and the “harsh penalt[y]” of “up to 10 years in prison” counseled against the statute being a strict liability public welfare offense. Id. at 71–72, 115 S.Ct. 464.

B.

The analysis in Dean v. United States, 556 U.S. 568, 129 S.Ct. 1849, 173 L.Ed.2d 785 (2009), might at first glance suggest a conflict among the three rules, but it too is consonant with them. The Court held that 18 U.S.C. § 924(c)(1)(A)(iii) does not require proof of mens rea. That subsection provides that if the firearm present during a violent or drug trafficking crime “is discharged,” the defendant shall “be sentenced to a term of imprisonment of not less than 10 years.” 18 U.S.C. § 924(c)(1)(A)(iii). This provision, unlike the ones considered by the Court in the previously discussed cases, was a sentencing factor rather than an offense element. Dean, 556 U.S. at 571, 574, 129 S.Ct. 1849. The Court noted that it “ordinarily resist[s] reading words or elements into a statute that do not appear on its face,” id. at 572, 129 S.Ct. 1849 (internal quotation marks and citation omitted), highlighted that the passive voice of the statute “focuses on an event ... without respect to any actor's intent or culpability,” id., and compared the provision with another provision of the statute that did include a mens rea: the “brandishing” provision, 18 U.S.C. § 924(c)(1)(A)(ii), see Dean, 556 U.S. at 572–73, 129 S.Ct. 1849. Applying the tools of statutory construction, the Court declined to “contort[ ] and stretch[ ] the statutory language to imply an intent requirement.” Id. at 574, 129 S.Ct. 1849.

Upon concluding that the text and structure of the statute did not support requiring proof of mens rea, the Supreme Court turned to the defendant's contention that the presumption of mens rea should apply. The Court rejected the applicability of the presumption, reasoning that the presumption line of cases involved situations where the conduct at issue would be innocent if the facts were as the defendant believed them. “It is unusual to impose criminal punishment for the consequences of purely accidental conduct. But it is not unusual to punish individuals for the unintended consequences of their unlawful acts.” Id. at 575, 129 S.Ct. 1849 (emphasis in original) (citing, inter alia, 4 W. Blackstone, Commentaries on the Laws of England 26–27 (1769)). The Court noted that “the defendant [was] already guilty of unlawful conduct twice over: a violent or drug trafficking offense and the use, carrying, or possession of a firearm in the course of that offense.” Id.

Thus, neither of the first two interpretative rules—grammatical rules of statutory construction nor the presence of otherwise innocent conduct—counseled in favor of requiring proof of mens rea, and the Court thus held that no such proof was required. Id. at 577, 129 S.Ct. 1849. In so holding, the Court did not, however, classify the provision as a public welfare offense. Nor did it frame the question before it as a choice between offenses that have mens rea requirements and public welfare offenses that do not. See id. at 580, 129 S.Ct. 1849 (Stevens, J., dissenting); Staples, 511 U.S. at 606, 114 S.Ct. 1793 (describing public welfare offenses as exception to rule that congressional silence as to mens rea is not read to mean Congress intended to dispense with requirement). This departure from the usual understanding of the public welfare offenses as being among the “few narrowly delineated exceptions,” Dean, 556 U.S. at 580, 129 S.Ct. 1849 (Stevens, J., dissenting), is explained by the fact that the “discharge” provision at issue in Dean was not an “offense” at all, and thus the public welfare offense exception, and its general limitation to statutes with minor punishments, see Staples, 511 U.S. at 616, 114 S.Ct. 1793, did not influence the Court's conclusion that the “discharge” provision required no proof of mens rea despite its ten year mandatory prison term. Instead, the provision was a sentencing factor, which “involve[s] special features of the manner in which a basic crime was carried out.” Id. at 574, 129 S.Ct. 1849 (internal quotation marks, citation, and alteration omitted).

Because the public welfare offense precedent was inapplicable to the sentencing factor at issue in Dean, the Court had no occasion to decide whether it was creating a new type of strict liability crime, outside the traditional conception of public welfare offenses. The general purpose of sentencing factors is to increase punishment for a proven offense. See Almendarez–Torres v. United States, 523 U.S. 224, 228, 118 S.Ct. 1219, 140 L.Ed.2d 350 (1998). Because the provision did not deal with an offense, the Court merely had to apply the traditional tools of statutory construction, including the presumption of mens rea for statutes that risk criminalizing otherwise innocent conduct.

C.

In the case before the court today, the interpretative rules, in their present form, cannot provide the answer to whether the machinegun provision requires proof of mens rea. Section 924(c)(1)(B)(ii) has no explicit mens rea to “travel” through the subsection, and it does not risk criminalizing “entirely innocent” conduct. Thus neither of these two rules counsel in favor of requiring proof of mens rea. But nor do they necessarily require imposition of strict liability either. The narrow exception to the rule favoring proof of mens rea, the public welfare offense exception, does not apply—the Supreme Court has concluded that firearm statutes do not fit within the type of crimes contemplated by the public welfare exception, see Staples, 511 U.S. at 611–12, 114 S.Ct. 1793, and has suggested that the exception has been generally limited to statutes with minor penalties, id. at 616, 618–19, 114 S.Ct. 1793.Section 924(c)(1)(B)(ii) would represent an entirely new category of strict liability for offense elements—one that imposes a sentence magnitudes larger than any of the statutes in the relevant cases the Court has previously considered.

In this uncharted territory, courts reasonably rely on clues from relevant Supreme Court case law. Taking the word “element” in Flores–Figueroa's description of the first rule (despite the context indicating a grammatical rule, as opposed to a distinction between “elements” and “sentencing factors”), 556 U.S. at 652, 129 S.Ct. 1886, and combining it with a dissected phrase from the second rule, Judge Kavanaugh concludes: “the presumption of mens rea means that, unless Congress plainly indicates otherwise, the Government must prove the defendant's mens rea for each element of the offense,” post at 528 (Kavanaugh, J., dissenting), regardless of whether that “element” is anywhere near an explicit mens rea, or whether the statute risks criminalizing otherwise innocent conduct. It may be somewhat of a stretch to take half of one inapplicable rule, plus half of another inapplicable rule, and declare it to be the “established and applied [ ] rule,” id. at 537, governing Burwell's case.

Rather than setting out a new rule such as this, which the majority fears potentially sweeps into its ambit a host of other applications, see ante at 507–09 (Brown, J., majority op.), this case can be resolved on a narrower ground. The Supreme Court in Staples stated that “[h]istorically, the penalty imposed under a statute has been a significant consideration in determining whether the statute should be construed as dispensing with mens rea.” 511 U.S. at 616, 114 S.Ct. 1793 (emphasis added). Although this statement was made in the context of holding that the public welfare exception was inapplicable, there is no obvious reason to limit the relevance of this consideration to determining whether that exception applies to impose strict liability. If an offense must have a minor punishment to be a public welfare offense, and in so limiting that category the Court was concerned with protecting the background principle disfavoring non- mens rea crimes, then the presence of a “severe penalty,” id. at 618, 114 S.Ct. 1793, should not, without an express statement by Congress, portend a new category of strict liability crimes.

The majority characterizes this conclusion as “broaden[ing] the reach of existing precedent ... [because] [i]n Staples, ... the Court's consideration of the severity of the penalty was decidedly narrow.” Ante at 513 (Brown, J., majority op.). Although the Court in Staples declined to adopt “a definitive rule of construction to decide [that] case,” 511 U.S. at 618, 114 S.Ct. 1793, the Court was not discussing the general relevance of the severity of a punishment to whether a statute requires proof of mens rea. Instead the Court was considering whether “absent a clear statement from Congress that mens rea is not required, [courts] should not apply the public welfare offense rationale to interpret any statute defining a felony offense as dispensing with mens rea.” Id. (emphasis added). The Court was not commenting on whether the severity of the penalty imposed is broadly relevant to the mens rea question outside the context of public welfare offenses, but simply declining to establish a rule definitively limiting the four-corners of public welfare offenses to exclude felonies.

Understood as an independent consideration, rather than as one tethered to the definition of a public welfare offense, the length of the punishment imposed by a statute must be considered in light of the purpose of the mens rea presumption and whether it is rebutted where the statute does not, as the interpretive rule has been defined, criminalize otherwise innocent conduct. In the context of statutes that, absent a showing of mens rea, risked criminalizing otherwise innocent conduct, the Supreme Court has described a punishment of ten years' imprisonment as “harsh” and “severe.” Staples, 511 U.S. at 616, 618, 114 S.Ct. 1793;X–Citement Video, 513 U.S. at 72, 115 S.Ct. 464. “In such a case, the usual presumption that a defendant must know the facts that make his conduct illegal should apply.” Staples, 511 U.S. at 619, 114 S.Ct. 1793. But where the traditional presumption of mens rea does not apply because the conduct prohibited is not, in the absence of intent, otherwise innocent, a ten-year penalty may not be so harsh as to require more than congressional silence to dispose of the mens rea requirement. See, e.g., ante at 508–09 (Brown, J., majority op.) (citing 18 U.S.C. § 2113, which increases maximum term of imprisonment for certain bank robbery offenses from one to ten years when value of stolen property, money, or thing of value exceeds $1,000); cf. Carter v. United States, 530 U.S. 255, 272–73, 120 S.Ct. 2159, 147 L.Ed.2d 203 (2000).

