This case presents the issue whether knowledge that the intended victim is a federal officer is a requisite for the crime of conspiracy, under 18 U.S.C. § 371, to commit an offense violative of 18 U.S.C. § 111,1 that is, an assault upon a federal officer while engaged in the performance of his official duties.

Respondent Feola and three others (Alsondo, Rosa, and Farr) were indicted for violations of §§ 371 and 111. A jury found all four defendants guilty of both charges.2 Feola received a sentence of four years for the conspiracy and one of three years, plus a $3,000 fine, for the assault. The three-year sentence, however, was suspended and he was given three years' probation 'to commence at the expiration of confinement' for the conspiracy. The respective appeals of Feola, Alsondo, and Rosa were considered by the United States Court of Appeals for the Second Circuit in a single opinion. After an initial ruling partially to the contrary, that court affirmed the judgment of conviction on the substantive charges, but reversed the conspiracy convictions. United States v. Alsondo, 486 F.2d 1339, 1346 (1973).3 Because of a conflict among the federal Circuits on the scienter issue with respect to a conspiracy charge,4 we granted the Government's petition for a writ of certiorari in Feola's case.5 416 U.S. 935, 94 S.Ct. 1932, 40 L.Ed.2d 285 (1974).

* The facts reveal a classic narcotics 'rip-off.' The details are not particularly important for our present purposes. We need note only that the evidence shows that Feola and his confederates arranged for a sale of heroin to buyers who turned out to be undercover agents for the Bureau of Narcotics and Dangerous Drugs. The group planned to palm off on the purchasers, for a substantial sum, a form of sugar in place of heroin and, should that ruse fail, simply to surprise their unwitting buyers and relieve them of the cash they had brought along for payment. The plan failed when one agent, his suspicions being aroused,6 drew his revolver in time to counter an assault upon another agent from the rear. Instead of enjoying the rich benefits of a successful swindle, Feola and his associates found themselves charged, to their undoubted surprise, with conspiring to assault, and with assaulting, federal officers.

At the trial, the District Court, without objection from the defense, charged the jurors that, in order to find any of the defendants guilty on either the conspiracy count or the substantive one, they were not required to conclude that the defendants were aware that their quarry were federal officers.7

The Court of Appeals reversed the conspiracy convictions on a ground not advanced by any of the defendants. Although it approved the trial court's instructions to the jury on the substantive charge of assaulting a federal officer,8 it nonetheless concluded that the failure to charge that knowledge of the victim's official identity must be proved in order to convict on the conspiracy charge amounted to plain error. 486 F.2d, at 1344. The court perceived itself bound by a line of cases, commencing with Judge Learned Hand's opinion in United States v. Crimmins, 123 F.2d 271 (CA2 1941), all holding that scienter of a factual element that confers federal jurisdiction, while unnecessary for conviction of the substantive offense, is required in order to sustain a conviction for conspiracy to commit the substantive offense. Although the court noted that the Crimins rationale 'has been criticized,' 486 F.2d, at 1343, and, indeed, offered no argument in support of it, it accepted 'the controlling precedents somewhat reluctantly.' Id., at 1344.

The Government's plea is for symmetry. It urges that since criminal liability for the offense described in 18 U.S.C. § 111 does not depend on whether the assailant harbored the specific intent to assault a federal officer, no greater scienter requirement can be engrafted upon the conspiracy offense, which is merely an agreement to commit the act proscribed by § 111. Consideration of the Government's contention requires us preliminarily to pass upon its premise, the proposition that responsibility for assault upon a federal officer does not depend upon whether the assailant was aware of the official identity of his victim at the time he acted.

That the 'federal officer' requirement is anything other than jurisdictional9 is not seriously urged upon us; indeed, both Feola10 and the Court of Appeals, 486 F.2d at 1342, concede that scienter is not a necessary element of the substantive offense under § 111. Although some early cases were to the contrary,11 the concession recognizes what is now the practical unanimity of the Courts of Appeals.12 Nevertheless, we are not always guided by concessions of the parties, and the very considerations of symmetry urged by the Government suggest that we first turn our attention to the substantive offense.

The Court has considered § 111 before. In Ladner v. United States, 358 U.S. 169, 79 S.Ct. 209, 3 L.Ed.2d 199 (1958), the issue was whether a single shotgun blast which wounded two federal agents effected multiple assaults, within the meaning of 18 U.S.C. § 254 (1940 ed.), one of the statutory predecessors to the present § 111.13 The Government urged that s 254 had been intended not only to deter interference with federal law enforcement activities but, as well, to forestall injury to individual officers, as 'wards' of the United States. Given the latter formulation of legislative intent, argued the Government, a single blast wounding two officers would constitute two offenses. The Court disagreed because it found an equally plausible reading of the legislative intent to be that 'the congressional aim was to prevent hindrance to the execution of official duty . . . and was not to protect federal officers except as incident to that aim,' 358 U.S., at 175—176, 79 S.Ct. at 213. Under that view of legislative purpose, to have punishment depend upon the number of officers impeded would be incongruous. With no clear choice between these alternative formulations of congressional intent, in light of the statutory language and sparse legislative history, the Court applied a policy of lenity and, for purposes of the case, adopted the less harsh reading. Id., at 177—178, 79 S.Ct., at 213—214. It therefore held that the single discharge of a shotgun constituted only a single violation of § 254.

