Full opinion text
MEMORANDUM OPINION ELLIS, District Judge. In this Espionage Act prosecution, defendants Steven Rosen and Keith Weiss-man have been charged in Count I of a superseding indictment with conspiring to transmit information relating to the national defense to those not entitled to receive it, in violation of 18 U.S.C. § 793(g). Defendants, by pretrial motion, attack the constitutionality of § 793 in three ways. First, they argue that the statute, as-applied to them, is unconstitutionally vague in violation of the Due Process Clause of the Fifth Amendment. Second, they argue that the statute, as-applied to them, abridges their First Amendment right to free speech and their First Amendment right to petition the government. Third, defendants assert the First Amendment rights of others by attacking the statute as facially overbroad. In the alternative, defendants urge the Court to avoid these constitutional issues by interpreting the statute as applying only to the transmission of tangible items, ie., documents, tapes, discs, maps and the like. In addition, defendant Rosen has been charged in Count III of the superseding indictment with aiding and abetting the transmission of information relating to the national defense to one not entitled to receive it, in violation of 18 U.S.C. § 793(d) and 2. He seeks dismissal of this count on the ground that the facts alleged in the superseding indictment in support of this count are legally insufficient. I. During the period of the conspiracy alleged in Count I, defendants Rosen and Weissman were employed by the American Israel Public Affairs Committee (AIPAC) in Washington, D.C. AIPAC is a pro-Israel organization that lobbies the United States executive and legislative branches on issues of interest to Israel, especially U.S. foreign policy with respect to the Middle East. Rosen was AIPAC’s Director of Foreign Policy Issues and was primarily engaged in lobbying officials of the executive branch with policy-making authority over issues of interest to AIPAC. Rosen did not have a security clearance during the period of the alleged conspiracy, and had not held a security clearance since his employment with the RAND Corporation in the late 1970s and early 1980s. Indeed, Rosen’s security clearance had been terminated on or about July 6, 1982. Defendant Weissman was AIPAC’s Senior Middle East Analyst and worked closely with Ro-sen in lobbying the executive branch of the U.S. government. Weissman has never held a security clearance. Alleged co-conspirator Lawrence Franklin worked on the Iran desk in the Office of the Secretary of the Department of Defense (DOD) and held a top secret security clearance during the alleged conspiracy. In general, the superseding indictment alleges that in furtherance of their lobbying activities, defendants (i) cultivated relationships with government officials with access to sensitive U.S. government information, including NDI, (ii) obtained the information from these officials, and (iii) transmitted the information to persons not otherwise entitled to receive it, including members of the media, foreign policy analysts, and officials of a foreign government. The government’s recitation of the acts constituting the conspiracy begins on April 13, 1999, when Rosen told an unnamed foreign official (FO-1) that he had “picked up an extremely sensitive piece of intelligence” which he described as “codeword protected intelligence.” Rosen proceeded to relate this piece of intelligence, which concerned terrorist activities in Central Asia, to the foreign official. Rosen and FO-1 continued this discussion over lunch a few weeks later. The superseding indictment alleges further that Weissman’s role in the conspiracy became apparent on June 11, 1999, when Weissman told the same foreign official that he had obtained a “secret FBI, classified FBI report” relating to the Khobar Towers bombing from three different sources, including a member of the United States government. Later that day, Weissman told FO-1 that he had interested a member of the media in the report. According to the superseding indictment, roughly eighteen months later, on December 12, 2000, Rosen and Weissman met with a United States government official (USGO-1) who had access to classified information relating to U.S. strategy pertaining to a certain Middle East country. Following this meeting, Rosen allegedly had a conversation with a member of the media in which he communicated classified information relating to the U.S. government’s deliberations on its strategy towards that particular Middle Eastern country. The next overt act in furtherance of the alleged conspiracy occurred over one year later, when, on January 18, 2002, Rosen met with another U.S. government official (USGO-2). After this meeting, Rosen prepared a memorandum referencing classified information provided by USGO-2 and distributed this memorandum to AIPAC staff. A few days later, Rosen relayed some of the information provided by USGO-2 to a foreign national. Rosen met again with USGO-2 on March 12, 2002 and discussed classified information regarding Al-Qaeda. Rosen allegedly disclosed this classified information to a fellow AIPAC employee the next day, and to another foreign embassy official (FO-2) the day after that. In August 2002, Rosen was introduced to Franklin through a contact at the DOD. The two agreed to meet on August 21, 2002, but the meeting was postponed. Ro-sen, Weissman, Franklin and another DOD employee finally met nearly six months later, on February 12, 2003. At this meeting, Franklin disclosed to Rosen and Weissman information relating to a classified draft internal United States government policy document concerning a certain Middle Eastern country. He told Ro-sen and Weissman that he had prepared a separate document based on the draft policy document. The three alleged co-conspirators met again on March 10, 2003 at Union Station in Washington, D.C. The three men conducted the meeting in successive restaurants and ended the meeting in an empty restaurant. Later that week, Rosen met with FO-2 and discussed the same draft internal policy document