(delivered the opinion of the Court)
In January 2003, petitioner Keshia Dixon purchased multiple firearms at two gun shows, during the course of which she provided an incorrect address and falsely stated that she was not under indictment for a felony. As a result of these illegal acts, petitioner was indicted and convicted on one count of receiving a firearm while under indictment in violation of 18 U. S. C. §922(n) and eight counts of making false statements in connection with the acquisition of a firearm in violation of §922(a)(6). At trial, petitioner admitted that she knew she was under indictment when she made the purchases and that she knew doing so was a crime; her defence was that she acted under duress because her boyfriend threatened to kill her or hurt her daughters if she did not buy the guns for him.
Petitioner contends that the trial judge’s instructions to the jury erroneously required her to prove duress by a preponderance of the evidence instead of requiring the Government to prove beyond a reasonable doubt that she did not act under duress. The Court of Appeals rejected petitioner’s contention, 413 F. 3d 520 (CA5 2005); given contrary treatment of the issue by other federal courts, we granted certiorari, 546 U. S. __ (2006).
At trial, in her request for jury instructions on her defence of duress, petitioner contended that she “should have the burden of production, and then that the Government should be required to disprove beyond a reasonable doubt the duress.” App. 300. Petitioner admitted that this request was contrary to Fifth Circuit precedent, and the trial court, correctly finding itself bound by Circuit precedent, denied petitioner’s request. Ibid. Instead, the judge’s instructions to the jury defined the elements of the duress defense and stated that petitioner has “the burden of proof to establish the defence of duress by a preponderance of the evidence.” Id., at 312.
Petitioner argues here, as she did in the District Court and the Court of Appeals, that federal law requires the Government to bear the burden of disproving her defence beyond a reasonable doubt and that the trial court’s erroneous instruction on this point entitles her to a new trial. There are two aspects to petitioner’s argument in support of her proposed instruction that merit separate discussion.
First, petitioner contends that her defence “controverted the mens rea required for conviction” and therefore that the Due Process Clause requires the Government to retain the burden of persuasion on that element. Brief for Petitioner 41.
Second, petitioner argues that Fifth Circuit’s rule is “contrary to modern common law.” Id., at 14.
The crimes for which petitioner was convicted require that she have acted “knowingly”, §922(a)(6), or “willfully”, §924(a)(1)(D). As we have explained, “unless the text of the statute dictates a different result, the term ‘knowingly’ merely requires proof of knowledge of the facts that constitute the offence.” Bryan v United States, 524 U. S. 184, 193 (1998) (footnote omitted). And the term “willfully” in §924(a)(1)(D) requires a defendant to have “acted with knowledge that his conduct was unlawful.” Ibid. In this case, then, the Government bore the burden of proving beyond a reasonable doubt that petitioner knew she was making false statements in connection with the acquisition of firearms and that she knew she was breaking the law when she acquired a firearm while under indictment. See In re Winship, 397 U. S. 358, 364 (1970). Although the Government may have proved these elements in other ways, it clearly met its burden when petitioner testified that she knowingly committed certain acts – she put a false address on the forms she completed to purchase the firearms, falsely claimed that she was the actual buyer of the firearms, and falsely stated that she was not under indictment at the time of the purchase – and when she testified that she knew she was breaking the law when, as an individual under indictment at the time, she purchased a firearm. App. 221–222.
Petitioner contends, however, that she cannot have formed the necessary mens rea for these crimes because she did not freely choose to commit the acts in question. But even if we assume that petitioner’s will was overborne by the threats made against her and her daughters, she still knew that she was making false statements and knew that she was breaking the law by buying a firearm. The duress defence, like the defence of necessity that we considered in United States v Bailey, 444 U. S. 394, 409–410 (1980), may excuse conduct that would otherwise be punishable, but the existence of duress normally does not controvert any of the elements of the offence itself. As we explained in Bailey, “[c]riminal liability is normally based upon the concurrence of two factors, ‘an evil-meaning mind [and] and evil-doing hand ....’” Id., at 402 (quoting Morissette v United States, 342 U. S. 246, 251 (1952)). Like the defence of necessity, the defence of duress does not negate a defendant’s criminal state of mind when the applicable offence requires a defendant to have acted knowingly or willfully; instead, it allows the defendant to “avoid liability .... because coercive conditions or necessity negates a conclusion of guilt even though the necessary mens rea was present.” Bailey, 444 U. S., at 402.
The fact that petitioner’s crimes are statutory offences that have no counterpart in the common law also supports our conclusion that her duress defence in no way disproves an element of those crimes. We have observed that “[t]he definition of the elements of a criminal offence is entrusted to the legislature, particularly in the case of federal crimes, which are solely creatures of statute.” Liparota v United States, 471 U. S. 419, 424 (1985). Here, consistent with the movement away from the traditional dichotomy of general versus specific intent and toward a more specifically defined hierarchy of culpable mental states, see Bailey, 444 U. S., at 403–404, Congress defined the crimes at issue to punish defendants who act “knowingly”, §922(a)(6), or “willfully”, §924(a)(1)(D). It is these specific mental states, rather than some vague “evil mind”, Brief for Petitioner 42, or “‘criminal’ intent”, Martin v Ohio, 480 U. S. 228, 235 (1987), that the Government is required to prove beyond a reasonable doubt, see Patterson v New York, 432 U. S. 197, n. 12 (1977) (“The applicability of the reasonable-doubt standard, however, has always been dependent on how a State defines the offence that is charged in any given case”). The jury instructions in this case were consistent with this requirement and, as such, did not run afoul of the Due Process Clause when they placed the burden on petitioner to establish the existence of duress by a preponderance of the evidence.
Having found no constitutional basis for placing upon the Government the burden of disproving petitioner’s duress defence beyond a reasonable doubt, we next address petitioner’s argument that the modern common law requires the Government to bear that burden. In making this argument, petitioner recognizes that, until the end of the 19th century, common-law courts generally adhered to the rule that “the proponent of an issue bears the burden of persuasion on the factual premises for applying the rule.” Fletcher, Two Kinds of Legal Rules: A Comparative Study of Burden-of-Persuasion Practices in Criminal Cases, 77 Yale L. J. 880, 898 (1967–1968). In petitioner’s view, however, two important developments have established a contrary common-law rule that now prevails in federal courts: this Court’s decision in Davis v United States, 160 U. S. 469 (1895), which placed the burden on the Government to prove a defendant’s sanity, and the publication of the Model Penal Code in 1962.