In the present case, we see again the possible consequences of an interpretation of § 111 that focuses on only one of the statute's apparent aims. If the primary purpose is to protect federal law enforcement personnel, that purpose could well be frustrated by the imposition of a strict scienter requirement. On the other hand, if § 111 is seen primarily as an anti-obstruction statute, it is likely that Congress intended criminal liability to be imposed only when a person acted with the specific intent to impede enforcement activities. Otherwise, it has been said: 'Were knowledge not required in obstruction of justice offenses described by these terms, wholly innocent (or even socially desirable) behavior could be transformed into a felony by the wholly fortuitous circumstance of the concealed identity of the person resisted.'14 Although we adhere to the conclusion in Ladner that either view of legislative intent is 'plausible,' we think it plain that Congress intended to protect both federal officers and federal functions, and that, indeed, furtherance of the one policy advances the other. The rejection of a strict scienter requirement is consistent with both purposes.

Section 111 has its origin in § 2 of the Act of May 18, 1934, c. 299, 48 Stat. 781. Section 1 of that Act, in which the present 18 U.S.C. § 1114 has its roots, made it a federal crime to kill certain federal law enforcement personnel while engaged in, or on account of, the performance of official duties,15 and § 2 forbade forcible resistance or interference with, or assault upon, any officer designated in § 1 while so engaged. The history of the 1934 Act, though scanty, offers insight into its multiple purposes. The pertinent committee reports consist, almost in their entirety, of a letter dated January 3, 1934, from Attorney General Cummings urging the passage of the legislation.16 In that letter the Attorney General states that this was needed 'for the protection of Federal officers and employees.' Compelled reliance upon state courts, 'however respectable and well disposed, for the protection of (federal) investigative and law-enforcement personnel' was inadequate, and there was need for resort to a federal forum.

Although the letter refers only to the need to protect federal personnel, Congress clearly was concerned with the safety of federal officers insofar as it was tied to the efficacy of law enforcement activities. This concern is implicit in the decision to list those officers protected rather than merely to forbid assault on any federal employee. Indeed, the statute as originally formulated would have prohibited attack on 'any civil official, inspector, agent, or other officer or employee of the United States.' See H.R.Rep.No.1455, 73d Cong., 2d Sess., 1 (1934). The House rejected this and insisted on the version that was ultimately enacted. Although the reason for the insistence is unexplained, it is fair to assume that the House was of the view that the bill as originally drafted strayed too far from the purpose of insuring the integrity of law enforcement pursuits.17

In resolving the question whether Congress intended to condition responsibility for violation of § 111 on the actor's awareness of the identity of his victim, we give weight to both purposes of the statute, but here again, as in Ladner, we need not make a choice between them. Rather, regardless of which purpose we would emphasize, we must take note of the means Congress chose for its achievement.

Attorney General Cummings, in his letter, emphasized the importance of providing a federal forum in which attacks upon named federal officers could be prosecuted. This, standing alone, would not indicate a congressional conclusion to dispense with a requirement of specific intent to assault a federal officer, for the locus of the forum does not of itself define the reach of the substantive offense. But the view that § 111 requires knowledge of the victim's office rests on the proposition that the reference to the federal forum was merely a shorthand expression of the need for a statute to fill a gap in the substantive law of the States. See United States v. Fernandez, 497 F.2d 730, 745 (CA9 1974) (concurring opinion), cert. pending, No. 73—6868. In that view § 111 is seen merely as a federal aggravated assault statute, necessary solely because some state laws mandate increased punishment only for assaults on state peace officers; assaults on federal personnel would be punishable, under state law, only for simple assault. As a federal aggravated assault statute, § 111 would be read as requiring the same degree of knowledge as its state-law counterparts. See Morissette v. United States, 342 U.S. 246, 263, 72 S.Ct. 240, 249, 96 L.Ed. 288 (1952). The argument fails, however, because it is fairly certain that Congress was not enacting § 111 as a federal counterpart to state proscriptions of aggravated assault.

The Attorney General's call for a federal forum in which to prosecute an attacker of a federal officer was directed at both sections of the proposed bill that became the 1934 Act. The letter concerned not only the section prohibiting assaults but also the section prohibiting killings. The latter, § 1, was not needed to fill a gap in existing substantive state law. The States proscribed murder, and, until recently, with the enactment of certain statutes in response to the successful attack on capital punishment, murder of a peace officer has not been deemed an aggravated form of murder, for all States usually have punished murderers with the most severe sanction the law allows. Clearly, then, Congress understood that it was not only filling one gap in state substantive law but in large part was duplicating state proscriptions in order to insure a federal forum for the trial of offenses involving federal officers. Fulfillment of the congressional goal to protect federal officers required then, as it does now, the highest possible degree of certainty that those who killed or assaulted federal officers were brought to justice. In the congressional mind, with the reliance upon the Attorney General's letter, certainty required that these cases be tried in the federal courts, for no matter how 'respectable and well disposed,' it would not be unreasonable to suppose that state officials would not always or necessarily share congressional feelings of urgency as to the necessity of prompt and vigorous prosecutions of those who violate the safety of the federal officer. From the days of prohibition to the days of the modern civil rights movement, the statutes federal agents have sworn to uphold and enforce have not always been popular in every corner of the Nation. Congress may well have concluded that § 111 was necessary in order to insure uniformly vigorous protection of federal personnel, including those engaged in locally unpopular activity.

We conclude, from all this, that in order to effectuate the congressional purpose of according maximum protection to federal officers by making prosecution for assaults upon them cognizable in the federal courts, § 111 cannot be construed as embodying an unexpressed requirement that an assailant be aware that his victim is a federal officer. All the statute requires is an intent to assault, not an intent to assault a federal officer. A contrary conclusion would give insufficient protection to the agent enforcing an unpopular law, and none to the agent acting under cover.18

This interpretation poses no risk of unfairness to defendants. It is no snare for the unsuspecting. Although the perpetrator of a narcotics 'rip-off,' such as the one involved here, may be surprised to find that his intended victim is a federal officer in civilian apparel, he nonetheless knows from the very outset that his planned course of conduct is wrongful. The situation is not one where legitimate conduct becomes unlawful solely because of the identity of the individual or agency affected. In a case of this kind the offender takes his victim as he finds him. The concept of criminal intent does not extend so far as to require that the actor understand not only the nature of his act but also its consequence for the choice of a judicial forum.