that Franklin had discussed with Rosen and Weissman. Both Rosen and Weissman had similar conversations with FO-1 later that same day. Rosen also called a senior fellow at a Washington, D.C. think tank and discussed the information concerning the government’s internal policy deliberations that had been provided by Franklin. A week after his meeting with Rosen and Weissman at Union Station, Franklin faxed to Rosen’s AIPAC office fax machine a document he had produced which contained information derived from the appendix of the U.S. draft internal policy document Franklin had discussed in his February meeting with Rosen and Weiss-man. The next day, Rosen discussed this information with a member of the media, prefacing his discussion with the statement, “I’m not supposed to know this.” Rosen had a similar discussion with another member of the media on May 30, 2003. In June 2003, Franklin, Rosen and Weissman arranged another lunch meeting. This meeting took place on June 26, 2003 at a restaurant in Arlington, Virginia. At the outset of the meeting Rosen told Franklin that he understood the difficult “constraints” under which Franklin was meeting, but notwithstanding these constraints, the three men proceeded to discuss the same draft internal policy document, as well as a newspaper article discussing the same classified document. The lunchtime discussion soon broadened to include internal United States policy deliberations, and at some point during the lunch, Franklin allegedly disclosed to Rosen and Weissman classified information relating to potential attacks on United States forces in Iraq. He told Rosen and Weissman that the information was “highly classified” and asked them not to use it. Later that day, Rosen described this information as “quite a story” and referring to Franklin, told Weissman “that this channel is one to keep wide open insofar as possible.” Consistent with this advice, Weissman took Franklin to a major league baseball game a few days later. At some point over the next year, Franklin was approached by law enforcement and he thereafter agreed to cooperate with the Federal Bureau of Investigation (FBI) in its investigation of Rosen and Weissman. On or about July 9, 2004 Weissman and Franklin, now acting as a cooperating witness, agreed to meet. At this meeting Franklin disclosed to Weiss-man NDI involving United States intelligence related to certain Middle Eastern countries. On July 21, 2004, Franklin again met with Weissman and allegedly disclosed to him classified national defense information concerning a foreign government’s covert actions in Iraq. Before disclosing the information, Franklin warned Weissman that the information he was about to receive was highly classified “Agency stuff’ and that Weissman could get into trouble by having the information. Following the meeting, Weissman returned to his office and related to Rosen what he had learned from Franklin. During the course of the day, Rosen and Weissman disclosed this information to another foreign official (FO-3) and a journalist, describing the information as “Agency information” and telling the journalist that the source of the information was “an American intelligence source” who was “100 percent credible.” Weissman also told a fellow AIPAC employee what he had learned earlier that day from Franklin. Nearly a month later, on August 20, 2004, Weissman again disclosed to a journalist the classified national defense information he had obtained from Franklin during their July 21, 2004 meeting. Within weeks of Weissman’s July 21, 2004 meeting with Franklin, the FBI contacted both Rosen and Weissman and asked them whether Franklin had ever disclosed classified information to either of them. Both Rosen and Weissman admitted knowing Franklin, but each denied that Franklin had ever disclosed classified information to them. After his interview, on August 27, 2004, Rosen contacted FO-2 and asked to meet with FO-2 and FO-3 to discuss a “serious matter.” Rosen also told FO-2 that the FBI had “made some allegations which are important” and added that he did not want to “discuss it on the phone” and did not want to go to FO-2’s embassy office. Accordingly, Rosen and FO-2 met later that day in a restaurant, and then proceeded to talk outside the restaurant where their conversation could not be monitored. These facts constitute the sum of Rosen’s and Weissman’s offense conduct as alleged in Count One of the superseding indictment. The superseding indictment also charges Rosen with aiding and abetting Franklin in the latter’s violation of 18 U.S.C. § 793(d). Specifically, Rosen is alleged to have aided and abetted Franklin’s March 17, 2003 transmission by fax of the document he had created from the classified draft internal policy document related to a certain Middle Eastern country. Rosen and Weissman have challenged the constitutionality of Count I of the superseding indictment on three separate but related grounds. First, the defendants argue that the government’s application of 18 U.S.C. § 793(e) in this prosecution violates the Fifth Amendment’s Due Process Clause under the vagueness doctrine because the statute’s indeterminate language failed to provide these defendants with adequate warning that their conduct was proscribed. In addition to this as-applied vagueness claim, defendants make two arguments based on the guarantees of the First Amendment. First, they argue that their conduct, as alleged in the superseding indictment, may not be proscribed without transgressing the First Amendment’s guarantees of free speech and the right to petition the government. Second, even assuming the statute’s constitutional application here, they raise a facial challenge to the statute pursuant to the First Amendment’s well-recognized overbreadth doctrine. Finally, in a separate motion to dismiss, Rosen challenges the sufficiency of the allegation that he aided and abetted Franklin’s violation of § 793(d). Each of these contentions is separately addressed. II. The operative statute at issue in defendant’s constitutional challenge is codified at 18 U.S.C. § 793 and provides, in pertinent part, as follows: (d) Whoever, lawfully having possession of, access to, control over, or being entrusted with any document, writing, code book, signal book, sketch, photograph, photographic negative, blueprint, plan, map, model, instrument, appliance, or note relating to the national defense, or information relating to the national defense which information the possessor has reason to believe could be used to the injury of the United States or to the advantage of any foreign nation, willfully communicates, delivers, transmits or causes to be communicated, delivered, or transmitted or attempts to communicate, deliver, transmit or cause to be communicated, delivered or transmitted the same to any person not entitled to receive it, or willfully retains the same and fails to deliver it on demand to the officer or employee of the United States entitled to receive it; or (e) Whoever having unauthorized possession of, access to, or control over any document, writing, code book, signal book, sketch, photograph, photographic negative, blueprint, plan, map, model, instrument, appliance, or note relating to the national defense, or information relating to the national defense which information the possessor has reason to believe could be used to the injury of the United States or to the advantage of any foreign nation, willfully communicates, delivers, transmits or causes to be communicated, delivered, or transmitted, or attempts to communicate, deliver, transmit or cause to be communicated, delivered, or transmitted the same to any person not entitled to receive it, or willfully retains the same and fails to deliver it to the officer or employee of the United States entitled to receive it.... Shall be fined under this title or imprisoned not more than ten years, or both. (g) If two or more persons conspire to violate any of the foregoing provisions of this section, and one or more of such persons do any act to effect the object of the conspiracy, each of the parties to such conspiracy shall be subject to the punishment provided for the offense which is the object of the conspiracy. 18 U.S.C. § 793. A brief history of this statute provides necessary context and helps illuminate the analysis of the questions presented. For much of this nation’s history, those who violated the nation’s trust by engaging in unauthorized disclosures of government secrets were prosecuted under generally applicable statutes punishing treason, unlawful entry into military bases, and theft of government property. See Harold Edgar and Benno C. Schmidt, Jr., The Espionage Statutes and Publication of Defense Information, 73 Colum. L.Rev. 929, 940 (1973) [hereinafter Espionage Statutes]. The first statute specifically intended to protect government secrets, and § 793’s progenitor, was the Defense Secrets Act of 1911. In terms that have survived largely unaltered for nearly a century, it prohibited the willful communication of knowledge concerning “anything connected with the national defense” to one “not entitled to receive it.” The statute did not define what was “connected to the national defense,” nor did it provide a clear basis for determining who was “entitled to receive” that knowledge. Notwithstanding these deficiencies, the drafters of the next legislative attempt to protect government secrets, which became known as the Espionage Act of 1917, were generally content to adopt the basic language of the 1911 statute. Thus, title I, section 1, subsection (d) of the Espionage Act provided that— whoever, lawfully or unlawfully having possession of, access to, control over, or being intrusted with any document, writing, code book, signal book, sketch, photograph, photographic negative, blue print, plan, map, model, instrument, appliance, or note relating to the national defense, wilfully communicates or transmits or attempts to communicate or transmit the same to any person not entitled to receive it, or willfully retains ttle same and fails to deliver it on demand to the officer or employee of the United States entitled to receive it .... shall be punished by a fine of not more than $10,000, or by imprisonment for not more than two years, or both. See 40 Stat. 217 (June 15, 1917). This provision, which is the precursor to both §§ 793(e) and (d), was altered in three material respects when Congress last amended the statute as part of the Internal Security Act of 1950. See 64 Stat. 987 (Sept. 23, 1950). First, Congress removed those with unlawful possession of NDI from the ambit of subsection (d), and created subsection (e) which focuses on this subset of persons. See id. Second, Congress expanded the category of what could not be communicated pursuant to §§ 793(d) and (e) to include “information relating to the national defense,” but modified this additional item by adding a scien-ter requirement to the effect that “the possessor has reason to believe [the information] could be used to the injury of the United States or to the advantage of any foreign nation.” See id. Finally, in contrast to subsection (d), Congress drafted subsection (e) to require one with unlawful possession of national defense information to return it to the government even in the absence of a demand for that information. See id.; see also New York Times Co. v. United States, 403 U.S. 713, 738 n. 9, 91 S.Ct. 2140, 29 L.Ed.2d 822 (1971) (White, J., concurring) (citing S.Rep. No. 81-2369, at 8-9 (1950)). Over the years, numerous commentators have criticized these provisions as excessively complex, confusing, indeed impenetrable. Yet, despite repeated calls for reform of these provisions in the more than half century since their last amendment in 1950, the statute has remained unchanged. Section 793’s litigation history is sparse, but nonetheless both pertinent and instructive. The modest number of reported decisions reflect that § 793 prosecutions are relatively rare and that over the years, the statute has successfully weathered several constitutional challenges on both vagueness and First Amendment grounds. While the Supreme Court has never considered a § 793(d) or (e) case, it has considered and rejected a vagueness challenge to the phrase “information relating to the national defense” as used in a related espionage statute. See Gorin v. United States, 312 U.S. 19, 61 S.Ct. 429, 85 L.Ed. 488 (1941). At the circuit level, authority is less sparse, but still relatively scarce. Particularly pertinent here is United States v. Morison, 844 F.2d 1057 (4th Cir.1988), in which the Fourth Circuit denied vagueness and First Amendment challenges to § 793 by a naval intelligence officer who transmitted classified satellite photographs of Soviet naval preparations to a British periodical. The Fourth Circuit has also considered and rejected vagueness challenges to § 793 and related espionage statutes in other cases. See United States v. Truong, 629 F.2d 908, 918-19 (4th Cir.1980) (rejecting vagueness challenge based on lack of evil intent in term willfulness); United States v. Dedeyan, 584 F.2d 36, 40 (4th Cir.1978) (rejecting vagueness and overbreadth challenges to the term “relating to the national defense” as used in § 793(f)); see also United States v. McGuinness, 35 M.J. 149, (C.M.A.1992) (rejecting a vagueness challenge to the term “unauthorized” as used in § 793(e)). Aware of these unsuccessful vagueness challenges to § 793, defendants attempt to distinguish their as-applied challenges by arguing that the instant prosecution is unprecedented in that it involves the alleged oral retransmission of information relating to the national defense, whereas other challenges to § 793 have involved the transmission of tangible items such as documents, or photographs. Indeed, a survey of the prosecutions under the modern version of § 793(e) discloses no prosecutions for the oral retransmission of information relating to the national defense. It is worth noting, however, that there have been prosecutions for the oral transmission of information relating to the national defense under 18 U.S.C. § 794 and its predecessor statutes, which prohibit the communication of information relating to the national defense to an agent of a foreign government. In addition, one case has been brought under both § 794(a) and § 793(d) for the oral transmission of national defense information to the Soviets. See United States v. Smith, 592 F.Supp. 424, 427 (E.D.Va.1984). Whether the fact that no person has been prosecuted under § 793(e) for the oral transmission of information relating to the national defense has any constitutional significance is addressed below. III. Before addressing defendants’ various constitutional challenges, it is first necessary to address the defendants’ statutory argument that the word “information” as used in § 793 should be construed as including only tangible information. This construction would preclude application of the statute to individuals who, like defendants, transmit NDI orally. If so construed, of course, § 793 would not reach the conduct alleged here and therefore obviate the need to address the defendants’ constitutional challenges. The phrase “information relating to the national defense” is not defined by the statute, and therefore, as with any issue of statutory interpretation, the appropriate place to begin the analysis is with the plain meaning of the statute’s words and the context in which they are used. See United States v. Groce, 398 F.3d 679, 681 (4th Cir.2005). The word “information” is a general term, the plain meaning of which encompasses knowledge derived both from tangible and intangible sources. See, e.g., The American Heritage College Dictionary 698 (1993) (defining information as “knowledge derived from study, experience, or instruction” and “knowledge of a specific event or situation; intelligence.”). Defendants do not dispute the plain meaning of the term “information,” but argue instead that plain meaning cannot control because construing the word “information” as encompassing intangible information renders the statute’s retention clauses absurd. And, it is well-established that the plain meaning of a term should not control if it leads to an absurd result. See Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 6, 120 S.Ct. 1942, 147 L.Ed.2d 1 (2000). In brief, the defendants’ argument proceeds as follows: The operative provisions in this prosecution, § 793(d) and (e), prohibit both (1) the communication of any document, note, map, etc., or information relating to the national defense to one not entitled to receive it, and (2) the willful retention of “the same.” While it is logical to punish the communication of intangible information, it is illogical to punish the retention of intangible information since a person cannot avoid remembering something he learned, thereby retaining it, nor can a person deliver their memory to one entitled to receive it, as the statute’s retention clauses would seem to require. Assuming, reasonably enough, that the two clauses apply to “the same” type of information, defendants assert that since it would be illogical to construe the term “information” as including intangible information in the retention clause, the communication clause must also apply only to tangible information. At the very least, defendants argue, this reflects that an ambiguity infects the statute, requiring the application of the canon of constitutional avoidance, the rule of lenity, and the related canons of ejusdem generis and nos-citur a sociis. While not without superficial appeal, this argument ultimately fails to persuade. A closer look at § 793’s history reveals that the absurdity identified by the defendants is a result of inadvertence and careless drafting, and not an indication that the drafters intended to restrict the prohibition of the first clause to tangible items. The grandfather of subsections (d) and (e) of § 793, the fourth clause of the Defense Secrets Act of 1911, clearly prohibited the willful communication of any “knowledge of anything connected with the national defense.” When this provision was reenacted as part of the Espionage Act of 1917, the drafters added the retention clause, and rather than again listing the items relating to the national defense that could not be retained, the drafters simply referred to those items listed in the first