Although undisputed in this case, it bears repeating that, at common law, the burden of proving “affirmative defences – indeed, ‘all circumstances .... of justification, excuse or alleviation’ – rested on the defendant.” Patterson, 432 U. S., at 202 (quoting 4 W. Blackstone, Commentaries *201); see also Martin v Ohio, 480 U. S., at 235; Mullaney v Wilbur, 421 U. S. 684, 693 (1975). This common-law rule accords with the general evidentiary rule that “the burdens of producing evidence and of persuasion with regard to any given issue are both generally allocated to the same party.” 2 J. Strong, McCormick on Evidence §337, p. 415 (5th ed. 1999). And, in the context of the defence of duress, it accords with the doctrine that “where the facts with regard to an issue lie peculiarly in the knowledge of a party, that party has the burden of proving the issue.” Id., at 413. Although she claims that the common-law rule placing the burden on a defendant to prove the existence of duress “was the product of flawed reasoning”, petitioner accepts that this was the general rule, at least until this Court’s decision in Davis. Brief for Petitioner 18. According to petitioner, however, Davis initiated a revolution that overthrew the old common-law rule and established her proposed rule in its place.
Davis itself, however, does not support petitioner’s position. In that case, we reviewed a defendant’s conviction for having committed murder “feloniously, wilfully, and of his malice aforethought.” 160 U. S., at 474. It was undisputed that the prosecution’s evidence “if alone considered, made it the duty of the jury to return a verdict of guilty of the crime charged”; the defendant, however, adduced evidence at trial tending to show that he did not have the mental capacity to form the requisite intent. Id., at 475. At issue before the Court was the correctness of the trial judge’s instruction to the jury that the law “‘presumes every man is sane, and the burden of showing it is not true is upon the party who asserts it.’” Id., at 476. Under this instruction, “if the evidence was in equilibrio as to the accused being sane, that is, capable of comprehending the nature and effect of his acts, he was to be treated just as he would be if there were no defence of insanity or if there was an entire absence of proof that he was insane.” Id., at 479.
In reversing the defendant’s conviction, we found ourselves “unable to assent to the doctrine that in a prosecution for murder .... it is the duty of the jury to convict where the evidence is equally balanced on the issue as to the sanity of the accused at the time of the killing.” Id., at 484 (emphasis added). Instead, we concluded that this defendant was “entitled to an acquittal of the specific crime charged if upon all the evidence there is reasonable doubt whether he was capable in law of committing the crime.” Ibid. (emphasis added). Our opinion focused on the “definition of murder”, explaining that “it is of the very essence of that heinous crime that it be committed by a person of ‘sound memory and discretion,’ and with ‘malice aforethought.’” Ibid. (citations omitted). Reviewing “the adjudged cases” and “elementary treatises upon criminal law”, we found that “[a]ll admit that the crime of murder necessarily involves the possession by the accused of such mental capacity as will render him criminally responsible for his acts.” Id., at 485. Thus, when we ultimately found that the burden of proving the accused’s sanity rested on the Government, our holding rested on the conclusion that - Id., at 488:
[Davis’] guilt cannot be said to have been proved beyond a reasonable doubt – his will and his acts cannot be held to have joined in perpetrating the murder charged – if the jury, upon all the evidence, have a reasonable doubt whether he was legally capable of committing crime, or (which is the same thing) whether he wilfully, deliberately, unlawfully, and of malice aforethought took the life of the deceased. As the crime of murder involves sufficient capacity to distinguish between right and wrong, the legal interpretation of every verdict of guilty as charged is that the jury believed from all the evidence beyond a reasonable doubt that the accused was guilty, and was therefore responsible, criminally, for his acts. How then upon principle or consistently with humanity can a verdict of guilty be properly returned, if the jury entertain a reasonable doubt as to the existence of a fact which is essential to guilt, namely, the capacity in law of the accused to commit that crime?
Our opinion in Davis, then, interpreted a defendant’s sanity to controvert the necessary mens rea for the crime of murder committed “feloniously, wilfully, and of his malice aforethought”, id., at 474, as “[o]ne who takes human life cannot be said to be actuated by malice aforethought, or to have deliberately intended to take life, or to have ‘a wicked, depraved, and malignant heart,’ .... unless at the time he had sufficient mind to comprehend the criminality or the right and wrong of such an act”, id., at 485. We required the Government to prove the defendant’s sanity beyond a reasonable doubt because the evidence that tended to prove insanity also tended to disprove an essential element of the offence charged. See Davis v United States, 165 U. S. 373, 378 (1897) (“[T]he fact of sanity, as any other essential fact in the case, must be established to the satisfaction of the jury beyond a reasonable doubt” (emphasis added)). Whether or not this reasoning correctly treated insanity as negating the mens rea for murder as defined in the statute at issue, cf., n. 4, supra, it does not help petitioner: The evidence of duress she adduced at trial does not contradict or tend to disprove any element of the statutory offences that she committed.
Nor does the proposition for which Davis has come to stand help petitioner’s cause. Although written more narrowly in the context of a prosecution for the crime of murder, Davis was later interpreted to establish a general “rule for federal prosecutions .... that an accused is ‘entitled to an acquittal of the specific crime charged if upon all the evidence there is reasonable doubt whether he was capable in law of committing crime.’” Leland v Oregon, 343 U. S. 790, 797 (1952) (quoting Davis, 160 U. S., at 484); see also Lynch v Overholser, 369 U. S. 705, 713 (1962) (explaining that the Davis rule applied in all federal courts). After Davis, if a federal defendant introduced sufficient evidence to raise a reasonable doubt as to his sanity, it was sufficient to create a question for the jury on which the Government bore the ultimate burden of persuasion beyond a reasonable doubt. See, e.g.,
Hall v United States, 295 F. 2d 26, 28 (CA4 1961);
Holloway v United States, 148 F. 2d 665, 666 (CADC 1945);
Post v United States, 135 F. 1, 10 (CA5 1905).
In apparent recognition of the fact that Davis relied on the heightened mens rea applicable to the particular statute at issue, we held in Leland that this rule was not constitutionally mandated, 343 U. S., at 797, and Congress overruled it by statute in 1984, requiring a defendant to prove his insanity by clear and convincing evidence, 98 Stat. 2057, codified at 18 U. S. C. §17(b). Moreover, Congress has treated the defence of insanity differently from that of duress not only by codifying it but by requiring defendants who intend to rely on an insanity defence to provide advance notice to the Government. See Fed. Rule Crim. Proc. 12.2(a). Thus, even if the rule arising from Davis may have once been relevant to an evaluation of other affirmative defences, Congress’ differential treatment of the insanity defence and its rejection of the Davis rule are inconsistent with petitioner’s invitation to follow Davis’ lead in this case.