Without deciding how the interpretative rules might unfold in a case with a different statutory punishment, I would take my lead from Staples and hold that the mandated thirty-year consecutive term of imprisonment imposed by section 924(c)(1)(B)(ii) is so severe that it outweighs the fact that the conduct prohibited is not otherwise totally innocent. In O'Brien, the Supreme Court defined the section 924(c)(1)(B)(ii) offense in terms of the automatic firing characteristic of the firearm, see130 S.Ct. at 2172, given “[t]he immense danger” and “moral depravity in choosing the weapon,” id. at 2178. Congress has determined that possessing a machinegun “during and in relation to any crime of violence or drug trafficking crime,” 18 U.S.C. § 924(c)(1)(A), warrants twenty years' additional imprisonment over such possession of a semi-automatic assault weapon, compare id. § 924(c)(1)(B)(i), with id. § 924(c)(1)(B)(ii). A leading commenter has observed that the hostility to strict liability for crimes protects a defendant “unaware of the magnitude of the wrong he is doing,” Wayne R. LaFave, Criminal Law 304 (5th ed.2010) (emphasis added); see post at 543 (Kavanaugh, J., dissenting); ante at 516–17 (Sentelle, C.J., concurring) (“I do not think it good enough to posit that someone should know conduct is illegal and therefore avoid it where the illegality of the conduct may be measured in degrees.”). The Court's observation in Staples, 511 U.S. at 614–15, 114 S.Ct. 1793, is directly on point: “The Government does not dispute ... that virtually any semiautomatic weapon may be converted ... into a machinegun.... Such a gun may give no externally visible indication that it is fully automatic. But in the Government's view, any person ... can be subject to imprisonment, despite absolute ignorance of the gun's firing capabilities, if the gun turns out to be automatic.” (internal citations omitted). For punishment, then, of an additional twenty-years' imprisonment for possessing an automatic, rather than a semi-automatic, firearm, neither silence, nor mere use of a passive voice or the meaning of surrounding provisions, see Dean, 556 U.S. at 572–74, 129 S.Ct. 1849, are sufficient to rebut the traditional mens rea requirement for criminal offenses, and the Government must prove a defendant knew the firearm he possessed was a machinegun in order to obtain a conviction under section 924(c)(1)(B)(ii).

This is not, as the majority charges, an “unbounded” approach, ante at 516 (Brown, J., majority op.); instead it is bound quite tightly by the uniquely severe mandatory term of consecutive imprisonment,as demonstrated by the dearth of cases considering whether proof of mens rea is required in the face of such severe punishment as section 924(c)(1)(B)(ii) mandates. Nor is it unusual, or posing a risk of “spillover,” for courts to balance competing considerations that point in opposite directions; this is what courts usually do anytime they are faced with a question informed by more than one consideration. Cf. United States v. O'Brien, 542 F.3d 921, 924 (1st Cir.2008) (observing that “the [Supreme] Court has developed unique policy and historical tests that complement, and sometimes work to modify, the most straightforward reading of the [statutory] language and structure.”). Indeed the majority does some balancing of its own for, as Judge Kavanaugh points out, see post at 546, by requiring proof of knowledge with respect to the use, carrying, or possession of a firearm under section 924(c), see ante at 503, 507 (Brown, J., majority op.), the majority itself does not limit the presumption solely to statutes that risk criminalizing otherwise innocent conduct—section 924(c) is only triggered upon conviction for a “crime of violence or drug trafficking crime.” I take the Supreme Court at its word when it stated that “the penalty imposed by a statute has been a significant consideration in determining whether the statute should be construed as dispensing with mens rea,” Staples, 511 U.S. at 616, 114 S.Ct. 1793 (emphasis added). It hardly seems a better solution to ignore that consideration in an effort to avoid balancing it against competing concerns, particularly when the Supreme Court has concluded that the “choosing” of a machinegun under section 924(c)(1)(B)(ii) involves “moral depravity,” O'Brien, 130 S.Ct. at 2178, and that mens rea is designed to protect a defendant against the risk of a more severe punishment by possessing a firearm that may appear physically indistinguishable from one to which Congress has attached a far lesser penalty, see Staples, 511 U.S. at 614–15, 114 S.Ct. 1793. Finally, the majority's critique, see ante at 516 (Brown, J., majority op.), that this approach goes beyond what the court has previously “proposed” is meaningless given that, absent en banc review, three-judge panels acting as the court have been bound by Harris. See LaShawn A. v. Barry, 87 F.3d 1389, 1395 (D.C.Cir.1996) (en banc).

Accordingly, I would vacate the judgment of conviction of Burwell under 18 U.S.C. § 924(c)(1)(B)(ii), and I respectfully dissent.

KAVANAUGH, Circuit Judge, with whom Circuit Judge TATEL joins, dissenting:

The presumption of mens rea embodies deeply rooted principles of law and justice that the Supreme Court has emphasized time and again. The presumption of mens rea is no mere technicality, but rather implicates “fundamental and far-reaching” issues, as this case well illustrates. Cf. Morissette v. United States, 342 U.S. 246, 247, 72 S.Ct. 240, 96 L.Ed. 288 (1952).

Under Section 924(c) of Title 18, a person who commits a robbery while carrying a semi-automatic gun faces a mandatory minimum sentence of 10 years. A person who commits a robbery while carrying an automatic gun is guilty of a more serious offense and faces a mandatory minimum sentence of 30 years. Congress specifically determined that carrying an automatic rather than semi-automatic gun during a robbery warrants an extra 20 years of mandatory imprisonment.

The majority opinion holds that a person who committed a robbery while carrying an automatic gun—but who genuinely thought the gun was semi-automatic—is still subject to the 30–year mandatory minimum sentence. The majority opinion thus gives an extra 20 years of mandatory imprisonment to a criminal defendant based on a fact the defendant did not know.

In my view, that extraordinary result contravenes the traditional presumption of mens rea long applied by the Supreme Court. Like many federal criminal statutes, this Section 924(c) offense contains no express mens rea requirement. But the presumption of mens rea means that, unless Congress plainly indicates otherwise, the Government must prove the defendant's mens rea for each element of the offense. And the Supreme Court has recently and unanimously ruled that the automatic character of the gun is an element of the Section 924(c) offense. See United States v. O'Brien, ––– U.S. ––––, 130 S.Ct. 2169, 176 L.Ed.2d 979 (2010). Therefore, in this case, the Government should have been required to prove that Burwell knew the gun was automatic in order to convict him under Section 924(c).

Several factors strongly reinforce the presumption of mens rea here. The Supreme Court has emphasized the particular importance of the presumption when penalties are high—a characterization the Court has applied to statutory maximum sentences of one year's imprisonment. Here, the punishment is dramatically more severe than that—20 extra years of mandatory prison time. Under the Supreme Court's precedents, that heavy sanction strongly reinforces the presumption of mens rea. Moreover, the Supreme Court has already applied the presumption to the automatic character of a gun. In Staples v. United States, 511 U.S. 600, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994), the Court addressed a statute that prohibited possession of an unregistered automatic gun and imposed a maximum sentence of 10 years' imprisonment. The statute was silent about mens rea. The Court held that the Government had to prove that the defendant knew the automatic character of the weapon. There is no good reason for a different result in this case: Here, too, the Government should have to prove that the defendant knew the automatic character of the weapon.

The majority opinion sidesteps the presumption of mens rea by treating the automatic character of the gun as if it's a sentencing factor, not an element of the Section 924(c) offense. But in O'Brien, the Supreme Court held that the gun's automatic character is an element of the Section 924(c) offense, not a sentencing factor. 130 S.Ct. 2169. The O'Brien decision matters here because the Supreme Court has repeatedly stated that the presumption of mens rea applies to elements of the offense, albeit not to sentencing factors. See Staples, 511 U.S. 600, 114 S.Ct. 1793 (element of the offense case); Dean v. United States, 556 U.S. 568, 129 S.Ct. 1849, 173 L.Ed.2d 785 (2009) (sentencing factor case). Because the automatic character of the gun is an element of the Section 924(c) offense, the presumption of mens rea applies in this case.

The majority opinion alternatively concludes that the presumption of mens rea applies to some elements of the offense, but not to others. In particular, the majority opinion states that the presumption of mens rea applies only when necessary to avoid criminalizing apparently innocent conduct—that is, when the conduct would be innocent if the facts were as the defendant believed. But the Supreme Court has never limited the presumption of mens rea in that fashion. The presumption of mens rea applies to each element of the offense. The presumption applies both when necessary to avoid criminalizing apparently innocent conduct (when the defendant would be innocent if the facts were as the defendant believed) and when necessary to avoid convicting the defendant of a more serious offense for apparently less serious criminal conduct (that is, when the defendant would receive a less serious criminal sanction if the facts were as the defendant believed).

In trying to cabin the presumption of mens rea so that it applies only when necessary to avoid criminalizing apparently innocent conduct, the majority opinion resurrects the Government's argument in the recent Flores–Figueroa v. United States case. But the Government's submission garnered zero votes in the Supreme Court. See556 U.S. 646, 129 S.Ct. 1886, 173 L.Ed.2d 853 (2009). With good reason. It would be illogical in the extreme to apply the presumption of mens rea to an element of the offense that would, say, increase the defendant's punishment from no prison time to a term of 2 years in prison, but not to apply the presumption of mens rea to an element of the offense that would aggravate the defendant's crime and increase the punishment from 10 years to 30 years. As Professor LaFave has crisply stated, such an approach would be “unsound, and has no place in a rational system of substantive criminal law.” Wayne R. LaFave, Criminal Law 305 (5th ed.2010).

In my view, the majority opinion is seriously mistaken because it does not properly account for the twin lines of Supreme Court precedent that dictate the result here: The presumption of mens rea applies to each element of the offense, and the automatic character of the gun is an element of the Section 924(c) offense. Twenty extra years of mandatory imprisonment hangs in the balance. I respectfully but emphatically dissent.

I

Criminal liability traditionally requires both a guilty act and a guilty mind, referred to as actus reus and mens rea. This case concerns mens rea. We separately analyze the mens rea requirements of a criminal statute for each element of the offense. See United States v. Bailey, 444 U.S. 394, 406, 100 S.Ct. 624, 62 L.Ed.2d 575 (1980) (“Clear analysis requires that the question of the kind of culpability required to establish the commission of an offense be faced separately with respect to each material element of the crime.”) (quoting Model Penal Code) (brackets omitted).