We are not to be understood as implying that the defendant's state of knowledge is never a relevant consideration under § 111. The statute does require a criminal intent, and there may well be circumstances in which ignorance of the official status of the person assaulted or resisted negates the very existence of mens rea. For example, where an officer fails to identify himself or his purpose, his conduct in certain circumstances might reasonably be interpreted as the unlawful use of force directed either at the defendant or his property. In a situation of that kind, one might be justified in exerting an element of resistance, and an honest mistake of fact would not be consistent with criminal intent.19

We hold, therefore, that in order to incur criminal liability under § 111 an actor must entertain merely the criminal intent to do the acts therein specified. We now consider whether the rule should be different where persons conspire to commit those acts.

Our decisions establish that in order to sustain a judgment of conviction on a charge of conspiracy to violate a federal statute, the Government must prove at least the degree of criminal intent necessary for the substantive offense itself. Ingram v. United States, 360 U.S. 672, 678, 79 S.Ct. 1314, 1319, 3 L.Ed.2d 1503 (1959). See Pettibone v. United States, 148 U.S. 197, 13 S.Ct. 542, 37 L.Ed. 419 (1893). Respondent Feola urges upon us the proposition that the Government must show a degree of criminal intent in the conspiracy count greater than is necessary to convict for the substantive offense; he urges that even though it is not necessary to show that he was aware of the official identity of his assaulted victims in order to find him guilty of assaulting federal officers, in violation of 18 U.S.C. § 111, the Government nonetheless must show that he was aware that his intended victims were undercover agents, if it is successfully to prosecute him for conspiring to assault federal agents. And the Court of Appeals held that the trial court's failure to charge the jury to this effect constituted plain error.

The general conspiracy statute, 18 U.S.C. § 371,20 offers no textual support for the proposition that to be guilty of conspiracy a defendant in effect must have known that his conduct violated federal law. The statute makes it unlawful simply to 'conspire . . . to commit any offense against the United States.' A natural reading of these words would be that since one can violate a criminal statute simply by engaging in the forbidden conduct, a conspiracy to commit that offense is nothing more than an agreement to engage in the prohibited conduct. Then where, as here, the substantive statute does not require that an assailant know the official status of his victim, there is nothing on the face of the conspiracy statute that would seem to require that those agreeing to the assault have a greater degree of knowledge.

We have been unable to find any decision of this Court that lends support to the respondent. On the contrary, at least two of our cases implicitly repudiate his position. The appellants in In re Coy, 127 U.S. 731, 8 S.Ct. 1263, 32 L.Ed. 274 (1888), were convicted of conspiring to induce state election officials to neglect their duty to safeguard ballots and election results. The offense occurred with respect to an election at which Indiana voters, in accordance with state law, voted for both local officials and members of Congress. Much like Feola here, those appellants asserted that they could not be punished for conspiring to violate federal law because they had intended only to affect the outcome of state races. In short, it was urged that the conspiracy statute embodied a requirement of specific intent to violate federal law. Id., at 753, 8 S.Ct. at 1269. The Court rejected this contention and held that the statute required only that the conspirators agree to participate in the prohibited conduct. See Anderson v. United States, 417 U.S. 211, 226, 94 S.Ct. 2253, 2263, 41 L.Ed.2d 20 (1974).

Similarly, in United States v. Freed, 401 U.S. 601, 91 S.Ct. 1112, 28 L.Ed.2d 356 (1971), we reversed the dismissal of an indictment charging defendants with possession of, and with conspiracy to possess, hand grenades that had not been registered, as reguired by 26 U.S.C. § 5861(d). The trial court dismissed the indictment for failure to allege that the defendants knew that the hand grenades in fact were unregistered. We held that actual knowledge that the grenades were unregistered was not an element of the substantive offense created by Congress and therefore upheld the indictment both as to the substantive offense and as to the charge of conspiracy. Again, we declined to require a greater degree of intent for conspiratorial responsibility than for responsibility for the underlying substantive offense.

With no support on the face of the general conspiracy statute or in this Court's decisions, respondent relies solely on the line of cases commencing with United States v. Crimmins, 123 F.2d 271 (CA2 1941), for the principle that the Government must prove'antifederal' intent in order to establish liability under § 371. In Crimmins, the defendant had been found guilty of conspiring to receive stolen bonds that had been transported in interstate commerce. Upon review, the Court of Appeals pointed out that the evidence failed to establish that Crimmins actually knew the stolen bonds had moved into the State. Accepting for the sake of argument the assumption that such knowledge was not necessary to sustain a conviction on the substantive offense, Judge Learned Hand nevertheless concluded that to permit conspiratorial liability where the conspirators were ignorant of the federal implications of their acts would be to enlarge their agreement beyond its terms as they understood them. He capsulized the distinction in what has become well known as his 'traffic light' analogy:

'While one may, for instance, be guilty of running past a traffic light of whose existence one is ignorant, one cannot be guilty of conspiring to run past such a light, for one cannot agree to run past a light unless one supposes that there is a light to run past.' Id., at 273.