Indeed, petitioner’s reliance on Davis ignores the fact that federal crimes “are solely creatures of statute”, Liparota, 471 U. S., at 424, and therefore that we are required to effectuate the duress defence as Congress “may have contemplated” it in the context of these specific offences, United States v Oakland Cannabis Buyers’ Cooperative, 532 U. S. 483, n. 3 (2001) (internal quotation marks omitted); see also id., at 499 (Stevens, J., concurring in judgment) (explaining that Court was addressing whether the statute at issue foreclosed a necessity defence to specific charges brought under the statute); Bailey, 444 U. S., at 410 (“We need not speculate now, however, on the precise contours of whatever defences of duress or necessity are available against charges brought under [18 U. S. C.] §751(a)”). The offences at issue in this case were created by statute in 1968, when Congress enacted the Omnibus Crime Control and Safe Streets Act (hereinafter Safe Streets Act or Act). See 82 Stat. 197. There is no evidence in the Act’s structure or history that Congress actually considered the question of how the duress defence should work in this context, and there is no suggestion that the offences at issue are incompatible with a defence of duress. Cf. Oakland Cannabis Buyers’ Cooperative, 532 U. S., at 491. Assuming that a defence of duress is available to the statutory crimes at issue, then, we must determine what that defence would look like as Congress “may have contemplated” it.
As discussed above, the common law long required the defendant to bear the burden of proving the existence of duress. Similarly, even where Congress has enacted an affirmative defence in the proviso of a statute, the “settled rule in this jurisdiction [is] that an indictment or other pleading .... need not negative the matter of an exception made by a proviso or other distinct clause .... and that it is incumbent on one who relies on such an exception to set it up and establish it.” McKelvey v United States, 260 U. S. 353, 357 (1922); see also United States v Dickson, 15Pet. 141, 165 (1841) (calling this “the general rule of law which has always prevailed, and become consecrated almost as a maxim in the interpretation of statutes”). Even though the Safe Streets Act does not mention the defence of duress, we can safely assume that the 1968 Congress was familiar with both the long-established common-law rule and the rule applied in McKelvey and that it would have expected federal courts to apply a similar approach to any affirmative defence that might be asserted as a justification or excuse for violating the new law.
This conclusion is surely more reasonable than petitioner’s hypothesis that Davis dramatically upset a well-settled rule of law. Petitioner cites only one federal case decided before 1968 for the proposition that it has been well established in federal law that the Government bears the burden of disproving duress beyond a reasonable doubt. But that case involved a defendant’s claim that he “lacked the specific intent to defraud required by the statute for the reason that he committed the offence under duress and coercion.” Johnson v United States, 291 F. 2d 150, 152 (CA8 1961). Thus, when the Court of Appeals explained that “there is no burden upon the defendant to prove his defence of coercion”, id., at 155, that statement is best understood in context as a corollary to the by-then-unremarkable proposition that “the burden of proof rests upon the Government to prove the defendant’s guilt beyond a reasonable doubt”, ibid. Properly understood, Johnson provides petitioner little help in her uphill struggle to prove that a dramatic shift in the federal common-law rule occurred between Davis and the enactment of the Safe Streets Act in 1968.
Indeed, for us to be able to accept petitioner’s proposition, we would need to find an overwhelming consensus among federal courts that it is the Government’s burden to disprove the existence of duress beyond a reasonable doubt. The existence today of disagreement among the Federal Courts of Appeals on this issue, however – the very disagreement that caused us to grant certiorari in this case, see n. 1, supra – demonstrates that no such consensus has ever existed. See also post, at 6–8 (Breyer, J., dissenting) (discussing differences in treatment of the duress defence by the various Courts of Appeals). Also undermining petitioner’s argument is the fact that, in 1970, the National Commission on Reform of Federal Criminal Laws proposed that a defendant prove the existence of duress by a preponderance of the evidence. See 1 Working Papers 278. Moreover, while there seem to be few, if any, post-Davis, pre-1968 cases placing the burden on a defendant to prove the existence of duress, or even discussing the issue in any way, this lack of evidence does not help petitioner. The long-established common-law rule is that the burden of proving duress rests on the defendant. Petitioner hypothesizes that Davis fomented a revolution upsetting this rule. If this were true, one would expect to find cases discussing the matter. But no such cases exist.
It is for a similar reason that we give no weight to the publication of the Model Penal Code in 1962. As petitioner notes, the Code would place the burden on the government to disprove the existence of duress beyond a reasonable doubt. See Model Penal Code §1.12, 10A U. L. A. 88 (2001) (hereinafter Model Penal Code or Code) (stating that each element of an offence must be proved beyond a reasonable doubt); §1.13(9)(c), at 91 (defining as an element anything that negatives an excuse for the conduct at issue); §2.09, at 131–132 (establishing affirmative defence of duress). Petitioner argues that the Code reflects “well established” federal law as it existed at the time. Brief for Petitioner 25. But, as discussed above, no such consensus existed when Congress passed the Safe Streets Act in 1968. And even if we assume Congress’ familiarity with the Code and the rule it would establish, there is no evidence that Congress endorsed the Code’s views or incorporated them into the Safe Streets Act.
In fact, the Act itself provides evidence to the contrary. Despite the Code’s careful delineation of mental states, see Model Penal Code §2.02, 10A U. L. A., at 94–95, the Safe Streets Act attached no explicit mens rea requirement to the crime of receiving a firearm while under indictment, §924(a), 82 Stat. 233 (“Whoever violates any provision of this chapter .... shall be fined not more than $5,000 or imprisoned not more than five years, or both”). And when Congress amended the Act to impose a mens rea requirement, it punished people who “willfully” violate the statute, see 100 Stat. 456, a mental state that has not been embraced by the Code, see Model Penal Code §2.02(2), 10A U. L. A., at 94–95 (defining “purposely”, “knowingly”, “recklessly”, and “negligently”); Explanatory Note, p. 97 (“Though the term ‘wilfully’ is not used in the definitions of crimes contained in the Code, its currency and its existence in offences outside the criminal code suggest the desirability of clarification”). Had Congress intended to adopt the Code’s structure when it enacted or amended the Safe Streets Act, one would expect the Act’s form and language to adhere much more closely to that used by the Code. It does not, and, for that reason, we cannot rely on the Model Penal Code to provide evidence as to how Congress would have wanted us to effectuate the duress defence in this context.