An element of an offense is said to impose strict liability if it does not require any proof of the defendant's mens rea (i.e., mental state) for that element. Cf. Staples v. United States, 511 U.S. 600, 607 n. 3, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994). Strict liability means that the defendant's ignorance or mistake of fact does not relieve the defendant of criminal liability. For example, consider a statute that criminalizes possession of stolen property. Under strict liability, a defendant would be guilty even if he genuinely did not know that the property he possessed was stolen (or that he possessed it at all). Or consider a statute that criminalizes fishing in prohibited areas. Under strict liability, a defendant would be guilty even if she genuinely did not realize that she was fishing in a prohibited area. Or consider an example the Supreme Court gave in Staples. Congress might make it a crime to operate a car with emissions above a certain limit. Under strict liability, a defendant would be guilty even if his “vehicle's emissions levels, wholly unbeknownst to him, began to exceed legal limits between regular inspection dates.” Id. at 614, 114 S.Ct. 1793.

Strict liability in criminal law is harsh and in serious tension with deeply rooted principles of justice and responsibility. See United States v. O'Mara, 963 F.2d 1288, 1293 (9th Cir.1992) (Kozinski, J., concurring). As a result, strict liability is extremely disfavored in the criminal laws of the United States. One commentator described the objection to strict liability for individual criminal punishment in words that have been often repeated:

The consensus can be summarily stated: to punish conduct without reference to the actor's state of mind is both inefficacious and unjust. It is inefficacious because conduct unaccompanied by an awareness of the factors making it criminal does not mark the actor as one who needs to be subjected to punishment in order to deter him or others from behaving similarly in the future, nor does it single him out as a socially dangerous individual who needs to be incapacitated or reformed. It is unjust because the actor is subjected to the stigma of a criminal conviction without being morally blameworthy. Consequently, on either a preventive or a retributive theory of criminal punishment, the criminal sanction is inappropriate in the absence of mens rea.
Herbert L. Packer, Mens Rea and the Supreme Court, 1962 Sup.Ct. Rev. 107, 109.

The fundamental question we confront here is how to interpret federal criminal statutes that are silent about mens rea. Like many federal criminal statutes, Section 924(c) does not expressly require proof of the defendant's mens rea for this offense. In particular, the statute does not expressly require proof that the defendant knew the weapon was automatic in order for the defendant to be convicted of the automatic weapon offense in Section 924(c).

The ordinary approach to statutory interpretation requires that we adhere to the text of the statute. See, e.g., Milner v. Dep't of the Navy, –––U.S. ––––, 131 S.Ct. 1259, 179 L.Ed.2d 268 (2011); Central Bank of Denver, N.A. v. First Interstate Bank of Denver, N.A., 511 U.S. 164, 114 S.Ct. 1439, 128 L.Ed.2d 119 (1994). At first blush, therefore, one might think courts should hold that there is strict liability whenever a statute is silent about mens rea—and thus that there is strict liability for all the elements of the Section 924(c) offense. Under that scenario, the Government wouldn't have to prove that the defendant in a Section 924(c) case even knew he had a gun—much less knew the gun was an automatic—in order to convict him of this Section 924(c) offense.

But the Supreme Court interprets statutes in light of traditional canons of construction. To take two well-known examples, the Supreme Court has long applied a presumption against extraterritoriality and a presumption against retroactivity. See, e.g., Morrison v. National Australia Bank Ltd., –––U.S. ––––, 130 S.Ct. 2869, 2877–78, 177 L.Ed.2d 535 (2010) (Scalia, J., for the Court) (extraterritoriality); Landgraf v. USI Film Products, 511 U.S. 244, 265–68, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994) (retroactivity); id. at 286–88, 114 S.Ct. 1483 (Scalia, J., concurring in judgment) (retroactivity).

Similarly, the Supreme Court has long applied a presumption of mens rea for criminal statutes. See, e.g., Staples, 511 U.S. at 605–06, 114 S.Ct. 1793 (Thomas, J., for the Court). Like other traditional canons, the presumption of mens rea constitutes a sturdy background principle against which Congress legislates and against which courts interpret statutes.

The presumption of mens rea applied by the Supreme Court stands on a bedrock historical foundation. The American legal tradition, as well as the English common-law tradition on which it was built, has long required proof of the defendant's mens rea as a pre-condition for imposing criminal liability. The “existence of a mens rea is the rule of, rather than the exception to, the principles of Anglo–American criminal jurisprudence.” Id. at 605, 114 S.Ct. 1793 (citation omitted); see 1 Joel Prentiss Bishop, Commentaries on the Criminal Law § 227 (2d ed. 1858) (“the essence of an offence is the wrongful intent, without which it cannot exist”); 4 William Blackstone, Commentaries *21 (1769) (“to constitute a crime against human laws, there must be, first, a vicious will”); 3 Edward Coke, Institutes of the Laws of England 6, 107 (London, E. & R. Brooke 1797) (1644) (“ actus non facit reum, nisi mens sit rea ”; that is, an act does not make a person guilty unless the mind is guilty); Francis Bacon, The Elements of the Common Lawes of England 65 (London, I. More 1630) (1596) (“All crimes have their conception in a corrupt intent”).

The Supreme Court added to that historical foundation when it decided the landmark case of Morissette v. United States, 342 U.S. 246, 72 S.Ct. 240, 96 L.Ed. 288 (1952). Morissette had taken what he thought were abandoned bomb casings from uninhabited land in rural Michigan. Turned out that the casings were government property. The Government charged Morissette with illegally converting (that is, taking and disposing of) government property even though he had thought the casings were abandoned. The relevant statute provided: “Whoever embezzles, steals, purloins, or knowingly converts to his use or the use of another ... any record, voucher, money, or thing of value of the United States” shall be punished by fine or imprisonment. 18 U.S.C. § 641 (1952); see Morissette, 342 U.S. at 248 n. 2, 72 S.Ct. 240. The word “knowingly” in its isolated position suggested that it attached only to the verb “converts,” and not to the fact that the property belonged to someone else (there, the Government).

The Government argued that federal criminal statutes are to be read literally and that Morissette was therefore guilty even if he truly thought that the casings were abandoned. When a statute does not explicitly contain a mens rea requirement, there is none—or so the Government argued.

The Morissette Court held otherwise: Such “adoption of the literal reasoning ... would sweep out of all federal crimes, except when expressly preserved, the ancient requirement of a culpable state of mind”—a result that would be “inconsistent with our philosophy of criminal law.” 342 U.S. at 250, 72 S.Ct. 240. The Court invoked the background presumption of mens rea to conclude that the term “knowingly” also required proof that the defendant knew that the property belonged to someone else.

Justice Jackson authored the Court's opinion in Morissette. Justice Jackson was of course “intimately familiar with the corruption of the criminal process in a totalitarian society.” Packer, Mens Rea and the Supreme Court, 1962 Sup.Ct. Rev. at 119. And in Morissette, Justice Jackson forcefully described the critical link between human liberty and mens rea requirements. He explained that mens rea in criminal law “is as universal and persistent in mature systems of law as belief in freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil.” 342 U.S. at 250, 72 S.Ct. 240. The concept, indeed, “is almost as instinctive as the child's familiar exculpatory ‘But I didn't mean to.’ ” Id. at 251, 72 S.Ct. 240.

Although some might have thought Morissette's case “profoundly insignificant,” Justice Jackson recognized the “fundamental and far-reaching” mens rea issues at stake. Id. at 247, 72 S.Ct. 240. Justice Jackson saw the wolf in sheep's clothing: “The Government asks us by a feat of construction radically to change the weights and balances in the scales of justice. The purpose and obvious effect of doing away with the requirement of a guilty intent is to ease the prosecution's path to conviction, to strip the defendant of such benefit as he derived at common law from innocence of evil purpose, and to circumscribe the freedom heretofore allowed juries.” Id. at 263, 72 S.Ct. 240.

Under the traditional presumption of mens rea as expounded by Morissette, courts presume a mens rea requirement for each element of the offense unless Congress plainly indicates otherwise. See id. at 254 n. 14, 72 S.Ct. 240 (presumption may be overridden by “clear command” from Congress).

II

The Supreme Court's case law since Morissette illustrates the force and breadth of the presumption of mens rea. The Court has applied the presumption to statutes that are silent about mens rea. The Court has likewise applied the presumption to statutes that contain a mens rea requirement for one element but are silent or ambiguous about mens rea for other elements.

A detailed review of those precedents demonstrates that the majority opinion in this case has jumped the rails.

Following Morissette, the Supreme Court again stressed the importance of the presumption of mens rea in United States v. U.S. Gypsum Co., 438 U.S. 422, 98 S.Ct. 2864, 57 L.Ed.2d 854 (1978). The case concerned criminal liability under the Sherman Antitrust Act. The text of the Sherman Act supplied no express mens rea requirement. But the Court relied on Morissette and said that it was “unwilling to construe the Sherman Act as mandating a regime of strict-liability criminal offenses.” Id. at 436, 98 S.Ct. 2864.

As in Morissette, the Court in U.S. Gypsum grounded the presumption in history and tradition. The Court recounted “the familiar proposition that the existence of a mens rea is the rule of, rather than the exception to, the principles of Anglo–American criminal jurisprudence.” Id. (internal quotation marks and brackets omitted). Invoking that background principle, the Court explained that it had “on a number of occasions read a state-of-mind component into an offense even when the statutory definition did not in terms so provide.” Id. at 437, 98 S.Ct. 2864 (emphasis added). “Certainly far more than the simple omission of the appropriate phrase from the statutory definition is necessary to justify dispensing with an intent requirement.” Id. at 438, 98 S.Ct. 2864. Indeed, the Court read Morissette “as establishing, at least with regard to crimes having their origin in the common law, an interpretative presumption that mens rea is required.” Id. at 437, 98 S.Ct. 2864.