clause by using the phrase “the same.” In an apparent attempt to avoid the absurdity noted above, the drafters simply dropped the term “knowledge” from the list of items detailed in the first clause. See 40 Stat. 217 (1917). See generally Edgar and Schmidt, Espionage Statutes, 73 Colum. L.Rev. at 1012. Thus, had the defendants been charged under section 1(d) of the Espionage Act of 1917, their argument that the statute does not cover oral transmissions would be more persuasive. But in 1950, the Congressional drafters of the current provision, concerned over this potential loophole in the statute’s coverage, attempted to fix it by adding to the statute the phrase “information relating to the national defense which the possessor has reason to believe could be used to the injury of the United States or to the advantage of a foreign nation.” See generally Edgar and Schmidt, Espionage Statutes, 73 Colum. L.Rev. at 1021-31, 1050. This formulation was not new, but was derived from similar language in section 2 of the Espionage Act, the predecessor to 18 U.S.C. § 794. As used in that provision, the term “information related to the national defense” was understood to apply to information existing in both tangible and intangible form, and it is reasonable to conclude that the 1950 drafters intended to adopt the same meaning. As the Fourth Circuit has stated in relation to these two provisions, “[w]hen a statute is a part of a larger Act as these statutes are, the starting point for ascertaining legislative intent is to look to other sections of the Act in pari mateña with the statute under review.” United States v. Morison, 844 F.2d 1057, 1064 (4th Cir.1988). Indeed, this conclusion is buttressed by a statement of the district court in Morison, in which it stated that the statute— defines all types of tangibles: “any document, writing, code book, signal book, sketch, photograph, photographic negative, blueprint, plan, map, model, instrument, appliance, or note relating to the national defense,” and also describes intangibles: “information relating to the national defense which information the possessor has reason to believe could be used to the injury of the United States or to the advantage of any foreign nation.” See United States v. Morison, 622 F.Supp. 1009, 1011 (D.Md.1985) (emphasis added); see also Edgar and Schmidt, Espionage Statutes, 73 Colum. L.Rev. at 1021. Because construing the term “information” as including both tangible and intangible information is consistent with the plain meaning of the term and supported by the legislative history, it is not necessary to resort to the canon of constitutional avoidance or to the rule of lenity, which both apply only when choosing between two equally plausible interpretations. See Clark, 543 U.S. at 385, 125 S.Ct. 716; Pasquantino, 544 U.S. at 383, 125 S.Ct. 1766. Nor is application of the canons of ejusdem generis or noscitur a sociis appropriate, since it is clear that Congress’s intent in amending the statute in 1950 was to plug the loophole created when the term “knowledge” was dropped from the Espionage Act. Thus, because the word “information” as used in the first clause of the statute applies both to tangible and intangible information, and because defendants are not charged under the second clause, the absurdity pointed out by the defendants is of no consequence to the present prosecution. For this reason, it is necessary to address the defendants’ constitutional challenges. IY. Defendants’ first constitutional challenge to the statute is based on the principle that the Due Process clause of the Fifth Amendment prohibits punishment pursuant to a statute so vague that “men of common intelligence must necessarily guess at its meaning and differ as to its application.” United States v. Lanier, 520 U.S. 259, 266, 117 S.Ct. 1219, 137 L.Ed.2d 432 (1997) (quoting Connally v. General Constr. Co., 269 U.S. 385, 391, 46 S.Ct. 126, 70 L.Ed. 322 (1926)). Specifically, defendants allege that, as-applied to them, both §§ 793(d) and (e) are fatally vague with respect to determining: (1) the content of information covered by the phrase “information relating to the national defense,” and (2) the individuals “not entitled to receive” that information. The vagueness doctrine is premised on the principle that due process of law requires the government to provide potential defendants fair warning that their conduct may be proscribed, and is further animated by the concern that vague statutes may encourage arbitrary and discriminatory enforcement. See City of Chicago v. Morales, 527 U.S. 41, 56, 119 S.Ct. 1849, 144 L.Ed.2d 67 (1999). Thus, to take a well known example of a statute that failed to provide adequate notice, the Supreme Court found unconstitutionally vague a statute making it “unlawful for any person willfully ... to make any unjust or unreasonable rate or charge in handling or dealing in or with any necessaries” because it failed to provide an “ascertainable standard of guilt ... adequate to inform persons accused of violation thereof of the nature and cause of the accusation against them.” United States v. L. Cohen Grocery Co., 255 U.S. 81, 90, 41 S.Ct. 298, 65 L.Ed. 516 (1921). The Supreme Court relied on the vagueness doctrine’s second rationale in Kolender v. Lawson, 461 U.S. 352, 103 S.Ct. 1855, 75 L.Ed.2d 903 (1983), where it struck down a penal statute requiring a person stopped for loitering to provide police officers with “credible and reliable identification” because it failed to “establish minimal guidelines to govern law enforcement” and therefore “furnishes a convenient tool for harsh and discriminatory enforcement by local prosecuting officials against particular groups deemed to merit their displeasure.” Id. at 360, 103 S.Ct. 1855 (internal quotations and citations omitted). See also Morales, 527 U.S. at 60, 119 S.Ct. 1849. (“The broad sweep of the ordinance also violates ‘the requirement that a legislature establish minimal guidelines to govern law enforcement.’ ”) (quoting Kolender, 461 U.S. at 358, 103 S.Ct. 1855). While acknowledging these general principles, courts applying the vagueness doctrine also recognize that the language of any statute will possess some level of indeterminacy, and therefore courts sensibly do not require the scope of a criminal statute to be defined with perfect precision and clarity. In the words of the Fourth Circuit, It is sufficient ... to satisfy requirements of “reasonable certainty,” that while “the prohibitions of a statute may not satisfy those intent on finding fault at any cost, they are set out in terms that the ordinary person exercising ordinary common sense can sufficiently understand and comply with, without sacrifice to the public interest.” United States v. Morison, 844 F.2d 1057, 1071 (4th Cir.1988) (quoting Arnett v. Kennedy, 416 U.S. 134, 159, 94 S.Ct. 1633, 40 L.Ed.2d 15 (1974)). It is also well established that “clarity at the requisite level may be supplied by judicial gloss on an otherwise uncertain statute.” Lanier, 520 U.S. at 266, 117 S.Ct. 1219; see also Morison, 844 F.2d at 1071 (“all vagueness may be corrected by judicial construction which narrows the sweep of the statute within the range of reasonable certainty.”). In sum, courts considering vagueness challenges require that criminal statutes “either standing alone or as construed, make reasonably clear at the relevant time that the defendant’s conduct was criminal.” Lanier, 520 U.S. at 266, 117 S.Ct. 1219. Finally, and especially pertinent to the present challenge, there exists a generally recognized proposition that an otherwise unconstitutionally vague statute can survive a challenge if it contains a specific intent requirement. As the Supreme Court cogently put it: “[W]here the punishment imposed is only for an act knowingly done with the purpose of doing that which the statute prohibits, the accused cannot be said to suffer from lack of warning or knowledge that the act which he does is a violation of law.” Screws v. United States, 325 U.S. 91, 102, 65 S.Ct. 1031, 89 L.Ed. 1495 (1945). Thus, an otherwise intolerably vague statute may avoid a finding of unconstitutional vagueness if its application is contingent on the accused’s knowledge that he is committing an unlawful act. These principles govern defendants’ vagueness challenge. A. Defendants first argue that the content of the information described by the phrase “information relating to the national defense” is insufficiently clear when such information is transmitted orally. In this respect, it has long been recognized that the phrase “information relating to the national defense” is quite broad and potentially too broad since, especially in time of war, any information could conceivably relate to the national defense. See United States v. Heine, 151 F.2d 813, 815 (2d Cir.1945) (“It seems plain that the section cannot cover information about all those activities which become tributary to ‘the national defense’ in time of war; for in modern war there are none which do not.”). Courts, facing the obvious need to find some limiting construction, have not limited the phrase by specific subject matter, but instead have chosen to limit the phrase by requiring the government to prove (i) that the information is closely held by the government and (ii) that the information is the type of information that, if disclosed, could harm the United States. A review of the most pertinent case law interpreting and cabining the phrase is instructive. In Gorin v. United States, 312 U.S. 19, 61 S.Ct. 429, 85 L.Ed. 488 (1941), the Supreme Court rejected a similar vagueness challenge to identical language used in section 2(a) of the Espionage Act, currently codified at 18 U.S.C. § 794(a). In that case, defendant Gorin, a citizen of the U.S.S.R., had obtained from defendant Sa-lich, a naval intelligence officer, the substance of over fifty reports relating to Japanese activities in the United States, which the two had conspired to transmit to the Soviet Union. Id. at 22, 61 S.Ct. 429. The Supreme Court rejected an attempt by defendants to tie the term “information relating to the national defense” to information relating to those places listed in section 1(a) of the statute, currently codified at § 793(a), stating instead that the term “national defense” had acquired a well-known meaning “as a generic concept of broad connotations, referring to the military and naval establishments and the related activities of national preparedness.” Gorin, 312 U.S. at 28, 61 S.Ct. 429. Significantly, as the trial court in Gorin instructed the jury, the term encompasses the United States’ own intelligence reports about another nation’s military activities, “[f]or from the standpoint of military or naval strategy it might not only be dangerous to us for a foreign power to know our weaknesses and our limitations, but it might also be dangerous to us when such a foreign power knows that we know that they know of our limitations.” Id. at 31, 61 S.Ct. 429. The considerable breadth of the subject matter falling within the phrase “related to the national defense” has been confirmed in more recent cases. Thus, in United States v. Truong Dinh Hung, 629 F.2d 908 (4th Cir.1980), the Fourth Circuit rejected the defendants’ argument that “information relating to the national defense” was restricted to military matters, holding instead that the U.S. diplomatic cables relating to the 1977 Paris peace negotiations with the North Vietnamese were “related to the national defense” within the meaning of §§ 793 and 794. Truong, 629 F.2d at 918. In response to Truong’s argument that the material he transmitted was not covered by the phrase, the Fourth Circuit explained that “Congress intended ‘national defense’ to encompass a broad range of information and rejected attempts to narrow the reach of the statutory language.” Id. (citing Edgar and Schmidt, Espionage Statutes, 73 Colum. L.Rev. at 972-74). Similarly, in Morison, the Fourth Circuit approved the district court’s instruction to the jury describing “information relating to the national defense” as including “all matters that directly or may reasonably be connected with the national defense of the United States against any of its enemies. It refers to the military and naval establishments