Congress can, if it chooses, enact a duress defence that places the burden on the Government to disprove duress beyond a reasonable doubt. In light of Congress’ silence on the issue, however, it is up to the federal courts to effectuate the affirmative defence of duress as Congress “may have contemplated” it in an specific-specific context. Oakland Cannabis Buyers’ Cooperative, 532 U. S., at 491, n. 3. In the context of the firearms offences at issue – as will usually be the case, given the long-established common-law rule – we presume that Congress intended the petitioner to bear the burden of proving the defence of duress by a preponderance of the evidence. Accordingly, the judgment of the Court of Appeals is affirmed.
It is so ordered.
No one disputes that, subject to constitutional constraints, Congress has the authority to determine the content of a duress defence with respect to federal crimes and to direct whether the burden of proof rests with the defence or the prosecution. The question here is how to proceed when Congress has enacted a criminal statute, the Omnibus Crime Control and Safe Streets Act of 1968, 82 Stat. 197 et seq. (hereinafter Safe Streets Act), without explicit instructions regarding the duress defence or its burden of proof. See ante, at 10.
When issues of congressional intent with respect to the nature, extent, and definition of federal crimes arise, we assume Congress acted against certain background understandings set forth in judicial decisions in the Anglo-American legal tradition. See United States v Bailey, 444 U. S. 394, n. 11 (1980). Those decisions, in turn, consult sources such as legal treatises and the American Legal Institute’s Model Penal Code. See, e.g., United States v Jimenez Recio, 537 U. S. 270, 275–276 (2003); Salinas v United States, 522 U. S. 52, 64–65 (1997). All of these sources rely upon the insight gained over time as the legal process continues. Absent some contrary indication in the statute, we can assume that Congress would not want to foreclose the courts from consulting these newer sources and considering innovative arguments in resolving issues not confronted in the statute and not within the likely purview of Congress when it enacted the criminal prohibition applicable in the particular case.
While the Court looks to the state of the law at the time the statute was enacted, see ante, at 12, the better reading of the Court’s opinion is that isolated authorities or writings do not control unless they were indicative of guiding principles upon which Congress likely would have relied. Otherwise, it seems altogether a fiction to attribute to Congress any intent one way or the other in assigning the burden of proof. It seems unlikely, moreover, that Congress would have wanted the burden of proof for duress to vary from statute to statute depending upon the date of enactment. Consistent with these propositions, the Court looks not only to our precedents and common-law traditions, but also to the treatment of the insanity defence in a 1984 statute and a proposal of the National Commission on Reform of Federal Criminal Laws, even though they both post-dated the passage of the Safe Streets Act. See ante, at 10, 13.
As there is no reason to suppose that Congress wanted to depart from the traditional principles for allocating the burden of proof, the proper approach is simply to apply these principles to the context of duress. See, e.g., Schaffer v Weast, 546 U. S. 49 (2005) (where the plain text of the statute is “silent on the allocation of the burden of persuasion”, we proceed to consider the “ordinary default rule” and its exceptions). The facts needed to prove or disprove the defence “lie peculiarly in the knowledge of” the defendant. 2 K. Broun, McCormick on Evidence §337, p. 475 (6th ed. 2006); see ante, at 6. The claim of duress in most instances depends upon conduct that takes place before the criminal act; and, as the person who allegedly coerced the defendant is often unwilling to come forward and testify, the prosecution may be without any practical means of disproving the defendant’s allegations. There is good reason, then, to maintain the usual rule of placing the burden of production and persuasion together on the party raising the issue. See 2 Broun, supra, §337; ante, at 6. The analysis may come to a different result, of course, for other defences.
With these observations, I join the Court’s opinion.
(with whom Justice Scalia joins, concurring)
I join the opinion of the Court with the understanding that it does not hold that the allocation of the burden of persuasion on the defence of duress may vary from one federal criminal statute to another.
Duress was an established defence at common law. See 4 W. Blackstone, Commentaries on the Laws of England 30 (1769). When Congress began to enact federal criminal statutes, it presumptively intended for those offences to be subject to this defence. Moreover, Congress presumptively intended for the burdens of production and persuasion to be placed, as they were at common law, on the defendant. Although Congress is certainly free to alter this pattern and place one or both burdens on the prosecution, either for all or selected federal crimes, Congress has not done so but instead has continued to revise the federal criminal laws and to create new federal crimes without addressing the issue of duress. Under these circumstances, I believe that the burdens remain where they were when Congress began enacting federal criminal statutes.
I do not assume that Congress makes a new, implicit judgment about the allocation of these burdens whenever it creates a new federal crime or, for that matter, whenever it substantially revises an existing criminal statute. It is unrealistic to assume that on every such occasion Congress surveys the allocation of the burdens of proof on duress under the existing federal case law and under the law of the States and tacitly adopts whatever the predominant position happens to be at the time. Such a methodology would create serious problems for the district courts and the courts of appeals when they are required to decide where the burden of persuasion should be allocated for federal crimes enacted on different dates. If the allocation differed for different offences, there might be federal criminal cases in which the trial judge would be forced to instruct the jury that the defendant bears the burden of persuasion on this defence for some of the offences charged in the indictment and that the prosecution bears the burden on others.
I would also not assume, as Justice Breyer does, see post, at 2–3, that Congress has implicitly delegated to the federal courts the task of deciding in the manner of a common-law court where the burden of persuasion should be allocated. The allocation of this burden is a debatable policy question with an important empirical component. In the absence of specific direction from Congress, cf. Fed. Rule Evid. 501, I would not assume that Congress has conferred this authority on the Judiciary.
(with whom Justice Souter joins, dissenting)
Courts have long recognized that “duress” constitutes a defence to a criminal charge. Historically, that defence “excuse[d] criminal conduct” if
a “threat of imminent death or serious bodily injury” led the defendant to commit the crime,
the defendant had no reasonable, legal alternative to breaking the law, and
the defendant was not responsible for creating the threat –
United States v Bailey, 444 U. S. 394, 409–410 (1980); see also 2 W. LaFave, Substantive Criminal Law §9.7(b), pp. 74–82 (2003) (hereinafter LaFave); ante, at 2, n. 1 (opinion of the Court). The Court decides today in respect to federal crimes that the defence must bear the burden of both producing evidence of duress and persuading the jury. I agree with the majority that the burden of production lies on the defendant, that here the burden of persuasion issue is not constitutional, and that Congress may allocate that burden as it sees fit. But I also believe that, in the absence of any indication of a different congressional intent, the burden of persuading the jury beyond a reasonable doubt should lie where such burdens normally lie in criminal cases, upon the prosecution.