The Court stated, moreover, that the presumption was strengthened by the severity of the sanctions under the Sherman Act—a maximum of 3 years' imprisonment. See id. at 442 n. 18, 98 S.Ct. 2864.

In United States v. Bailey, 444 U.S. 394, 100 S.Ct. 624, 62 L.Ed.2d 575 (1980), the Court applied the presumption to a statute punishing prison escape. The law required the Government to prove that the prisoner had escaped from custody. The statute did not contain any express mens rea requirement, raising this question: Could a prisoner be convicted of escape even if he genuinely but mistakenly thought he had permission to leave—on a work detail, for example? The Court embarked on an “element-by-element analysis,” “dissecting” the statute and “assigning a level of culpability to each element.” Id. at 406, 100 S.Ct. 624. Reiterating that the omission of mens rea “does not mean, of course,” that the statute “defines a ‘strict liability’ crime for which punishment can be imposed without proof of any mens rea at all,” the Court required proof of knowledge—proof that “an escapee knew his actions would result in his leaving physical confinement without permission.” Id. at 406 n. 6, 408, 100 S.Ct. 624. In other words, a defendant who genuinely but mistakenly thought he had permission to leave could not be convicted under the statute.

In Posters ‘N’ Things, Ltd. v. United States, 511 U.S. 513, 114 S.Ct. 1747, 128 L.Ed.2d 539 (1994), the statute provided: “It is unlawful for any person ... to make use of the services of the Postal Service or other interstate conveyance as part of a scheme to sell drug paraphernalia....” 21 U.S.C. § 857(a) (1988); see511 U.S. at 516, 114 S.Ct. 1747. The statute was silent as to mens rea. So the question arose: Did the Government have to prove that the defendant knew the possessed materials—a scale and pipe, for example—were drug paraphernalia likely to be used with illegal drugs? The Court said yes. “Certainly far more than the simple omission of the appropriate phrase from the statutory definition is necessary to justify dispensing with an intent requirement.” 511 U.S. at 522, 114 S.Ct. 1747 (citation omitted). Applying the presumption of mens rea, the Court held that the statute “requires the Government to prove that the defendant knowingly made use of an interstate conveyance as part of a scheme to sell items that he knew were likely to be used with illegal drugs.” Id. at 524, 114 S.Ct. 1747 (emphases added).

The Court's decision that same Term in Staples v. United States, 511 U.S. 600, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994), is particularly instructive here. The statute at issue in Staples provided only: “It shall be unlawful for any person ... to receive or possess a firearm which is not registered to him in the National Firearms Registration and Transfer Record....” 26 U.S.C. § 5861 (1994); see511 U.S. at 605, 114 S.Ct. 1793. The statutory definition of “firearm” included automatic weapons, but not semiautomatic weapons. See511 U.S. at 602–04, 615, 114 S.Ct. 1793. The statute contained no express mens rea requirement.

The question in Staples—very similar to the question in our case—was whether the Government had to prove that the defendant knew the gun was an automatic. The Court said yes. The Court cited “the background rule of the common law favoring mens rea ” and noted that the rule applies to statutory criminal offenses. Id. at 605–06, 619, 114 S.Ct. 1793. According to the Court, “some indication of congressional intent” to dispense with mens rea is necessary before a court will find strict liability. Id. at 606, 114 S.Ct. 1793. The Court stated: “Silence does not suggest that Congress dispensed with mens rea ” for the automatic weapon element. Id. at 619, 114 S.Ct. 1793. The Court also pointed out that the sentence involved—a statutory maximum of 10 years' imprisonment—further supported requiring proof of mens rea. See id. at 616–19, 114 S.Ct. 1793. As for what level of mens rea was required, “knowledge” of “the facts that make his conduct fit the definition of the offense” would suffice “to establish mens rea.” Id. at 608 n. 3, 114 S.Ct. 1793. Therefore, “to obtain a conviction, the Government should have been required to prove that” the defendant “knew of the features of his AR–15 that brought it within the scope of the Act”—that is, knew that the gun was an automatic. Id. at 619, 114 S.Ct. 1793.

In Carter v. United States, 530 U.S. 255, 120 S.Ct. 2159, 147 L.Ed.2d 203 (2000), the law in question punished taking bank property either by force and violence or by intimidation. See id. at 259, 120 S.Ct. 2159. The relevant subsection of the statute had no express mens rea requirement. Of course, it's hard to imagine unknowingly taking property by force and violence or by intimidation. But even so, the Court still required proof of the defendant's knowledge because the statute “should not be interpreted to apply to the hypothetical person who engages in forceful taking of money while sleepwalking (innocent, if aberrant activity).” Id. at 269, 120 S.Ct. 2159.

In its long line of mens rea precedents, the Court has applied the presumption of mens rea not just to statutes that are silent about mens rea (such as those just discussed), but also to statutes that contain a mens rea requirement for one element but are silent or ambiguous about mens rea for other elements.

In Liparota v. United States, 471 U.S. 419, 105 S.Ct. 2084, 85 L.Ed.2d 434 (1985), for example, the Court considered a statute providing that “whoever knowingly uses, transfers, acquires, alters, or possesses coupons or authorization cards in any manner not authorized by this chapter or the regulations” shall be fined or imprisoned. 7 U.S.C. § 2024(b)(1) (1982); see471 U.S. at 420 n. 1, 105 S.Ct. 2084. The Court had to decide how far the term “knowingly” traveled down the sentence. Did the defendant need to know that his use of food stamps was unauthorized? The Court stated that the statutory text provided “little guidance” and that either interpretation “would accord with ordinary usage.” 471 U.S. at 424, 105 S.Ct. 2084. Based on the presumption of mens rea, the Court concluded: “Absent indication of contrary purpose in the language or legislative history of the statute, we believe that § 2024(b)(1) requires a showing that the defendant knew his conduct to be unauthorized by statute or regulations.” Id. at 425, 105 S.Ct. 2084.

In United States v. X–Citement Video, Inc., 513 U.S. 64, 115 S.Ct. 464, 130 L.Ed.2d 372 (1994), the Court followed a similar path. The Court there interpreted a statute targeting distribution of child pornography. The Court had to decide whether the statutory mens rea “knowingly” applied not just to the transport of a pornographic video or photo but also to the fact that a girl or boy under age 18 appeared in the video or photo. See id. at 68, 115 S.Ct. 464. As the Court acknowledged, the “most natural grammatical reading” of the statute suggested that the mens rea requirement did not apply to the fact that a girl or boy under 18 was shown in the video or photo. Id. But the Court rejected that plain-text reading as inconsistent with the presumption of mens rea. The Court instead required proof that the defendant knew that a girl or boy under 18 was depicted. See id. at 70, 78, 115 S.Ct. 464.

The Court's recent decision in Flores–Figueroa v. United States, 556 U.S. 646, 129 S.Ct. 1886, 173 L.Ed.2d 853 (2009), adhered to much the same approach. The statute at issue in Flores–Figueroa applied to a person who—while committing a listed predicate crime—“knowingly transfers, possesses, or uses, without lawful authority, a means of identification of another person.” 18 U.S.C. § 1028A; see129 S.Ct. at 1888–89. The Court decided that “knowingly” applied not just to the fact of the defendant's possession or use of the fake identification card but also to the fact that the identity on the card was “of another person.” The Court reached its decision by relying in part on the presumption of mens rea, which in that case meant a strong default rule of statutory interpretation that applies when there is an express mens rea as to one element of the offense: “The manner in which the courts ordinarily interpret criminal statutes is fully consistent with this ordinary English usage. That is to say courts ordinarily read a phrase in a criminal statute that introduces the elements of a crime with the word ‘knowingly’ as applying that word to each element.” 129 S.Ct. at 1891. The Court further explained that its decision was consistent with precedents, such as Liparota and X–Citement Video, that had applied the presumption of mens rea to similar statutory schemes. See id. In justifying its conclusion, the Court asked rhetorically: “Would we apply a statute that makes it unlawful ‘ knowingly to possess drugs' to a person who steals a passenger's bag without knowing that the bag has drugs inside?” Id. at 1890.

Justice Alito's concurrence in Flores–Figueroa concisely summarized the presumption of mens rea applied by the Court, and added that the presumption was of course not conclusive and could be overcome by context in certain circumstances: “In interpreting a criminal statute such as the one before us, I think it is fair to begin with a general presumption that the specified mens rea applies to all the elements of an offense, but it must be recognized that there are instances in which context may well rebut that presumption.... I therefore concur in the judgment and join the opinion of the Court except insofar as it may be read to adopt an inflexible rule of construction that can rarely be overcome by contextual features pointing to a contrary reading.” Id. at 1895–96 (Alito, J., concurring in part and concurring in judgment).

The Supreme Court's case law demonstrates that the Court has applied the presumption of mens rea consistently, forcefully, and broadly. The presumption applies to statutes that are silent as to mens rea. See Morissette, 342 U.S. 246, 72 S.Ct. 240;U.S. Gypsum, 438 U.S. 422, 98 S.Ct. 2864;Bailey, 444 U.S. 394, 100 S.Ct. 624;Posters ‘N’ Things, 511 U.S. 513, 114 S.Ct. 1747;Staples, 511 U.S. 600, 114 S.Ct. 1793;Carter, 530 U.S. 255, 120 S.Ct. 2159. The presumption also applies to statutes that contain an explicit mens rea requirement for one element but are silent or ambiguous about mens rea for other elements. See Morissette, 342 U.S. 246, 72 S.Ct. 240;Liparota, 471 U.S. 419, 105 S.Ct. 2084;X–Citement Video, 513 U.S. 64, 115 S.Ct. 464;Flores–Figueroa, 556 U.S. 646, 129 S.Ct. 1886. And whether the statute is completely silent as to mens rea, or only partially silent, the presumption applies to each element of the offense. Recall that in Posters ‘N’ Things, for example, the statute said simply: “It is unlawful for any person ... to make use of the services of the Postal Service or other interstate conveyance as part of a scheme to sell drug paraphernalia....” 21 U.S.C. § 857(a) (1988); see511 U.S. at 516, 114 S.Ct. 1747. The statute contained no express mens rea requirement. The Court held that the presumption of mens rea applied to each element of the offense, saying the statute “requires the Government to prove that the defendant knowingly made use of an interstate conveyance as part of a scheme to sell items that he knew were likely to be used with illegal drugs.” 511 U.S. at 524, 114 S.Ct. 1747 (emphases added).