Judge Hand's attractive, but perhaps seductive, analogy has received a mixed reception in the Courts of Appeals. The Second Circuit, of course, has followed it;21 others have rejected it.22 It appears that most have avoided it by the simple expedient of inferring the requisite knowledge from the scope of the conspiratorial venture.23 We conclude that the analogy, though effective prose, is, as applied to the facts before us, bad law.24

The question posed by the traffic light analogy is not before us, just as it was not before the Second Circuit in Crimmins. Criminal liability, of course, may be imposed on one who runs a traffic light regardless of whether he harbored the 'evil intent' of disobeying the light's command; whether he drove so recklessly as to be unable to perceive the light; whether, thinking he was observing all traffic rules, he simply failed to notice the light; or whether, having been reared elsewhere, he thought that the light was only an ornament. Traffic violations generally fall into that category of offenses that dispense with a mens rea requirement. See United States v. Dotterweich, 320 U.S. 277, 64 S.Ct. 134, 88 L.Ed. 48 (1943). These laws embody the social judgment that it is fair to punish one who intentionally engages in conduct that creates a risk to others, even though no risk is intended or the actor, through no fault of his own, is completely unaware of the existence of any risk. The traffic light analogy poses the question whether it is fair to punish parties to an agreement to engage intentionally in apparently innocent conduct where the unintended result of engaging in that conduct is the violation of a criminal statute.

But this case does not call upon us to answer this question, and we decline to do so, just as we have once before. United States v. Freed, 401 U.S., at 609 n. 14, 91 S.Ct. at 1118. We note in passing, however, that the analogy comes close to stating what has been known as the 'Powell doctrine,' originating in People v. Powell, 63 N.Y. 88 (1875), to the effect that a conspiracy, to be criminal, must be animated by a corrupt motive or a motive to do wrong. Under this principle, such a motive could be easily demonstrated if the underlying offense involved an act clearly wrongful in itself; but it had to be independently demonstrated if the acts agreed to were wrongful solely because of statutory proscription. See Note, Developments in the Law—Criminal Conspiracy, 72 Harv.L.Rev. 920, 936—937 (1959). Interestingly, Judge Hand himself was one of the more severe critics of the Powell doctrine.25

That Judge Hand should reject the Powell doctrine and then create the Crimmins doctrine seems curious enough. Fatal to the latter, however, is the fact that it was announced in a case to which it could not have been meant to apply. In Crimmins, the substantive offense, namely, the receipt of stolen securities that had been in interstate commerce, proscribed clearly wrongful conduct. Such conduct could not be engaged in without an intent to accomplish the forbidden result. So, too, it is with assault, the conduct forbidden by the substantive statute, § 111, presently before us. One may run a traffic light 'of whose existence one is ignorant,' but assaulting another 'of whose existence one is ignorant,' probably would require unearthly intervention. Thus, the traffic light analogy, even if it were a correct statement of the law, is inapt, for the conduct proscribed by the substantive offense, here assault, is not of the type outlawed without regard to the intent of the actor to accomplish the result that is made criminal. If the analogy has any vitality at all, it is to conduct of the latter variety; that, however, is a question we save for another day. We hold here only that where a substantive offense embodies only a requirement of mens rea as to each of its elements, the general federal conspiracy statute requires no more.

The Crimmins rule rests upon another foundation: that it is improper to find conspiratorial liability where the parties to the illicit agreement were not aware of the fact giving rise to federal jurisdiction, because the essence of conspiracy is agreement and persons cannot be punished for acts beyond the scope of their agreement. 123 F.2d, at 273. This 'reason' states little more than a conclusion, for it is clear that one may be guilty as a conspirator for acts the precise details of which one does not know at the time of the agreement. See Blumenthal v. United States, 332 U.S. 539, 557, 68 S.Ct. 248, 256, 92 L.Ed. 154 (1947). The question is not merely whether the official status of an assaulted victim was known to the parties at the time of their agreement, but whether the acts contemplated by the conspirators are to be deemed legally different from those actually performed solely because of the official identity of the victim. Put another way, does the identity of the proposed victim alter the legal character of the acts agreed to, or is it no more germane to the nature of those acts than the color of the victim's hair?

Our analysis of the substantive offense in Part II, supra, is sufficient to convince us that for the purpose of individual guilt or innocence, awareness of the official identity of the assault victim is irrelevant. We would expect the same to obtain with respect to the conspiracy offense unless one of the policies behind the imposition of conspiratorial liability is not served where the parties to the agreement are unaware that the intended target is a federal law enforcement official.

It is well settled that the law of conspiracy serves ends different from, and complementary to, those served by criminal prohibitions of the substantive offense. Because of this, consecutive sentences may be imposed for the conspiracy and for the underlying crime. Callanan v. United States, 364 U.S. 587, 81 S.Ct. 321, 5 L.Ed.2d 312 (1961); Pinkerton v. United States, 328 U.S. 640, 66 S.Ct. 1180, 90 L.Ed. 1489 (1946). Our decisions have identified two independent values served by the law of conspiracy. The first is protection of society from the dangers of concerted criminal activity, Callanan v. United States, supra, 364 U.S., at 593, 81 S.Ct., at 325; Dennis v. United States, 341 U.S. 494, 573 574, 71 S.Ct. 857, 899—900, 95 L.Ed. 1137 (1951) (Jackson, J., concurring). That individuals know that their planned joint venture violates federal as well as state law seems totally irrelevant to that purpose of conspiracy law which seeks to protect society from the dangers of concerted criminal activity. Given the level of criminal intent necessary to sustain conviction for the substantive offense, the act of agreement to commit the crime is no less opprobrious and no less dangerous because of the absence of knowledge of a fact unnecessary to the formation of criminal intent. Indeed, unless imposition of an 'antifederal' knowledge requirement serves social purposes external to the law of conspiracy of which we are unaware, its imposition here would serve only to make it more difficult to obtain convictions on charges of conspiracy, a policy with no apparent purpose.