My disagreement with the majority in part reflects my different view about how we should determine the relevant congressional intent. Where Congress speaks about burdens of proof, we must, of course, follow what it says. But suppose, as is normally the case, that the relevant federal statute is silent? The majority proceeds on the assumption that Congress wished courts to fill the gap by examining judicial practice at the time that Congress enacted the particular criminal statute in question. Ante, at 10–14 (opinion of the Court). I would not follow that approach.
To believe Congress intended the placement of such burdens to vary from statute to statute and time to time is both unrealistic and risks unnecessary complexity, jury confusion, and unfairness. It is unrealistic because the silence could well mean only that Congress did not specifically consider the “burden of persuasion” in respect to a duress defence. It simply did not think about that secondary matter. Had it done so, would Congress have wanted courts to freeze current practice statute-by-statute? Would it have wanted to impose different burden-of-proof requirements where claims of duress are identical, where statutes are similar, where the only relevant difference is the time of enactment? Why? Indeed, individual instances of criminal conduct often violate several statutes. In a trial for those violations, is the judge to instruct the jury to apply different standards of proof to a duress defence depending upon when Congress enacted the particular statute in question? What if in this very case the defendant’s boyfriend had given her drug money and insisted (under threat of death) not only that she use some of the money to buy him a gun, but that she launder the rest? See 18 U. S. C. §1956 (2000 ed. and Supp. II); see infra, at 6–7.
I would assume instead that Congress’ silence typically means that Congress expected the courts to develop burden rules governing affirmative defences as they have done in the past, by beginning with the common law and taking full account of the subsequent need for that law to evolve through judicial practice informed by reason and experience. See Davis v United States, 160 U. S. 469 (1895); McNabb v United States, 318 U. S. 332, 341 (1943); ante, at 11, n. 8 (opinion of the Court) (proposed general revision of the federal criminal code would have instructed courts to determine the contours of affirmative defences “‘according to the principles of the common law as they may be interpreted in the light of reason and experience’”); 9 J. Wigmore, Evidence §2486, p. 291 (J. Chadbourn rev. ed. 1981) (allocation of the burdens of proof present courts with questions “of policy and fairness based on experience in the different situations”). That approach would produce uniform federal practice across different affirmative defences, as well as across statutes passed at different points in time.
My approach leads me to conclude that in federal criminal cases, the prosecution should bear the duress defence burden of persuasion. The issue is a close one. In Blackstone’s time the accused bore the burden of proof for all affirmative defences. See 4 W. Blackstone, Commentaries *201; Patterson v New York, 432 U. S. 197, 201–202 (1977). And 20th-century experts have taken different positions on the matter. The Model Penal Code, for example, recommends placing the burden of persuasion on the prosecution. ALI, Model Penal Code §1.12, p. 16, §1.13(9)(c), p. 18, §2.09, pp. 37–38 (1985). The Brown Commission recommends placing it upon the defendant. National Commission on Reform of Federal Criminal Laws, 1 Working Papers 278 (1970). And the proposed revision of the federal criminal code, agnostically, would have turned the matter over to the courts for decision. S. 1722, 96th Cong., 1st Sess., §501 (1979). Moreover, there is a practical argument that favours the Government’s position here, namely that defendants should bear the burden of persuasion because defendants often have superior access to the relevant proof.
Nonetheless, several factors favour placing the burden on the prosecution. For one thing, in certain respects the question of duress resembles that of mens rea, an issue that is always for the prosecution to prove beyond a reasonable doubt. See In re Winship, 397 U. S. 358, 364 (1970); Martin v Ohio, 480 U. S. 228, 234 (1987). The questions are not the same. The defendant’s criminal activity here was voluntary; no external principle, such as the wind, propelled her when she acted. The Nicomachean Ethics of Aristotle, p. 54 (R. W. Browne transl. 1865). Moreover, her actions were intentional. Whether she wanted to buy the guns or not, and whether she wanted to lie while doing so or not, she decided to do these things and knew that she was doing them. Indeed, her action was willful in the sense that she knew that to do them was to break the law. Ante, at 3–4 (opinion of the Court); see also Ratzlaf v United States, 510 U. S. 135, 136–137 (1994).
Nonetheless, where a defendant acts under duress, she lacks any semblance of a meaningful choice. In that sense her choice is not free. As Blackstone wrote, the criminal law punishes “abuse[s] of th[e] free will”; hence “it is highly just and equitable that a man should be excused for those acts, which are done through unavoidable force and compulsion.” 4 Commentaries *27. And it is in this “force and compulsion”, acting upon the will, that the resemblance to lack of mens rea lies. Cf. Austin, Ifs and Cans, in Proceedings of the British Academy 123–124 (1956) (noting difference between choosing to do something where one has the opportunity and ability to do otherwise and choosing to do something where one lacks any such opportunity or ability). Davis v United States, supra, allocated the federal insanity defence burden to the Government partly for these reasons. That case, read in light of Leland v Oregon, 343 U. S. 790, 797 (1952), suggests that, even if insanity does not always show the absence of mens rea, it does show the absence of a “‘vicious will.’” Davis, supra, at 484 (citing Blackstone; emphasis added).
For another thing, federal courts (as a matter of statutory construction or supervisory power) have imposed the federal-crime burden of persuasion upon the prosecution in respect to self-defence, insanity, and entrapment, which resemble the duress defence in certain relevant ways. In respect to both duress and self-defence, for example, the defendant’s illegal act is voluntary, indeed, intentional; but the circumstances deprive the defendant of any meaningful ability or opportunity to act otherwise, depriving the defendant of a choice that is free. Insanity, as I said, may involve circumstances that resemble, but are not identical to, a lack of mens rea. And entrapment requires the prosecution to prove that the defendant was “predisposed” to commit the crime – a matter sometimes best known to the defendant.