To sum up so far, the Supreme Court has established and applied a rule of statutory interpretation for federal crimes: A requirement of mens rea applies to each element of the offense unless Congress has plainly indicated otherwise.

III

We thus far have established that the presumption of mens rea applies to each element of the offense. To use the presumption of mens rea correctly, we must keep in mind a critical distinction: The presumption of mens rea applies to elements of the offense, but not to sentencing factors.

An element of the offense is a “fact necessary to constitute the crime.” Almendarez–Torres v. United States, 523 U.S. 224, 240, 118 S.Ct. 1219, 140 L.Ed.2d 350 (1998) (citation omitted). “Characteristics of the offense itself are traditionally treated as elements....” United States v. O'Brien, –––U.S. ––––, 130 S.Ct. 2169, 2176, 176 L.Ed.2d 979 (2010).

By contrast, a sentencing factor is a fact that is not necessary to define the crime but that typically is used by a sentencing court to increase punishment. See Almendarez–Torres, 523 U.S. at 228, 118 S.Ct. 1219. “Sentencing factors traditionally involve characteristics of the offender—such as recidivism, cooperation with law enforcement, or acceptance of responsibility.” O'Brien, 130 S.Ct. at 2176.

As the Supreme Court has recognized: “Much turns on the determination that a fact is an element of an offense rather than a sentencing” factor. Jones v. United States, 526 U.S. 227, 232, 119 S.Ct. 1215, 143 L.Ed.2d 311 (1999); see also McMillan v. Pennsylvania, 477 U.S. 79, 84–86, 106 S.Ct. 2411, 91 L.Ed.2d 67 (1986). For example, under the Fifth and Sixth Amendments, an element of the offense “must be charged in the indictment, submitted to a jury, and proven by the Government beyond a reasonable doubt.” Jones, 526 U.S. at 232, 119 S.Ct. 1215. By contrast, a sentencing factor need not be charged in the indictment, submitted to a jury, or proved by the Government beyond a reasonable doubt. See id.

As to the presumption of mens rea, the Supreme Court has long stated that an element of the offense triggers the presumption of mens rea. See, e.g., Staples v. United States, 511 U.S. 600, 606–07 & n. 3, 619, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994). By contrast, as the Supreme Court indicated in Dean v. United States, a sentencing factor does not trigger the presumption of mens rea. 556 U.S. 568, 129 S.Ct. 1849, 173 L.Ed.2d 785 (2009); see also O'Brien, 130 S.Ct. at 2179, 2180 (expressly describing Dean as a case about “sentencing factors”).

In Dean, the Court resolved a circuit split that had arisen over whether the presumption of mens rea applies to sentencing factors. The Court concluded that the presumption does not apply to sentencing factors. Justice Stevens vigorously dissented and lamented the distinction the Court had drawn between elements of the offense and sentencing factors for the presumption of mens rea: “Although mandatory minimum sentencing provisions are of too recent genesis to have any common-law pedigree, there is no sensible reason for treating them differently from offense elements for purposes of the presumption of mens rea. Sentencing provisions of this type have substantially the same effect on a defendant's liberty as aggravated offense provisions.... The common-law tradition of requiring proof of mens rea to establish criminal culpability should thus apply equally to such sentencing factors.” 129 S.Ct. at 1858 (Stevens, J., dissenting) (citation omitted).

The distinction that the Court has drawn between elements of the offense and sentencing factors derives in part from the Court's traditional view of sentencing as a more flexible, open-ended proceeding that takes account of a wide variety of circumstances. Cf. Pepper v. United States, ––– U.S. ––––, 131 S.Ct. 1229, 1235, 179 L.Ed.2d 196 (2011) (“sentencing judges exercise a wide discretion in the types of evidence they may consider when imposing sentence”) (internal quotation marks omitted); Williams v. New York, 337 U.S. 241, 246, 69 S.Ct. 1079, 93 L.Ed. 1337 (1949) (“courts in this country and in England practiced a policy under which a sentencing judge could exercise a wide discretion in the sources and types of evidence used”).

To be sure, as indicated by Justice Stevens's dissent in Dean, some Justices continue to voice weighty arguments that the protections attached to elements of the offense—including Fifth and Sixth Amendment rights, as well as the presumption of mens rea—should also attach to sentencing factors. According to those Justices, there is little if any difference in certain modern criminal statutes between the facts labeled as elements and the facts labeled as sentencing factors. And in the view of those Justices, the Court has allowed legislatures to too broadly deploy the “sentencing factor” label and thereby evade the protections, including the presumption of mens rea, that attach to elements of the offense. See O'Brien, 130 S.Ct. at 2181–83 (Stevens, J., concurring); id. at 2183–84 (Thomas, J., concurring in judgment); Dean, 129 S.Ct. at 1857–60 (Stevens, J., dissenting); Harris v. United States, 536 U.S. 545, 572–83, 122 S.Ct. 2406, 153 L.Ed.2d 524 (2002) (Thomas, J., dissenting); Monge v. California, 524 U.S. 721, 737–41, 118 S.Ct. 2246, 141 L.Ed.2d 615 (1998) (Scalia, J., dissenting); McMillan, 477 U.S. at 93–94, 106 S.Ct. 2411 (Marshall, J., dissenting); id. at 95–104, 106 S.Ct. 2411 (Stevens, J., dissenting).

But that continuing argument about adding protections for sentencing factors is not relevant here. That's because, as I will explain now, this case indisputably concerns an element of the offense, which in turn means that the presumption of mens rea indisputably applies.

IV

As the preceding discussion reveals, deciding whether the presumption of mens rea applies in this case turns on the following question: Is the automatic character of the weapon an element of the Section 924(c) offense, or is it a sentencing factor?

Determining whether a particular fact is an element of the offense or a sentencing factor can be difficult in some cases. In this case, however, the question is not difficult because the Supreme Court has already answered it.

In its 2010 decision in United States v. O'Brien, the Court unanimously concluded that a firearm's automatic character is an element of the Section 924(c) offense as written and intended by Congress—and not a sentencing factor. ––– U.S. ––––, 130 S.Ct. 2169, 176 L.Ed.2d 979 (2010). The question there concerned whether the Government had to prove the fact that the weapon was automatic to a jury beyond a reasonable doubt. See id. at 2173–74. The O'Brien Court said yes, because the automatic character of the weapon is an element of the Section 924(c) offense. The Court reasoned that the nature of the weapon was a characteristic of the offense, and characteristics of the offense “are traditionally treated as elements.” Id. at 2176. In addition, the Court explained that the “drastic” increase in the minimum sentence when the firearm is automatic “strongly suggests a separate substantive crime,” not merely a sentencing factor. Id. at 2177.

The Court in O'Brien thus concluded that the automatic character of the gun is an element of the Section 924(c) offense.

Because the automatic character of the gun is an element of the offense, and because the presumption of mens rea applies to each element of the offense, the presumption of mens rea applies here. We therefore must presume that the Section 924(c) offense requires the Government to prove that the defendant knew his gun was automatic.

V

The majority opinion avoids the presumption of mens rea by treating the automatic character of the gun as if it's a sentencing factor rather than an element of the offense. That is apparent from the majority opinion's repeated invocation of the Supreme Court's decision in Dean v. United States, a sentencing factor case. 556 U.S. 568, 129 S.Ct. 1849, 173 L.Ed.2d 785 (2009).

The fundamental problem for the majority opinion is that the decision in Dean addressed a sentencing factor—namely, discharge of the firearm—and not an element of the offense. The Supreme Court itself has referred to Dean as a case about “sentencing factors.” United States v. O'Brien, –––U.S. ––––, 130 S.Ct. 2169, 2179, 2180, 176 L.Ed.2d 979 (2010). In the predecessor case to Dean, the Supreme Court likewise held that discharge of the firearm during the crime was a “sentencing factor,” not an element of the offense. Harris v. United States, 536 U.S. 545, 552, 556, 122 S.Ct. 2406, 153 L.Ed.2d 524 (2002).

Because the Supreme Court concluded that discharge of the firearm was a sentencing factor and not an element of the offense, the presumption of mens rea did not apply in Dean. The Dean Court therefore determined that the Government need not prove that the defendant intended to discharge the firearm.

By contrast, as the Supreme Court held in O'Brien, the automatic character of the gun is an element of the offense. As a result, the presumption of mens rea applies in this case.

To rely on Dean here—as the majority opinion does relentlessly—is to miss the boat on the crucial distinction between sentencing factors and elements of the offense for purposes of the presumption of mens rea.

For the same reason, the majority opinion's reliance on this Court's 1992 decision in United States v. Harris, 959 F.2d 246 (D.C.Cir.1992), is misplaced. Harris said that the automatic character of a gun was a “sentence enhancement”; therefore, the Court did not require proof that the defendant knew the gun was automatic. Id. at 259. In United States v. Brown, moreover,we specifically described Harris as a sentencing factor case: “The government also relies on United States v. Harris, where we analyzed a part of § 924(c)(1) that imposes a 30–year minimum when the weapon used is a machine gun. Assuming that this created a sentencing factor, we found that the government didn't need to prove that the defendant knew the precise nature of the weapon he used....” 449 F.3d 154, 157–58 (D.C.Cir.2006) (citations omitted).