The second aspect is that conspiracy is an inchoate crime. This is to say, that, although the law generally makes criminal only antisocial conduct, at some point in the continuum between preparation and consummation, the likelihood of a commission of an act is sufficiently great and the criminal intent sufficiently well formed to justify the intervention of the criminal law. See Note, Developments in the Law—Criminal Conspiracy, 72 Harv.L.Rev., at 923—925. The law of conspiracy identifies the agreement to engage in a criminal venture as an event of sufficient threat to social order to permit the imposition of criminal sanctions for the agreement alone, plus an overt act in pursuit of it, regardless of whether the crime agreed upon actually is committed. United States v. Bayer, 331 U.S. 532, 542, 67 S.Ct. 1394, 1399, 91 L.Ed. 1654 (1947). Criminal intent has crystallized, and the likelihood of actual, fulfilled commission warrants preventive action.

Again, we do not see how imposition of a strict 'anti-federal' scienter requirement would relate to this purpose of conspiracy law. Given the level of intent needed to carry out the substantive offense, we fail to see how the agreement is any less blameworthy or constitutes less of a danger to society solely because the participants are unaware which body of law they intend to violate. Therefore, we again conclude that imposition of a requirement of knowledge of those facts that serve only to establish federal jurisdiction would render it more difficult to serve the policy behind the law of conspiracy without serving any other apparent social policy.

We hold, then, that assault of a federal officer pursuant to an agreement to assault is not, even in the words of Judge Hand, 'beyond the reasonable intendment of the common understanding,' United States v. Crimmins, 123 F.2d, at 273. The agreement is not thereby enlarged, for knowledge of the official identity of the victim is irrelevant to the essential nature of the agreement, entrance into which is made criminal by the law of conspiracy.

Again we point out, however, that the state of knowledge of the parties to an agreement is not always irrelevant in a proceeding charging a violation of conspiracy law. First, the knowledge of the parties is relevant to the same issues and to the same extent as it may be for conviction of the substantive offense. Second, whether conspirators knew the official identity of their quarry may be important, in some cases, in establishing the existence of federal jurisdiction. The jurisdictional requirement is satisfied by the existence of facts tying the proscribed conduct to the area of federal concern delineated by the statute. Federal jurisdiction always exists where the substantive offense is committed in the manner therein described, that is, when a federal officer is attacked. Where, however, there is an unfulfilled agreement to assault, it must be established whether the agreement, standing alone, constituted a sufficient threat to the safety of a federal officer so as to give rise to federal jurisdiction. If the agreement calls for an attack on an individual specifically identified, either by name or by some unique characteristic, as the putative buyers in the present case, and that specifically identified individual is in fact a federal officer, the agreement may be fairly characterized as one calling for an assault upon a federal officer, even though the parties were unaware of the victim's actual identity and even though they would not have agreed to the assault had they known that identity. Where the object of the intended attack is not identified with sufficient specificity so as to give rise to the conclusion that had the attack been carried out the victim would have been a federal officer, it is impossible to assert that the mere act of agreement to assault poses a sufficient threat to federal personnel and functions so as to give rise to federal jurisdiction.

To summarize, with the exception of the infrequent situation in which reference to the knowledge of the parties to an illegal agreement is necessary to establish the existence of federal jurisdiction, we hold that where knowledge of the facts giving rise to federal jurisdiction is not necessary for conviction of a substantive offense embodying a mens rea requirement, such knowledge is equally irrelevant to questions of responsibility for conspiracy to commit that offense.

The judgment of the Court of Appeals with respect to the respondent's conspiracy conviction is reversed.

It is so ordered.

Judgment of Court of Appeals with respect to conspiracy conviction reversed.

Mr. Justice STEWART, with whom Mr. Justice DOUGLAS joins, dissenting.

Does an assault on a federal officer violate 18 U.S.C. § 1111 even when the assailant is unaware, and has no reason to know, that the victim is other than a private citizen or, indeed, a confederate in crime? This important question, never decided by the Court, is squarely presented in a petition for certiorari that has been pending here for many months: No. 73—6868, Fernandez v. United States.2 But this question was not contained in the petition for certiorari in the present case, and has not been addressed in either the briefs or oral arguments. The parties have merely assumed the answer to the question, and directed their attention to the separate question whether scienter is an element of conspiring to violate § 111. Nevertheless the Court sets out sua sponte to decide the basic question presented in Fernandez without the benefit of either briefing or oral argument by counsel.

This conspicuous disregard of the most basic principle of our adversary system of justice seems to me indefensible. Clearly, the petition for certiorari in Fernandez should have been granted, and that case decided after briefing and oral argument on its merits, before the subsidiary issue in the present case was considered. It is not too late to correct the serious judicial mistake the Court has made. We should grant certiorari in Fernandez now, and set the present case for rehearing after the argument in Fernandez has been had. But the Court rejects that course, and I perforce address the fundamental Fernandez question.

The Court recognizes that '(t)he question . . . is not whether the ('federal officer') requirement is jurisdictional, but whether it is jurisdictional only.' Ante, p. 677 n. 9. Put otherwise, the question is whether Congress intended to write an aggravated assault statute, analogous to the many state statutes which protect the persons and functions of state officers against assault, or whether Congress intended merely to federalize every assault which happens to have a federal officer as its victim. The Court chooses the latter interpretation, reading the federal-officer requirement to be jurisdictional only. This conclusion is inconsistent with the pertinent legislative history, the verbal structure of § 111, accepted canons of statutory construction, and the dictates of common sense.