As to self-defence, see
First Circuit Pattern Criminal Jury Instructions §5.04 (1998);
United States v Thomas, 34 F. 3d 44, 47 (CA2 1994);
Government of Virgin Islands v Smith, 949 F. 2d 677, 680 (CA3 1991);
United States v Harris, Nos. 95–5637, 95–5638, 1996 U. S. App. LEXIS 22040, *4–*5 (CA4, Aug. 27, 1996);
United States v Branch, 91 F. 3d 699, 714, n. 1 (CA5 1996);
Sixth Circuit Pattern Criminal Jury Instructions §6.06 (2005);
United States v Jackson, 569 F. 2d 1003, 1008, n. 12 (CA7 1978);
United States v Pierre, 254 F. 3d 872, 876 (CA9 2001);
United States v Corrigan, 548 F. 2d 879, 883 (CA10 1977);
United States v Alvarez, 755 F. 2d 830, 842 (CA11 1985);
Bynum v United States, 408 F. 2d 1207 (CADC 1968);
see also Mullaney v Wilbur, 421 U. S. 684, n. 30 (1975) (noting this as the “‘majority rule’”).
As to insanity, see Davis, 160 U. S., at 486; Leland, supra, at 797 (making clear that Davis determined burden allocations as a matter of federal, but not constitutional, law); but see 18 U. S. C. §17(b) (overruling this default rule to place the burden on the defendant by clear and convincing evidence). As to entrapment, see Jacobson v United States, 503 U. S. 540, 554 (1992) (reversing the judgment affirming the conviction because “the prosecution failed, as a matter of law, to adduce evidence to support the jury verdict that petitioner was predisposed, independent of the Government’s acts and beyond a reasonable doubt” to commit the crime). See also Patterson, 432 U. S., at 202 (noting that Davis “had wide impact on the practice in the federal courts with respect to the burden of proving various affirmative defences”); Patterson, supra, at 231 (Powell, J., dissenting) (“[S]ince this Court’s decision in Davis .... federal prosecutors have borne the burden of persuasion with respect to factors like insanity, self-defence, and malice or provocation, once the defendant has carried this burden of production”).
Further, most federal courts, in respect to most federal crimes, have imposed the burden of persuasion in respect to the duress defence upon the Government, following Johnson v United States, 291 F. 2d 150, 155 (CA8 1961), and authorities such as E. Devitt & C. Blackmar, Federal Jury Practice and Instructions §13.14, p. 293 (2d ed. 1970), and the Federal Judicial Center Pattern Criminal Jury Instructions §56 (1988). By the mid-1990’s, seven circuits had squarely placed the burden of persuasion upon the prosecution; one Circuit (the Fifth) placed the burden on the defendant; and four (the Third, Fourth, Eleventh, and District of Columbia) did not, as far as I can tell, have a definitive practice. Compare
United States v Arthurs, 73 F. 3d 444, 448(CA1 1996);
United States v Mitchell, 725 F. 2d 832, 836 (CA2 1983);
United States v Campbell, 675 F. 2d 815, 821 (CA6 1982);
United States v Talbott, 78 F. 3d 1183, 1186 (CA7 1996) (per curiam);
United States v Campbell, 609 F. 2d 922, 925 (CA8 1979);
United States v Hearst, 563 F. 2d 1331, 1336, and n. 2 (CA9 1977);
United States v Falcon, 766 F. 2d 1469, 1477 (CA10 1985),
with United States v Willis, 38 F. 3d 170, 179 (CA5 1994) (putting the burden on the defendant by a preponderance). Compare also
First Circuit Pattern Criminal Jury Instructions §5.05 (1998);
Sixth Circuit Pattern Criminal Jury Instructions §6.05 (1991);
Seventh Circuit Pattern Criminal Jury Instructions §6.08 (1998); and
Eighth Circuit Pattern Criminal Jury Instructions §§3.09, 9.02 (2000),
with Fifth Circuit Pattern Criminal Jury Instructions §1.36 (2001). Petitioner adds, without contradiction, that the States allocate the burden similarly by a ratio of 2 to 1. Brief for Petitioner 32–34; Brief for United States 38, n. 30.
Beginning in 1991, the matter became more complicated because the Ninth Circuit began to require the defendant to bear the burden of proving duress in certain circumstances - United States v Dominguez-Mestas, 929 F. 2d 1379, 1382, 1384 (1991) (per curiam). And a few years later the Third, Sixth, and Eleventh Circuits followed suit in cases concerning a closely related justification defence. See
United States v Dodd, 225 F. 3d 340, 347–350 (CA3 2000);
United States v Brown, 367 F. 3d 549, 555–556 (CA6 2004);
United States v Deleveaux, 205 F. 3d 1292, 1298-1300 (CA11 2000);
Eleventh Circuit Pattern Criminal Jury Instructions §16 (2003).
But see Sixth Circuit Pattern Criminal Jury Instructions §6.05 (2005) (stating that the burden of proof issue for duress is undecided in that Circuit).
These latter cases, however, put the burden on the defendant only where the criminal statute narrows its mens rea requirement, i.e., the burden is the defendant’s where the statute requires that the defendant act with “knowledge” but not, suggest these courts, where the statute requires that the defendant act “willfully”, “intentionally”, or “voluntarily.” See, e.g.,
Dominguez-Mestas, supra, at 1382, 1384;
United States v Meraz-Solomon, 3 F. 3d 298, 300 (CA9 1993);
Ninth Circuit Pattern Criminal Jury Instructions §§6.5, 6.6 (2003);
but see United States v Fei Lin, 139 F. 3d 1303, 1307–1308 (CA9 1998). See also Eleventh Circuit Pattern Criminal Jury Instructions §16 (2003); United States v Diaz, 285 F. 3d 92, 97 (CA1 2002) (indicating that this bifurcated rule might be appropriate, but noting Circuit precedent to the contrary). Similarly, the Tenth Circuit placed the burden of proving duress upon the defendant in “strict liability” cases where mens rea is not an element of the crime at all. United States v Unser, 165 F. 3d 755, 763–765 (CA10 1999).
The apparent upshot is that four Circuits now place the burden of persuasion on the prosecution across the board; one places the burden on the prosecution if the statute requires mens rea but not otherwise; and four have held or suggested that the burden should be on the prosecution if the statute requires an intentional or willful state of mind, but not if the statute requires only knowledge. While the Circuits are divided, apparently only one (the Fifth) agrees with the position taken by the Court today.