The determination made in Harris and reiterated in Brown—that the automatic character of the gun is a sentencing factor—has been undermined by the Supreme Court's decision in O'Brien. In O'Brien, to repeat, the Supreme Court expressly ruled that the automatic character of the gun is an element of the Section 924(c) offense, not a sentencing factor. See130 S.Ct. 2169. The Supreme Court's decision in O'Brien thus knocked out the fundamental underpinnings of this Court's decision in Harris.

Moreover, in concluding that the Government did not need to prove that the defendant knew the automatic character of the gun, Harris reasoned that “there does not seem to be a significant difference in mens rea between a defendant who commits a drug crime using a pistol and one who commits the same crime using a machine gun; the act is different, but the mental state is equally blameworthy.” 959 F.2d at 259. But in O'Brien, the Supreme Court directly rejected that line of reasoning, stating that the dramatically enhanced sanctions in Section 924(c) for carrying an automatic weapon reflect “the moral depravity in choosing the weapon.” 130 S.Ct. at 2178.O'Brien thus rejected not only the sentencing factor label that Harris used, but also the reasoning that Harris employed. For that reason as well, Harris has been undermined by the Supreme Court on the question whether the presumption of mens rea applies to the automatic character of the gun in a Section 924(c) case.

VI

The majority opinion alternatively asserts that the presumption of mens rea applies only to some elements of the offense—namely, only to those elements that are “essential.” See Maj. Op. at 503, 505–07. Under the majority opinion's approach, the presumption applies only when proof of mens rea is necessary to avoid criminalizing apparently innocent conduct—that is, when the defendant would be innocent if the facts were as the defendant believed. Under the majority opinion's theory, the presumption does not apply when proof of mens rea is necessary to avoid convicting a defendant of a more serious offense for apparently less serious criminal conduct—that is, when the defendant would receive a less serious criminal sanction if the facts were as he believed.

But where does the majority opinion find that limitation on the presumption of mens rea? The Supreme Court has never drawn such a distinction when employing the presumption of mens rea.

To be sure, the Supreme Court has said that the presumption of mens rea is important when the defendant otherwise may have been innocent of any wrongdoing. See, e.g., Staples v. United States, 511 U.S. 600, 610, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994) (Court has taken “particular care” to avoid criminalizing apparently innocent conduct). But the Court has never cabined the presumption of mens rea to those circumstances. As this Court said in United States v. Villanueva–Sotelo—a decision approved by the Supreme Court in Flores–Figueroa v. United States—the Supreme Court has found it “particularly appropriate to extend a mens rea requirement when failure to do so would result in a statute criminalizing nonculpable conduct,” but “the Court has never held that avoiding such a result is the only reason to do so.” 515 F.3d 1234, 1242 (D.C.Cir.2008) (internal quotation marks omitted). The majority opinion here is thus trying to confine the presumption of mens rea in a way that the Supreme Court has never done.

Moreover, it would be incoherent to limit the presumption of mens rea to only those cases where it's necessary to avoid criminalizing what the defendant thought was innocent conduct. The key trigger for the presumption of mens rea is whether the fact at issue is an element of the offense. If a fact is an element of the offense and not a sentencing factor, the presumption applies. And the presumption applies both when necessary to avoid criminalizing apparently innocent conduct and when necessary to avoid convicting of a more serious offense for apparently less serious criminal conduct. As Professor LaFave has explained, rules of mens rea apply both to a defendant who is unaware of the facts that make his conduct criminal and to a defendant who is “unaware of the magnitude of the wrong he is doing.” Wayne R. LaFave, Criminal Law 304 (5th ed.2010). The idea that “the mistake by the defendant may be disregarded because of the fact that he actually intended to do some legal or moral wrong” is—in Professor LaFave's words—“unsound, and has no place in a rational system of substantive criminal law.” Id. at 304–05; see alsoGlanville Williams, Criminal Law: The General Part 185–99 (2d ed.1961).

Taking a step back: What sense would it make to have a presumption of mens rea for an element of the offense that increases the defendant's mandatory minimum punishment from no prison time to a term of 2 years' imprisonment, for example, but not to have a presumption of mens rea for an element of the offense that aggravates the defendant's offense and elevates the defendant's mandatory minimum punishment from 10 years to 30 years? The answer is that it would make no sense at all: “After all, a comparable degree of inequity exists in (1) punishing a person who, but for the strict liability application to the element, would have received zero punishment (the ‘innocent’ case protected by X–Citement Video ) and (2) punishing with more years of imprisonment a person who, but for the strict liability application to the element, would still have received substantial punishment.” Leonid Traps, “ Knowingly” Ignorant: Mens Rea Distribution in Federal Criminal Law After Flores–Figueroa, 112 Colum. L.Rev.. 628, 661 (2012).

The majority opinion retorts that we are not confronted with “an altar boy who made an innocent mistake.” Maj. Op. at 507 (citation and brackets omitted). But the fact that the defendant is a “bad person” who has done “bad things” does not justify dispensing with the presumption of mens rea in this fashion and imposing 20 years of additional mandatory prison time. An example helps illustrate that point: If an altar boy steals the collection bag, he is guilty of larceny. If the bag also happens to contain a stash of cocaine sewn into the lining, but the altar boy did not know about the hidden drugs, he should not be guilty of drug possession. In other words, the fact that he is guilty of larceny doesn't justify rendering him guilty of possessing drugs, at least absent some plain indication of legislative intent to eliminate a mens rea requirement. Yet the majority opinion's approach here would mean that the altar boy in that hypothetical scenario is indeed guilty of both larceny and drug possession.

When the facts as the defendant believed them would have warranted conviction of a lesser offense and called for a lesser punishment, no legitimate purpose of criminal law—whether it be retribution, deterrence, or rehabilitation—is served by convicting him of an aggravated offense and imposing a more severe punishment.

The Supreme Court's recent decision in Flores–Figueroa underscores that the presumption of mens rea applies not just when the presumption is necessary to avoid criminalizing apparently innocent conduct, but also when the presumption is necessary to avoid convicting a defendant of a more serious offense for apparently less serious conduct.

Recall that the statute in Flores–Figueroa punished someone who “knowingly transfers, possesses, or uses, without lawful authority, a means of identification of another person” while committing an enumerated predicate crime. 18 U.S.C. § 1028A; see129 S.Ct. 1886, 1888–89 (2009). The question was whether the Government had to prove that the defendant knew the identification card contained the identity of another actual person (as opposed to thinking that the identification card merely contained the identity of a made-up person). Because the statute applied only to those who had committed a predicate crime and who had illegally used a fake ID card, proof that the defendant knew the identification card contained the identity of another actual person was not necessary to avoid criminalizing apparently innocent conduct. The Government argued that it therefore should not have to prove that the defendant knew the card contained the identity of another actual person.

No Justice on the Court accepted the Government's argument that the presumption of mens rea applies only when necessary to avoid criminalizing apparentlyinnocent conduct. Indeed, the Court specifically rejected the Government's argument that “the defendant's necessary guilt of a predicate crime and the defendant's necessary knowledge that he has acted ‘without lawful authority,’ make it reasonable” to “read the statute's language as dispensing with the knowledge requirement.” Flores–Figueroa, 129 S.Ct. at 1893. The Court ruled that the Government still must prove the defendant knew the card contained the identity of another person, even though the defendant was already committing two other crimes—the predicate crime and the use of a fake ID card. The Court relied in part on the presumption of mens rea, described by the Court as a strong default rule of statutory interpretation: “The manner in which the courts ordinarily interpret criminal statutes is fully consistent with this ordinary English usage. That is to say courts ordinarily read a phrase in a criminal statute that introduces the elements of a crime with the word ‘knowingly’ as applying that word to each element.” Id. at 1891.

As one respected commentary has explained: “Earlier cases had indicated that when a federal statute criminalizes otherwise innocent conduct, courts should interpret the mens rea requirement in the statute broadly. Some lower courts had taken this line of decisions to mean that when a federal criminal statute criminalizes behavior that would not be innocent in the absence of that statute, the mens rea requirement should be read to apply to fewer elements of the crime. In Flores–Figueroa, the Court corrected this misreading.” The Supreme Court, 2008 Term—Leading Cases, 123 Harv. L.Rev.. 153, 317 (2009).

In trying to narrow the presumption of mens rea so that it applies only when necessary to avoid criminalizing apparently innocent conduct, the majority opinion echoes the Government's failed submission in Flores–Figueroa. It is instructive to compare the Government's position in Flores–Figueroa with the majority opinion's analysis here.

In Flores–Figueroa, the Government tried to distinguish Morissette, U.S. Gypsum, Liparota, Staples, and X–Citement Video on the ground that those cases involved statutes that “criminalize conduct that might reasonably be viewed as innocent or presumptively lawful in nature.” Brief for the United States at 42–43, Flores–Figueroa, 556 U.S. 646, 129 S.Ct. 1886, 173 L.Ed.2d 853 (No. 08–108). The Government further contended that the Supreme Court's mens rea precedents “should not be understood apart from the Court's primary stated concern of avoiding criminalization of otherwise non-culpable conduct.” Id. at 18 (citation and brackets omitted). And the Government claimed that “the Court has interpreted the scope of a statutorily prescribed mens rea requirement against a background presumption that statutes should not generally be read as criminalizing apparently innocent conduct.” Id. at 40 (internal quotation marks omitted). The Supreme Court rejected those arguments wholesale.

Like the Government in Flores–Figueroa, the majority opinion here tries to distinguishthe relevant Supreme Court precedents on the ground that those cases “involved” statutes “that criminalized otherwise lawful behavior.” Maj. Op. at 506. The majority opinion concludes that the precedents' “concerns animating the presumption in favor of mens rea ”—the “danger of ensnaring an altar boy who made an innocent mistake”—“simply are not present here.” Maj. Op. at 507 (internal quotation marks and brackets omitted). And the majority opinion says that the “Court developed the presumption in favor of mens rea for one particular reason: to avoid criminalizing otherwise lawful conduct.” Maj. Op. at 505.