Many States provide an aggravated penalty for assaults upon state law enforcement officers; typically the victim-status element transforms the assault from a misdemeanor to a felony.3 These statutes have a twofold purpose: to reflect the societal gravity associated with assaulting a public officer and, by providing an enhanced deterrent against such assault, to accord to public officers and their functions a protection greater than that which the law of assault otherwise provides to private citizens and their private activities.4 Consonant with these purposes, the accused's knowledge that his victim had an official status or function is invariably recognized by the States as an essential element of the aggravated offense.5 Where an assailant had no such knowledge, he could not of course be deterred by the statutory threat of enhanced punishment, and it makes no sense to regard the unknowing assault as being any more reprehensible, in a moral or retributive sense, than if the victim had been, as the assailant supposed, a private citizen.

The state statutes protect only state officers. I would read § 111 as filling the gap and supplying analogous protection for federal officers and their functions. An aggravated penalty should apply only where an assailant knew, or had reason to know, that his victim had some official status or function. It is immaterial whether the assailant knew the victim was employed by the federal, as opposed to a state or local, government. That is a matter of 'jurisdiction only,' for it does not affect the moral gravity of the act. If the victim was a federal officer, § 111 applies; if he was a state or local officer, an analogous state statute or local ordinance will generally apply. But where the assailant reasonably thought his victim a common citizen or, indeed, a confederate in crime, aggravation is simply out of place, and the case should be tried in the appropriate forum under the general law of assault, as are unknowing assaults on state officers.

The history of § 111 permits no doubt that this is an aggravated assault statute, requiring proof of scienter. The provision derives from a 1934 statute, 18 U.S.C. § 254 (1940 ed.), set out in the margin.6 The Attorney General proposed the statute in a letter to the Chairman of the Senate Committee on the Judiciary; the Attorney General's reasons are the only ones on record for the provision.7 The federal officers covered were listed in a companion provision, simultaneously enacted, proscribing the killing of federal officers.8 The present § 111 emerged from the 1948 recodification of Title 18,9 'with changes in phraseology and substance necessary to effect the consolidation' of the former § 254 with a minor 1909 statute proscribing assaults on officers of the 'Bureau of Animal Industry of the Department of Agriculture.'10 As the Court has recognized, the purport of the present § 111 must be derived from its major source, the 1934 enactment. See Ladner v. United States, 358 U.S. 169, 176 n. 4, 79 S.Ct. 209, 213, 3 L.Ed.2d 199.

Rummaging through the spare legislative history of the 1934 law, the Court manages to persuade itself that Congress intended to reach unknowing assaults on federal officers. Ante, pp. 679-684. But if that was the congressional intention, which I seriously doubt, it found no expression in the legislative product. The fact is that the 1934 statute expressly required scienter for an assault conviction. An assault on a federal officer was proscribed only if perpetrated 'on account of the performance of his official duties.' See n. 6, supra. That is, it was necessary not only that the assailant have notice that his victim possessed official status or duties but also that the assailant's motive be retaliation against the exercise of those duties.

It was not until the 1948 recodification that the proscription was expanded to cover assaults on federal officers 'while engaged in,' as well as 'on account of,' the performance of official duties. This was, as the Reviser observed, a technical alteration; it produced no instructive legislative history. See n. 10, supra. As presently written, the statute does clearly reach knowing assaults regardless of motive. But to suggest that it also reaches wholly unknowing assaults is to convert the 1948 alteration into one of major substantive importance, which it concededly was not.

The Court has also managed to convince itself that § 254 was not an aggravated assault statute. The surest evidence that § 254 was an aggravated assault statute may be found in its penalty provision.11 A single unarmed assault was made, and remains, punishable by a sentence of three years' imprisonment and a $5,000 fine. One need not make an exhaustive survey of state law to appreciate that this is a harsher penalty than is typically imposed for an unarmed assault on a private citizen. In 1934, federal law already defined and proscribed all varieties of assault occurring within the admiralty, maritime, and territorial jurisdiction of the United States: The penalty structure extended in graded steps, turning on the intent and methods of the assailant, from three months' to 20 years' imprisonment.12 If Congress had intended the victim-status element in § 254 to be 'jurisdictional only'—to provide merely another jurisdictional basis for trying assaults in the federal courts—there would have been no need to append a new and unique penalty provision to § 254. Instead, Congress could simply have made cross-reference to the pre-existing penalty structure for assaults within federal jurisdiction. This is not idle speculation. It was precisely the solution adopted, in the same 1934 Act, for the new offense of killing a federal officer: Congress provided that that new offense be defined and punished according to the pre-existing, graded, penalty structure for homicides within the maritime, admiralty, and territorial jurisdiction of the United States.13

This deliberated difference in definition and penalty treatment between the homicide and the assault statutes has an obvious significance. Congress gave to the new assault statute a unique and substantively novel definition and penalty. Unless we wish to assume that Congress was scatterbrained, we must conclude that it regarded the victim-status element as of substantive—and not merely jurisdictional—importance. That element was seen as an aggravating circumstance, just as is true in the state statutes, and not merely as a factor giving federal prosecutors and judges jurisdiction to deal with the offense.

The Court reasons otherwise. Positing that the victim-status element in the homicide statute is jurisdictional only, the Court concludes that the same must be true of the assault statute. Ante, pp. 683-684. Even assuming the premise, the conclusion does not follow. Quite apart from the radically different ways in which the two statutes provide for offense-definition and penalties, it requires little imagination to appreciate how Congress could regard the victimstatus element as 'jurisdictional only' in the homicide case but substantively significant in the assault case. The Court itself supplies a possible reason:

'(The homicide statute) was not needed to fill a gap in existing substantive state law. The States proscribed murder, and, until recently, with the enactment of certain statutes in response to the successful attack on capital punishment, murder of a peace officer has not been deemed an aggravated form of murder, for all States usually have punished murderers with the most severe sanction the law allows.' Ante, pp. 683.