Further, while I concede the logic of the Government’s practical argument – that defendants have superior access to the evidence – I remain uncertain of the argument’s strength. After all, “[i]n every criminal case the defendant has at least an equal familiarity with the facts and in most a greater familiarity with them than the prosecution.” Tot v United States, 319 U. S. 463, 469 (1943). And the strict contours of the duress defence, as well as the defendant’s burden of production, already substantially narrow the circumstances under which the defence may be used. A defendant may find it difficult, for example, to show duress where the relevant conduct took place too long before the criminal act. Cf. ante, at 2 (opinion of Kennedy, J.). That is because the defendant must show that he had no alternative to breaking the law - supra, at 1. And that will be the more difficult to show the more remote the threat. See also LaFave, § 9.7, p. 77-79 (duress generally requires an “immediate” or “imminent” threat, that the defendant “take advantage of a reasonable opportunity to escape”, and that the defendant “terminate his conduct ‘as soon as the claimed duress .... had lost its coercive force’”). More important, the need to prove mens rea can easily present precisely the same practical difficulties of proof for the prosecutor. Suppose for example the defendant claims that an old lady told him that the white powder he transported across the border was medicine for her dying son. Cf. United States v Mares, 441 F. 3d 1152 (CA10 2006). See also Mullaney v Wilbur, 421 U. S., at 702 (requiring the government to prove an absence of passion in a murder conviction imposes “no unique hardship on the prosecution”).
It is particularly difficult to see a practical distinction between this affirmative defence and, say, self-defense.The Government says that the prosecution may “be unable to call the witness most likely to have information bearing on the point”, namely, the defendant. Brief for United States 21. But what is the difference in this respect between the defendant here, who says her boyfriend threatened to kill her, and a battered woman who says that she killed her husband in self-defence, where the husband’s evidence is certainly unavailable? See also Jacobson, 503 U. S. 540 (entrapment; need to prove “propensity”). Regardless, unless the defendant testifies, it could prove difficult to satisfy the defendant’s burden of production; and, of course, once the defendant testifies, cross-examination is possible.
In a word, I cannot evaluate the claim of practicality without somewhat more systematic evidence of the existence of a problem, say, in those Circuits that for many years have imposed the burden on the prosecutor. And, of course, if I am wrong about the Government’s practical need (and were my views to prevail), the Government would remain free to ask Congress to reallocate the burden.
Finally, there is a virtue in uniformity, in treating the federal statutory burden of persuasion similarly in respect to actus reus, mens rea, mistake, self-defence, entrapment, and duress. The Second Circuit, when imposing the burden of persuasion for duress on the prosecution, wrote that differences in this respect create “a grave possibility of juror confusion.” United States v Mitchell, 725 F. 2d 832, 836 (1983) (Newman, J., joined by Feinberg, C. J., and Friendly, J.). They risk unfairness as well.
For these reasons I believe that, in the absence of an indication of congressional intent to the contrary, federal criminal law should place the burden of persuasion in respect to the duress defence upon the prosecution, which, as is now common in respect to many affirmative defences, it must prove beyond a reasonable doubt. With respect, I dissent.
 Cf., e.g., United States v Talbott, 78 F. 3d 1183, 1186 (CA7 1996) (per curiam); United States v Riffe, 28 F. 3d 565, 568, n. 2 (CA6 1994); United States v Simpson, 979 F. 2d 1282, 1287 (CA8 1992).
 There is no federal statute defining the elements of the duress defence. We have not specified the elements of the defence, see, e.g., United States v Bailey, 444 U. S. 394, 409–410 (1980), and need not do so today. Instead, we presume the accuracy of the District Court’s description of these elements:
The defendant was under an unlawful and imminent threat of such a nature as to induce a well-grounded apprehension of death or serious bodily injury;
the defendant had not recklessly or negligently placed herself in a situation in which it was probable that she would be forced to perform the criminal conduct;
the defendant had no reasonable, legal alternative to violating the law, that is, a chance both to refuse to perform the criminal act and also to avoid the threatened harm; and,
that a direct causal relationship may be reasonably anticipated between the criminal act and the avoidance of the threatened harm.
See App. 312–313; see generally United States v Harper, 802 F. 2d 115, 118 (CA5 1986).
 Although §922(n) does not contain a mens rea requirement, the relevant sentencing provision, §924(a)(1)(D), requires that a violation be committed willfully.
 As the Government recognized at oral argument, there may be crimes where the nature of the mens rea would require the Government to disprove the existence of duress beyond a reasonable doubt. See Tr. of Oral Arg. 26–27; see also, e.g., 1 W. LaFave, Substantive Criminal Law §5.1, p. 333 (2d ed. 2003) (hereinafter LaFave) (explaining that some common-law crimes require that the crime be done “‘maliciously’”); Black’s Law Dictionary 968 (7th ed. 1999) (defining malice as “[t]he intent, without justification or excuse, to commit a wrongful act”).
 Professor LaFave has explained the duress defence as follows: “The rationale of the defence is not that the defendant, faced with the unnerving threat of harm unless he does an act which violates the literal language of the criminal law, somehow loses his mental capacity to commit the crime in question. Nor is it that the defendant has not engaged in a voluntary act. Rather it is that, even though he has done the act the crime requires and has the mental state which the crime requires, his conduct which violates the literal language of the criminal law is excused ....” 2 LaFave §9.7(a), at 72 (footnote omitted).
 While Congress’ findings in support of the Safe Streets Act show that Congress was concerned because “the ease with which any person can acquire firearms .... is a significant factor in the prevalence of lawlessness and violent crime in the United States”, §901(a)(2), 82 Stat. 225, it would be unrealistic to read this concern with the proliferation of firearm-based violent crime as implicitly doing away with a defence as strongly rooted in history as the duress defence, see, e.g., 4 W. Blackstone, Commentaries on the Laws of England 30 (1769).
 We have previously made this assumption when addressing common-law affirmative defences, see United States v Oakland Cannabis Buyers’ Cooperative, 532 U. S. 483, 491 (2001); Bailey, 444 U. S., at 410, and the parties give us no reason to question it here.
 Indeed, when a congressional committee did consider codifying the duress defence, it would have had the courts determine the defence “according to the principles of the common law as they may be interpreted in the light of reason and experience.” S. 1437, 95th Cong., 2d Sess., §501 (1978).
 Duress, like the defence at issue in McKelvey, is an excuse that allows an exception from liability. See, e.g., 2 LaFave §9.7, at 72 (“The rationale of the defence of duress is that the defendant ought to be excused when he ‘is the victim of a threat that a person of reasonable moral strength could not fairly be expected to resist’”).