The majority opinion is thus rehashing the same theory that the Government unsuccessfully advanced to the Supreme Court in Flores–Figueroa.

Finally, it bears mention that even the majority opinion ultimately backs off its “apparently innocent conduct” limitation to the presumption of mens rea. Under the majority opinion's theory, there should be no mens rea requirement for any of the elements of the Section 924(c) offense, including the carrying of the gun. After all, Section 924(c) applies only to someone who has committed a separate violent or drug trafficking crime. Requiring proof that the defendant (i) knew he was carrying an object and (ii) further knew that the object he was carrying was a gun is therefore not necessary to avoid criminalizing apparently innocent conduct. But realizing the harsh absurdities that could result from that conclusion—namely, that a 30–year mandatory minimum sentence could be imposed based on the presence of a gun that the defendant did not even know was there—the majority opinion retreats. The majority opinion concedes that Section 924(c) must be interpreted to require knowledge that the defendant was carrying a gun and knowledge that the object carried was a gun. See Maj. Op. at 503, 507. What that means is that the majority opinion itself actually ends up applying the presumption of mens rea in circumstances where it is not necessary to avoid criminalizing apparently innocent conduct. Given that concession, what is left of the majority opinion's attempt to limit the presumption of mens rea to circumstances where the presumption is necessary to avoid criminalizing apparently innocent conduct? Not much. The majority opinion fashions a limitation on the presumption of mens rea that the Supreme Court has never applied and that was rejected in Flores–Figueroa, that we likewise rejected in Villanueva–Sotelo, and that makes little sense under fundamental criminal law principles. And then the majority opinion—apparently recognizing that its theory generates harsh absurdities—carves out exceptions in an ad hoc manner that leaves its attempted limitation on the presumption of mens rea in shambles.

In sum, under Supreme Court precedent, the presumption of mens rea applies to each element of the offense, not just when necessary to avoid criminalizing apparently innocent conduct. Therefore, the presumption of mens rea applies to the automatic character of the weapon in Section 924(c) cases.

VII

By now, we have cleared a lot of brush in determining that the presumption of mens rea applies to the automatic character of the weapon in Section 924(c) cases. Of course, the presumption of mens rea is a presumption; it thus may be overcome by a plainly contrary congressional intent, as revealed in the statutory text or context. Here, the presumption of mens rea is not overcome.

To begin with, three aspects of Section 924(c) strongly reinforce the presumption of mens rea: the severity of the additional sentence for carrying an automatic gun; the difficulty of distinguishing an automatic gun from a semiautomatic gun; and the inconsistency that would otherwise be created with the Supreme Court's decision in Staples, which required proof that the defendant knew the gun was automatic in order to convict him of possessing an unregistered automatic weapon.

First, the severe penalties at issue here support requiring proof of the defendant's mens rea. The Supreme Court has repeatedly stated that “the penalty imposed under a statute has been a significant consideration in determining whether the statute should be construed as dispensing with mens rea.” Staples v. United States, 511 U.S. 600, 616, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994). In X–Citement Video, the Court said that the “concern with harsh penalties looms equally large respecting § 2252.” United States v. X–Citement Video, Inc., 513 U.S. 64, 72, 115 S.Ct. 464, 130 L.Ed.2d 372 (1994). In U.S. Gypsum, the Court said: “The severity of these sanctions provides further support for our conclusion that the Sherman Act should not be construed as creating strict-liability crimes.” United States v. U.S. Gypsum Co., 438 U.S. 422, 442 n. 18, 98 S.Ct. 2864, 57 L.Ed.2d 854 (1978). And in Morissette, the Court found larceny not to be a strict liability crime in part because “the penalty is high and, when a sufficient amount is involved, the infamy is that of a felony, which, says Maitland, is ‘as bad a word as you can give to man or thing.’ ” Morissette v. United States, 342 U.S. 246, 260, 72 S.Ct. 240, 96 L.Ed. 288 (1952) (ellipsis omitted).

The “harsh penalties” in Staples and X–Citement Video were statutory maximums of 10 years' imprisonment. 513 U.S. at 72, 115 S.Ct. 464,511 U.S. at 602–03, 114 S.Ct. 1793. The “sever[e]” sanction in U.S. Gypsum was a statutory maximum of 3 years' imprisonment. 438 U.S. at 442 n. 18, 98 S.Ct. 2864. And the “high” penalty in Morissette was a statutory maximum of one year in prison. 342 U.S. at 248 & n. 2, 72 S.Ct. 240. The Supreme Court deemed those penalties sufficiently stringent to support a requirement of mens rea.

In this case, the additional imprisonment that turns on the automatic character of the gun is an extra 20 years; the extra 20 years is a mandatory minimum; and the resulting 30–year mandatory minimum sentence for this Section 924(c) offense must be served consecutively to (that is, in addition to) the sentence for the underlying robbery. The penalty at issue here is thus dramatically more severe than the penalties in those earlier Supreme Court mens rea cases. In the words of the Supreme Court, the extra 20 years of mandatory, consecutive prison time here is an “extreme sentencing increase.” United States v. O'Brien, ––– U.S. ––––, 130 S.Ct. 2169, 2178, 176 L.Ed.2d 979 (2010).

Congress could have decided that carrying either a semi-automatic or an automatic gun during a robbery is equally depraved and equally worthy of a 30–year mandatory sentence. Congress did not do so. It believed that carrying an automatic gun is far more serious and depraved, warranting an extra 20 years of mandatory prison time above that for carrying a semi-automatic gun. Almost a fortiori from the Supreme Court's decisions in Staples, X–Citement Video, U.S. Gypsum, and Morissette, the additional 20 years of mandatory prison time strongly supports requiring proof of mens rea—namely, proof that the defendant knew the automatic character of the gun.

Second, the difficulty of distinguishing an automatic gun from a semi-automatic gun supports requiring proof of mens rea. Automatic and semi-automatic guns may appear externally similar if not identical, as the Staples Court explained. See511 U.S. at 615, 114 S.Ct. 1793. And “virtually any semiautomatic weapon may be converted, either by internal modification or, in some cases, simply by wear and tear,” into an automatic weapon. Id. The Supreme Court thus recognized that it is quite possible for someone who possesses an automatic weapon to “genuinely and reasonably” believe that he possesses “a conventional semi-automatic” weapon. Id. (citation omitted).

Third, requiring proof of mens rea is supported by the Supreme Court's decision in Staples, which applied the presumption of mens rea in very similar circumstances. There, the Court considered a statute that criminalized possession of an unregistered automatic gun. The statute contained no express mens rea requirement. The Supreme Court applied the presumption of mens rea and required the Government to prove that the defendant knew that the gun was automatic.

The decision in Staples raises a straightforward question: If the presumption of mens rea applied in Staples to the gun's automatic character, why shouldn't it also apply here to the gun's automatic character? The majority opinion offers no persuasive answer to that question.

So there are three significant textual and contextual factors that strongly reinforce the presumption of mens rea and support requiring proof that the defendant knew the automatic character of the gun. There is nothing in the statutory text or context that undermines the presumption of mens rea.

It is of course true, as the majority opinion notes, that “Section 924(c)(1)(B)(ii) is silent regarding a mens rea requirement.” Maj. Op. at 511. But as the Supreme Court has explained again and again, “mere omission from the statute of any mention of intent will not be construed as eliminating that element from the crimes denounced.” United States v. Bailey, 444 U.S. 394, 406 n. 6, 100 S.Ct. 624, 62 L.Ed.2d 575 (1980) (quoting Morissette, 342 U.S. at 263, 72 S.Ct. 240) (brackets omitted). To state the obvious: If the presumption of mens rea were overcome by statutory silence, it would not be much of a presumption. But the presumption of mens rea is quite potent. Indeed, the Supreme Court has stated that statutes containing no express mens rea requirement still unambiguously contain a mens rea requirement. You read that correctly. See Staples, 511 U.S. at 619 n. 17, 114 S.Ct. 1793.

It is also true that other subsections in Section 924, as well as other statutes, expressly require a specific mens rea. Because Congress knows how to require mens rea and indeed has required it in other subsections, the majority opinion suggests that the omission of mens rea from the automatic weapon element in Section 924(c) must have been deliberate. But the Supreme Court has repeatedly rejected that approach to mens rea issues. For example, the statute in Morissette punished “[w]hoever embezzles, steals, purloins, or knowingly converts to his use or the use of another” anything owned by the United States. 18 U.S.C. § 641 (1952) (emphasis added); see342 U.S. at 248 n. 2, 72 S.Ct. 240. Despite the explicit knowledge requirement for one element, the Court held that “mere omission” of a mens rea requirement for another element does not eliminate a mens rea requirement for that other element. 342 U.S. at 263, 72 S.Ct. 240. Similarly, in Carter v. United States, the Court faced parallel subsections of a bank robbery statute. While subsection (b) required “a specific ‘intent to steal or purloin,’ ” subsection (a) contained “no explicit mens rea requirement of any kind.” 530 U.S. 255, 267, 120 S.Ct. 2159, 147 L.Ed.2d 203 (2000). But once again, the Court refused to apply strict liability to subsection (a). Instead, relying on “the presumption in favor of scienter,” the Court “read subsection (a) as requiring proof” of the defendant's knowledge. Id. at 268, 120 S.Ct. 2159. And in Posters ‘N’ Things, Ltd. v. United States, the Court interpreted a section of a statute—21 U.S.C. § 857 (1988)—enacted as part of the Anti–Drug Abuse Act of 1986. See511 U.S. 513, 516, 114 S.Ct. 1747, 128 L.Ed.2d 539 (1994). The adjacent section of the statute, enacted in the same Act, imposed an explicit knowledge requirement. See21 U.S.C. § 856(a) (1988). Yet the Court still held that “the fact that Congress did not include the word ‘knowingly’ in the text of § 857” cannot “justif[y] the conclusion that Congress intended to dispense entirely with a scienter requirement.” Posters ‘N’ Things, 511 U.S. at 522, 114 S.Ct. 1747.Cf. Sackett v. EPA, ––– U.S. ––––, 132 S.Ct. 1367, 1373, 182 L.Ed.2d 367 (2012) (“if the express provision of judicial review in one section of a long and complicated statute were alone enough to overcome the APA's presumption of reviewability for all final agency action, it would not be much of a presumption at all”).