In other words, the Court suggests that the widely perceived distinction, in morality and social policy, between assaults, depending upon the assailant's knowledge of the identity of the victim, found little or no echo in the law of homicide. From this, the natural conclusion—fortified by the penalty provisions—would be that Congress discriminated between the two statutes, recognizing the substantive distinction in the one and not in the other. For reasons I cannot fathom, the Court instead assumes that Congress was unable to discriminate in this fashion—that what had been self-evident to state legislatures was beyond the capacity of the National Legislature to comprehend. The Court says it cannot believe 'Congress silently chose to treat assaults and homicides differently. . . . (W)e have before us one bill with a single legislative history, and we decline to bifurcate our interpretation.' Ante, p. 685 n. 18. But it was Congress itself that 'bifurcated' the 1934 statute—by treating homicides and assaults differently as regards penalty and offense definition, and by proscribing only those assaults that were 'on account of the performance of official duties.' What the Court 'declines' to do is to read the statute that Congress wrote.

While the legislative history of the 1934 law is 'scant,' Ladner v. United States, 358 U.S., at 174, 79 S.Ct., at 212, it is sufficient to locate a congressional purpose consistent only with implication of a scienter requirement. As the Court said in Ladner: '(T)he congressional aim was to prevent hindrance to the execution of official duty, and thus to assure the carrying out of federal purposes and interests, and was not to protect federal officers except as incident to that aim.' Id., at 175—176, 79 S.Ct., at 213. This purpose is, of course, exactly analogous to the purposes supporting the state statutes which provide enhanced punishment for assault on state officers. A statute proscribing interference with official duty does not 'prevent hindrance' with that duty where the assailant thinks his victim is a mere private citizen, or indeed, a confederate in his criminal activity.

To avoid this self-evident proposition, the Court effectively overrules Ladner and concludes that the assault statute aims as much at protecting individual officers as it does at protecting the functions they execute. Ante, pp. 677-682. If the Ladner Court had shared this opinion, it would not have held, as it did, that a single shotgun blast wounding two federal agents was to be considered a single assault. But in any event, even today's revisionist treatment of Ladner does not succeed in getting the Court where it wants to go. So far as the scienter requirement is concerned, it makes no difference whether the statute aims to protect individuals, or functions, or both. The Court appears to think that extending § 111 to unknowing assaults will deter such assaults—will 'give . . . protection . . . to the agent acting under cover.' Ante, p. 684. This, of course, is nonsense. The federal statute 'protects' an officer from assault only when the assailant knows that the victim is an officer. Absent such knowledge, the only 'protection' is that provided by the general law of assault, for that is the only law which the potential assailant reasonably, if erroneously, believes applicable in the circumstances.

The Court also suggests that implication of a scienter requirement 'would give insufficient protection to the agent enforcing an unpopular law.' This is to repeat the same error. Whatever the 'popularity' of the laws he is executing, and whatever the construction placed on § 111, a federal officer is 'protected' from assault by that statute only where the assailant has some indication from the circumstances that his victim is other than a private citizen. Assuming, arguendo, that Congress thought that local prosecutors and judges were insufficiently enthusiastic about trying cases involving assaults on federal officers, it remains the fact that a federal statute proscribing knowing assaults meets this concern in every case where local attitudes might conceivably embolden the populace to interfere with federal officers enforcing an 'unpopular' law.

The fact is that there is absolutely no indication that before 1934 local prosecutors and judges were lax in trying cases involving assaults on federal officers, that Congress thought so, or—and this is the major point—that Congress was so obsessed by the esoteric 'problem' of unknowing assaults on officers who, if known, would be unpopular, as to enact a statute severely aggravated in penalty but blind to the commonsense distinction between knowing and unknowing assaults. The list of covered officers was long and varied in 1934; it has since become even more so.14 I can perceive no design to single out officers charged with the execution of 'unpopular' laws or given to using undercover techniques. The Attorney General's letter15 in support of the 1934 enactment disavowed any criticism of the integrity or good faith of local law enforcement authorities. He was at pains to stress that the 'Federal Government should not be compelled to rely upon the courts of the states, however respectable and well disposed . . ..' His particular concern was that '(i)n these cases resort must usually be had to the local police court, which affords but little relief to us, under the circumstances, in our effort to further the legitimate purposes of the Federal Government.' This is most reasonably read as a reference to the fact that, absent some statute aggravating the offense, assault was and is merely a misdemeanor—a 'police court' offense—in many States. To deal with this problem, the Attorney General sought enactment of a federal aggravated assault statute, Congress obliged, and this Court should give the statute its natural interpretation.

Turning from the history of the statute to its structure, the propriety of implying a scienter requirement becomes manifest. The statute proscribes not only assault but also a whole series of related acts. It applies to any person who 'forcibly assaults, resists, opposes, impedes, intimidates, or interferes with (a federal officer) . . . while engaged in or on account of the performance of his official duties.' (Emphasis added.) It can hardly be denied that the emphasized words imply a scienter requirement. Generally speaking, these acts are legal and moral wrongs only if the actor knows that his 'victim' enjoys a moral or legal privilege to detain him or order him about. These are terms of art, arising out of the common and statutory law proscribing obstruction of justice.16 Indeed, in urging enactment of § 254, the Attorney General referred to obstruction statutes, having either express or implied scienter requirements, as an instructive analogue.17 Whether it be express or implied, scienter has always been regarded in this country as an essential element of obstruction of justice. Pettibone v. United States, 148 U.S. 197, 204—207, 13 S.Ct. 542, 545—547, 37 L.Ed. 419. The sole innovation in § 111 is its protection of executive officers and functions, rather than judicial officers and functions. Obviously this distinction should have no effect on the scienter requirement.