 In D’Aquino v United States, 192 F. 2d 338, 358, n. 11 (CA9 1951), the trial court instructed the jury that it would be warranted in acquitting the defendant on the basis that she acted under duress “‘if you believe from the evidence that the defendant committed these acts that the Government alleges .... under a well grounded apprehension of immediate death or serious bodily injury ....’” This instruction did not require the Government to disprove duress beyond a reasonable doubt, and it seemingly placed the burden on the defendant to prove the existence of duress.
Bryan v United States, 524 U. S. 184 (1998)
D’Aquino v United States, 192 F. 2d 338, 358, n. 11 (CA9 1951)
Davis v United States, 160 U. S. 469 (1895)
Davis v United States, 165 U. S. 373 (1897)
Bynum v United States, 408 F. 2d 1207 (CADC 1968)
Government of Virgin Islands v Smith, 949 F. 2d 677 (CA3 1991)
Hall v United States, 295 F. 2d 26 (CA4 1961)
Holloway v United States, 148 F. 2d 665 (CADC 1945)
In re Winship, 397 U. S. 358 (1970)
Johnson v United States, 291 F. 2d 150 (CA8 1961)
Jacobson v United States, 503 U. S. 540 (1992)
Johnson v United States, 291 F. 2d 150 (CA8 1961)
Davis v United States, 160 U. S. 469 (1895)
Leland v Oregon, 343 U. S. 790 (1952)
Liparota v United States, 471 U. S. 419 (1985)
Lynch v Overholser, 369 U. S. 705 (1962)
Morissette v United States, 342 U. S. 246 (1952)
Martin v Ohio, 480 U. S. 228 (1987)
McNabb v United States, 318 U. S. 332 (1943)
McKelvey v United States, 260 U. S. 353 (1922)
Mullaney v Wilbur, 421 U. S. 684 (1975)
Oakland Cannabis Buyers’ Cooperative, 532 U. S.
Patterson v New York, 432 U. S. 197 (1977)
Post v United States, 135 F. 1 (CA5 1905)
Martin v Ohio, 480 U. S. 228 (1987)
Ratzlaf v United States, 510 U. S. 135 (1994)
Salinas v United States, 522 U. S. 52 (1997)
Schaffer v Weast, 546 U. S. 49 (2005)
Tot v United States, 319 U. S. 463 (1943)
United States v Alvarez, 755 F. 2d 830 (CA11 1985)
United States v Arthurs, 73 F. 3d 444 (CA1 1996)
United States v Bailey, 444 U. S. 394 (1980)
United States v Branch, 91 F. 3d 699 (CA5 1996)
United States v Brown, 367 F. 3d 549 (CA6 2004)
United States v Campbell, 675 F. 2d 815 (CA6 1982)
United States v Campbell, 609 F. 2d 922 (CA8 1979)
United States v Corrigan, 548 F. 2d 879 (CA10 1977)
United States v Deleveaux, 205 F. 3d 1292 (CA11 2000)
United States v Diaz, 285 F. 3d 92 (CA1 2002)
United States v Dickson, 15Pet. 141 (1841)
United States v Dodd, 225 F. 3d 340 (CA3 2000)
United States v Dominguez-Mestas, 929 F. 2d 1379 (1991)
United States v Falcon, 766 F. 2d 1469 (CA10 1985)
United States v Fei Lin, 139 F. 3d 1303 (CA9 1998)
United States v Harris, Nos. 95–5637, 1996 U. S. App. LEXIS 22040, *4–*5 (CA4, Aug. 27, 1996)
United States v Harper, 802 F. 2d 115 (CA5 1986)
United States v Hearst, 563 F. 2d 1331 (CA9 1977)
United States v Jackson, 569 F. 2d 1003 (CA7 1978)
United States v Jimenez Recio, 537 U. S. 270 (2003)
United States v Mares, 441 F. 3d 1152 (CA10 2006)
United States v Meraz-Solomon, 3 F. 3d 298 (CA9 1993)
United States v Mitchell, 725 F. 2d 832 (CA2 1983)
United States v Oakland Cannabis Buyers’ Cooperative, 532 U. S. 483 (2001)
United States v Pierre, 254 F. 3d 872 (CA9 2001)
United States v Riffe, 28 F. 3d 565 (CA6 1994)
United States v Simpson, 979 F. 2d 1282 (CA8 1992)
United States v Thomas, 34 F. 3d 44 (CA2 1994)
United States v Talbott, 78 F. 3d 1183 (CA7 1996)
United States v Unser, 165 F. 3d 755 (CA10 1999)
United States v Willis, 38 F. 3d 170 (CA5 1994)
Omnibus Crime Control and Safe Streets Act of 1968, 82 Stat. 197
Authors and other references
Fletcher, Two Kinds of Legal Rules: A Comparative Study of Burden-of-Persuasion Practices in Criminal Cases, 77 Yale L. J. 880 (1967–1968)
2 J. Strong, McCormick on Evidence §337 (5th ed. 1999)
National Commission on Reform of Federal Criminal Laws, 1 Working Papers 278
American Legal Institute’s Model Penal Code
2 K. Broun, McCormick on Evidence §337 (6th ed. 2006)
2 W. LaFave, Substantive Criminal Law §9.7(b) (2003)
9 J. Wigmore, Evidence (J. Chadbourn rev. ed. 1981)
The Nicomachean Ethics of Aristotle (R. W. Browne transl. 1865)
Austin, Ifs and Cans, in Proceedings of the British Academy (1956)
E. Devitt & C. Blackmar, Federal Jury Practice and Instructions (2d ed. 1970)
Black’s Law Dictionary (7th ed. 1999)
4 W. Blackstone, Commentaries on the Laws of England (1769)
Federal Judicial Center Pattern Criminal Jury Instructions (1988)
First Circuit Pattern Criminal Jury Instructions (1998)
Sixth Circuit Pattern Criminal Jury Instructions (2005)
Ninth Circuit Pattern Criminal Jury Instructions (2003)
Eleventh Circuit Pattern Criminal Jury Instructions (2003)
J. Craig Jett (Jeffrey T. Green with him) for the petitioner.
Irving L. Gornstein (Solicitor General Clement, Assistant Attorney General Fisher, Deputy Solicitor General Dreeben, Deanne E. Maynard, and Deborah Watson with him) for the United States.
Elliot H. Scherker, Julissa Rodriguez, Karen M. Gottlieb, Peter Goldberger, and Pamela Harris filed a brief for the National Association of Criminal Defense Lawyers et al. as amici curiae urging reversal.
all rights reserved