In short, the Supreme Court's precedents definitively establish that neither (i) silence on mens rea, nor (ii) the inclusion of a mens rea requirement in another statute, nor (iii) the inclusion of a mens rea requirement in another part of the same statute suffices to defeat the presumption of mens rea.

More broadly, the majority opinion suggests that the presumption of mens rea is overcome here because the purpose of Section 924(c) is to deter violent use of the most dangerous guns. But the goal of every criminal statute is to deter disfavored or dangerous activity. The purpose of deterring criminal activity has not justified dispensing with the presumption of mens rea for elements of the offense. See, e.g., Flores–Figueroa v. United States, 556 U.S. 646, 129 S.Ct. 1886, 1892, 173 L.Ed.2d 853 (2009) (“The question, however, is whether Congress intended to achieve this enhanced protection by permitting conviction of those who do not know the ID they unlawfully use refers to a real person, i.e., those who do not intend to cause this further harm.”); Morissette, 342 U.S. at 259, 72 S.Ct. 240 (“Of course, the purpose of every statute would be ‘obstructed’ by requiring a finding of intent, if we assume that it had a purpose to convict without it.”).

In sum, the presumption of mens rea is not overcome in this statute. Indeed, the relevant textual and contextual considerations strongly support requiring proof of mens rea in this case. Therefore, the Government should have been required to prove the defendant's mens rea—namely, to prove that the defendant knew the gun was automatic.

VIII

In seeking to limit the presumption of mens rea, the Government suggests that the presumption is not workable or practical for statutes like this one. That is wrong. The presumption of mens rea is eminently workable and practical.

First, the presumption of mens rea eliminates the need for difficult statute-by-statute inquiry into whether a particular statute requires proof of the defendant's mens rea. The presumption of mens rea applies to each element of the offense in federal criminal statutes. A stable and consistently applied presumption of mens rea yields greater clarity and predictability for courts, prosecutors, and defendants. It saves resources that otherwise might be wasted in wrangling over whether a particular element warrants the presumption. It means that Congress need not go back and scour all existing statutes to ensure that mens rea was properly addressed. Nor need Congress worry that inadvertent ambiguity about mens rea will produce harsh and unintended results. The background principle is straightforward: Only a deliberately and plainly expressed choice by Congress will override the presumption of mens rea that attaches to elements of the offense. Cf. Morrison v. National Australia Bank Ltd., ––– U.S. ––––, 130 S.Ct. 2869, 2881, 177 L.Ed.2d 535 (2010) (“The results of judicial-speculation-made-law—divining what Congress would have wanted if it had thought of the situation before the court—demonstrate the wisdom of the presumption against extraterritoriality. Rather than guess anew in each case, we apply the presumption in all cases, preserving a stable background against which Congress can legislate with predictable effects.”).

Second, the presumption of mens rea avoids the significant constitutional questions that would arise if a defendant could be severely punished based on a fact the defendant did not know. If we followed the Government's lead and read criminal statutes literally with respect to mens rea, we would have to open up an entire new body of constitutional mens rea law. See Lambert v. California, 355 U.S. 225, 228–30, 78 S.Ct. 240, 2 L.Ed.2d 228 (1957); see also United States v. X–Citement Video, Inc., 513 U.S. 64, 78, 115 S.Ct. 464, 130 L.Ed.2d 372 (1994) (cases “suggest that a statute completely bereft of a scienter requirement as to the age of the performers would raise serious constitutional doubts”); United States v. U.S. Gypsum Co., 438 U.S. 422, 437–38, 98 S.Ct. 2864, 57 L.Ed.2d 854 (1978) (“While strict-liability offenses are not unknown to the criminal law and do not invariably offend constitutional requirements, the limited circumstances in which Congress has created and this Court has recognized such offenses attest to their generally disfavored status.”) (emphasis added) (citations omitted).

Third, the presumption of mens rea carefully balances the competing interests of the prosecution and the defense. The Government suggests that it would be impractical and unfair to the prosecution to require proof of the defendant's mens rea in these circumstances. The Government has advanced such claims many times before. Yet the Supreme Court has repeatedlystated that the presumption of mens rea does not unfairly burden the prosecution. Indeed, in Staples v. United States, the Court rejected the exact same argument from the Government and said that the very mens rea requirement at issue here—knowledge that the gun was automatic—would be easy enough to prove: “The Government contends that ... requiring proof of knowledge would place too heavy a burden on the Government and obstruct the proper functioning of § 5861(d). But knowledge can be inferred from circumstantial evidence, including any external indications signaling the nature of the weapon. And firing a fully automatic weapon would make the regulated characteristics of the weapon immediately apparent to its owner. In short, we are confident that when the defendant knows of the characteristics of his weapon that bring it within the scope of the Act, the Government will not face great difficulty in proving that knowledge.” 511 U.S. 600, 615–16 n. 11, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994) (citation omitted); see also Flores–Figueroa v. United States, 556 U.S. 646, 129 S.Ct. 1886, 1893, 173 L.Ed.2d 853 (2009) (“The difficulties of proof along with the defendant's necessary guilt of a predicate crime and the defendant's necessary knowledge that he has acted ‘without lawful authority,’ make it reasonable, in the Government's view, to read the statute's language as dispensing with the knowledge requirement. We do not find this argument sufficient, however, to turn the tide in the Government's favor.”).

In rejecting the Government's repeated claims that the presumption of mens rea makes it too difficult to convict, the Supreme Court has tartly replied that strict liability can make it too easy to convict: “The Government asks us by a feat of construction radically to change the weights and balances in the scales of justice. The purpose and obvious effect of doing away with the requirement of a guilty intent is to ease the prosecution's path to conviction, to strip the defendant of such benefit as he derived at common law from innocence of evil purpose, and to circumscribe the freedom heretofore allowed juries.” Morissette v. United States, 342 U.S. 246, 263, 72 S.Ct. 240, 96 L.Ed. 288 (1952).

If proving mens rea as to a specific element is indeed considered too burdensome for the prosecution, Congress can always eliminate the mens rea requirement for a particular element or crime, subject to constitutional limits. As the Supreme Court said in Staples: “Of course, if Congress thinks it necessary to reduce the Government's burden at trial to ensure proper enforcement of the Act, it remains free to amend § 5861(d) by explicitly eliminating a mens rea requirement.” 511 U.S. at 616 n. 11, 114 S.Ct. 1793. So too here.

The presumption of mens rea is a core element of the rule of law; it also is workable and practical.

* * *

Convicting a defendant of this Section 924(c) offense and imposing an extra 20 years of mandatory imprisonment based on a fact the defendant did not know is unjust and incompatible with deeply rooted principles of American law. The Supreme Court has applied the presumption of mens rea precisely to avoid such injustice. Justice Jackson's opinion in Morissette described the link between mens rea and bedrock American principles of justice and responsibility: “The contention that an injury can amount to a crime only when inflicted by intention is no provincial or transient notion. It is as universal and persistent in mature systems of law as belief in freedom of the human will and a consequent ability and duty of the normal individual to choose between good and evil. A relation between some mental element and punishment for a harmful act is almost as instinctive as the child's familiar exculpatory ‘But I didn't mean to,’ and has afforded the rational basis for a tardy and unfinished substitution of deterrence and reformation in place of retaliation and vengeance as the motivation for public prosecution.” Morissette v. United States, 342 U.S. 246, 250–51, 72 S.Ct. 240, 96 L.Ed. 288 (1952) (footnote omitted).

The debate over mens rea is not some philosophical or academic exercise. It has major real-world consequences for criminal defendants. And it takes on added significance in an era of often lengthy mandatory minimum sentences. In this statute, dispensing with mens rea means an extra 20 years of mandatory imprisonment for the defendant, tripling the mandatory minimum sentence from 10 years to 30 years. And the 30–year sentence must be served consecutively to (that is, in addition to) the sentence for the underlying robbery. That is an extraordinarily harsh result for a fact the defendant did not know.

It's tempting to conclude that Burwell got what he deserved—that carrying a semi-automatic gun during a robbery (as Burwell allegedly believed he was doing) is just as depraved and blameworthy as carrying an automatic gun during a robbery. But neither Congress nor the Supreme Court agrees. Congress deliberately selected 10 years as the mandatory minimum sentence for a person who commits a robbery while carrying a semi-automatic gun. And Congress deliberately chose 30 years as the mandatory minimum sentence for a person who commits a robbery while carrying an automatic gun. As the Supreme Court explained in O'Brien, Congress drew that dramatic distinction because it believed that carrying an automatic gun during the robbery reflected significantly greater moral depravity by the defendant. But that link between the automatic weapon and greater moral depravity does not hold if the defendant actually thought his gun was a semi-automatic.

I would conclude that the presumption of mens rea applies to the automatic weapon element of Section 924(c). Applying that presumption, I would hold that the Government had to prove that Burwell knew his firearm was automatic. Of course, a properly instructed jury might or might not find Burwell guilty, but he is entitled to a jury instruction on whether he had the mens rea required for this offense. Because the District Court did not require the Government to prove that Burwell knew his gun was automatic, I would vacate Burwell's Section 924(c)(1)(B)(ii) conviction. I respectfully dissent.