If the words grouped in the statute with 'assaults' require scienter, it follows that scienter is also required for an assault conviction. One need hardly rely on such Latin Phrases as ejusdem generis and noscitur a sociis to reach this obvious conclusion. The Court suggests that assault may be treated differently, 'with no risk of unfairness,' because an assailant—unlike one who merely 'opposes' or 'resists'—'knows from the very outset that his planned course of conduct is wrongful' even though he 'may be surprised to find that his intended victim is a federal officer in civilian apparel.' Ante, p. 685. This argument will not do, either as a matter of statutory construction or as a matter of elementary justice.

The Court is saying that because all assaults are wrong, it is 'fair' to regard them all as equally wrong. This is a strange theory of justice. As the States recognize, an unknowing assault on an officer is less reprehensible than a knowing assault; to provide that the former may be punished as harshly as the latter is to create a very real 'risk of unfairness.' It is not unprecedented for Congress to enact stringent legislation, but today it is the Court that rewrites a statute so as to create an inequity which Congress itself had no intention of inflicting.

To treat assaults differently from the other acts associated with it in the statute is a pure exercise in judicial legislation. In Ladner v. United States, 358 U.S., at 176, 79 S.Ct., at 213, the Court noted that the 'Government frankly conceded on the oral argument that assault can be treated no differently from the other outlawed activities.' The Court characterized this concession as 'necessary in view of the lack of any indication that assault was to be treated differently, and in light of 18 U.S.C. § 111, the present recodification of § 254, which lumps assault in with the rest of the offensive actions,' id., at 176 n. 4, 79 S.Ct. 213. This analysis was not mere dictum but strictly necessary to the result reached in Ladner. No contrary analysis can be squared with the statutory history.18

The implication of scienter here is as necessary and proper as it was in Morissette v. United States, 342 U.S. 246, 72 S.Ct. 240, 96 L.Ed. 288. The Court there read a scienter requirement into a federal larceny statute over the government's objection that the need for scienter should not be implied for a federal offense when the statute that created the offense was silent on the subject. The Court said:

'Congressional silence as to mental elements in an Act merely adopting into federal statutory law a concept of crime already so well defined in common law and statutory interpretation by the states may warrant quite contrary inferences than the same silence in creating an offense new to general law, for whose definition the courts have no guidance except the Act. . . .

'. . . (W)here Congress borrows terms of art in which are accumulated the legal tradition and meaning of centuries of practice, it presumably knows and adopts the cluster of ideas that were attached to each borrowed word in the body of learning from which it was taken and the meaning its use will convey to the judicial mind unless otherwise instructed.' Id., at 262—263, 72 S.Ct. at 249—250.

The same principle applies here. The terms and purposes of § 111 flow from well-defined and familiar law proscribing obstructions of justice, and the provision complements a pattern of state aggravated assault statutes which are uniform and unambiguous in requiring scienter.

We see today the unfortunate consequences of deciding an important question without the benefit of the adversary process.19 In this rush to judgment, settled precedents, such as Ladner v. United States, supra, and Pettibone v. United States, supra, are subverted. Legislative history is ignored or imaginatively reconstructed. Statutory terms are broken from their context and given unnatural readings. On top of it all, the Court disregards two firmly established cannons of statutory construction—'two wise principles this Court has long followed':

'First, as we have recently reaffirmed, 'ambiguity concerning the ambit of criminal statutes should be resolved in favor of lenity.' Rewis v. United States, 401 U.S. 808, 812, 91 S.Ct. 1056, 1059, 28 L.Ed.2d 493 (1971). See also Ladner v. United States, 358 U.S. 169, 177, 79 S.Ct. 209, 213, 3 L.Ed.2d 199 (1958); Bell v. United States, 349 U.S. 81, 75 S.Ct. 620, 99 L.Ed. 905 (1955); United States v. Five Gambling Devices etc., 346 U.S. 441, 74 S.Ct. 190, 98 L.Ed. 179 (1953) (plurality opinion for affirmance) . . .

'. . . (S)econd . . .: unless Congress conveys its purpose clearly, it will not be deemed to have significantly changed the federal-state balance. Congress has traditionally been reluctant to define as a federal crime conduct readily denounced as criminal by the States. . . . In traditionally sensitive areas, such as legislation affecting the federal balance, the requirement of clear statement assures that the legislature has in fact faced, and intended to bring into issue, the critical matters involved in the judicial decision.' United States v. Bass, 404 U.S. 336, 347, 92 S.Ct. 515, 522, 30 L.Ed.2d 488.

If the Congress desires to sweep all assaults upon federal employees into the federal courts, a suitable statute could be easily enacted. I should hope that in so doing the Congress, like every State which has dealt with the matter, would make a distinction in penalty between an assailant who knows the official identity of the victim and one who does not. That result would have a double advantage over the result reached by the Court today. It would be a fair law, and it would be the product of the lawmaking branch of our Government.

For the reasons stated, I believe that before there can be a violation of 18 U.S.C. § 111, an assailant must know or have reason to know that the person he assaults is an officer. It follows a fortiori that there can be no criminal conspiracy to violate the statute in the absence of at least equivalent knowledge. Accordingly, I respectfully dissent from the opinion and judgment of the Court.