and the related activities of national preparedness.” Morison, 844 F.2d at 1071. In sum, the phrase “information relating to the national defense” has consistently been construed broadly to include information dealing with military matters and more generally with matters relating to United States foreign policy and intelligence capabilities. Rather than limiting the subject matter scope of the phrase “information relating to the national defense,” or restricting it to tangible material, courts have carefully cabined the phrase’s scope in two ways. First, courts have limited the term by requiring that the information be closely held by the government. This requirement was recognized by the Supreme Court in Gorin, and served as the basis for Judge Hand’s decision in United States v. Heine, 151 F.2d 813 (2d Cir.1945), in which he reversed Heine’s conviction under the predecessor to § 794 because the information about airplane production Heine delivered to the Germans was publicly available. See id. at 817. As Judge Hand put it, As declared in Gorin ... it is obviously lawful to transmit any information about weapons and munitions of war which the services had themselves made public; and if that be true, we can see no warrant for making a distinction between such information, and information which the services have never thought it necessary to withhold at all. Id. at 816. Similarly, the Fourth Circuit’s rejection of a vagueness challenge to the term “information relating to the national defense” in Morison, was based, in part, on the district judge’s instruction to the jury that “the government must prove that the documents or the photographs are closely held in that they have not been made public and are not available to the general public.” Morison, 844 F.2d at 1071-72. Contrary to the government’s assertion, the Fourth Circuit’s decision in United States v. Squillacote, 221 F.3d 542 (4th Cir.2000), does not compel the conclusion that information not officially disclosed, but in the public domain, retains its status as “information relating to the national defense.” In that case, the Fourth Circuit held that the government’s assessment of the reliability of publicly available information — as opposed to the information itself — might itself be information relating to the national defense. Id. at 578-79. As the Fourth Circuit put it: “That our government believes the estimates to be correct in and of itself is a fact that would be highly valuable to other countries.” Id. at 578. Because the disclosure of classified documents discloses the “government’s implicit stamp of correctness and accuracy,” the disclosure of official documents may violate the statute even if the information contained within the documents is publicly available. Id. Thus, it is the confirmation of the accuracy (or presumably also the inaccuracy) of material in the public domain, and not the public domain material itself, that a jury may consider to be “information relating to the national defense.” Because the instant case does not involve the disclosure of classified documents, this distinction will matter only if the defendants’ oral disclosure of information in the public domain included an official confirmation of what had previously been mere rumor or speculation. Further, although the confirmation of publicly available information relating to the national defense may at times constitute a disclosure of information relating to the national defense, this is not always the case, for as the Fourth Circuit has cautioned, “one may imagine situations in which information has been so widely circulated and is so generally believed to be true, that confirmation by one in a position to know would add nothing to its weight.” Knopf, 509 F.2d at 1370-71. As explained below, the distinction between a confirmation of information relating to the national defense already in the public domain that can be NDI and one that cannot depends on whether the confirmation may potentially harm the national security. In sum, information related to the national defense typically cannot qualify as such if it is in the public domain; it must be closely held by the government. Yet, in appropriate circumstances, this NDI can include the government’s closely held assessment or confirmation of certain public domain information. The second judicially imposed limitation on the phrase “information relating to the national defense” is the requirement that its “disclosure would be potentially damaging to the United States or useful to an enemy of the United States.” Morison, 844 F.2d at 1071-72. This important requirement is implicit in the purpose of the statute and assures that the government cannot abuse the statute by penalizing citizens for discussing information the government has no compelling reason to keep confidential. As the Supreme Court has instructed, the statute only applies to information for which there is an “occasion for secrecy,” and there is no “occasion for secrecy” unless disclosure of the information the government seeks to protect implicates an important government interest such as the national security. See Gorin, 312 U.S. at 28, 61 S.Ct. 429. This second NDI judicial gloss was explicitly relied upon in the concurring opinions of Judges Wilkinson and Phillips in Morison as necessary to save the statute from Morison’s First Amendment challenge. As Judge Wilkinson explained; The district court’s limiting instructions properly confine prosecution under the statute to disclosures of classified information potentially damaging to the military security of the United States. In this way the requirements of the vagueness and overbreadth doctrines restrain the possibility that the broad language of this statute would ever be used as a means of punishing mere criticism of incompetence and corruption in the government. Morison, 844 F.2d at 1084 (Wilkinson, J., concurring); see also id. at 1086 (Phillips, J., concurring) (“I agree that the hmiting instruction which required proof that the information leaked was either ‘potentially damaging to the United States or might be useful to an enemy sufficiently remedied [the statute’s vagueness and over-breadth].”). Thus, the phrase “information relating to the national defense,” while potentially quite broad, is limited and clarified by the requirements that the information be a government secret, i.e., that it is closely held by the government, and that the information is the type which, if disclosed, could threaten the national security of the United States. So cabined, the phrase “information relating to the national defense” avoids fatal vagueness and passes Due Process muster; given these two limitations the phrase provides fair notice of what it encompasses and is also an adequate safeguard against arbitrary enforcement. B. Defendants also argue that they lacked constitutionally adequate notice as to who was “entitled to receive” the national defense information, especially given the fact that the information was transmitted orally and therefore possessed no markings of “SECRET,” “CONFIDENTIAL” or other indicia typical of classified material. It is true that the statute itself provides no definition of the phrase “entitled to receive,” nor does it expressly delegate to the executive branch the authority to determine who is entitled to receive national defense information. Yet, this is not the end of the analysis. The Fourth Circuit addressed a similar challenge in Morison, and held that drafters of the Espionage Act correctly assumed that the President had the inherent authority to limit the communication of information relating to the national defense and that these preexisting rules and regulations of the Executive Branch would determine who is entitled to receive NDI. Morison, 844 F.2d at 1065-66. Indeed, during the debates over the passage of the Espionage Act in 1917, Senator Sutherland observed that “the phrase ‘lawfully entitled’ means nothing more and nothing less than that the particular information must have been forbidden not necessarily by an act of Congress; because in dealing with military matters the President has very great powers.” Id. (quoting 54 Cong.Rec. 3489). In other words, Congressional drafters viewed the phrase “entitled to receive” as an unfilled vessel into which the Executive Branch could pour more detailed content consistent with the phrase’s plain meaning and the statute’s purpose. Precisely this occurred. ■ In the instant case, as in Morison, the rule regulating who is “entitled to receive” is the Executive Order setting forth a uniform classification system for national security information. See Exec.Order No. 13,292, 68 Fed.Reg. 15,315 (March 25, 2003) amending Exee.Order No. 12,958, 60 Fed.Reg. 19,825 (April 17, 1995). The current classification system provides for the classification of information into one of three categories — Top Secret, Secret, and Classified — depending on the harm to the United States that would result from the information’s disclosure, and restricts ae-cess to classified information to those with a corresponding security clearance and a need-to-know. Id. at 15,315-16, 15,324. The classification system also acknowledges that classified information may be disseminated beyond the executive branch by those with authority to do so. See id. at 1325. Thus, while the language of the statute, by itself, may lack precision, the gloss of judicial precedent has clarified that the statute incorporates the executive branch’s classification regulations, which provide the requisite constitutional clarity. C. Defendants seek to distinguish this clear precedent clarifying the phrases “information relating to the national defense” and “entitled to receive” by arguing that because they received the information orally it was difficult to know at the time whether it was classified. And in this respect, although evidence that the information was classified is neither strictly necessary nor always sufficient to obtain a prosecution under § 793, the classification of the information by the executive branch is highly probative of whether the information at issue is “information relating to the national defense” and whether the person to whom they disclosed the information was “entitled to receive” the information. This is so because: (1) the subject matter of classified information must concern the national security or military preparedness of the nation, (2) access to classified information is restricted to a small number of people and accordingly is “closely held,” and (3) classified information must be the type of information the disclosure of which could damage the nation’s security. In addition, as noted, the classification status of information alleged to be related to the national defense governs who is “entitled to receive” the information. Citing these parallels, defendants argue that the difficulty in determining whether orally transmitted information is classified is highly relevant to whether the statute provides constitutionally adequate notice. A comparison of the application of the statute as-applied here to intangible information and the application of the statute in the typical § 793 prosecution to the delivery of classified documents (or any other tangible item) illustrates this point. All classified documents are clearly marked with a classification level and are often marked classified or unclassified at the paragraph level. For this reason, a person possessing such a document can easily determine: (i) whether the possession is authorized, (ii) which portions of the information the government is attempting to keep secret, and (iii) who else is entitled to receive the document. In contrast, a conversation about classified information, even one accompanied by a generic warning that “this information is classified,” is not likely to apprise the listener of precisely which portions of the information transmitted in the conversation are classified, or whether a more general description of the information retains its classification status such that it is sufficiently closely held and potentially damaging to the United States to violate the statute. Defendants argue that the difficulty in determining the classification of intangible information renders the application of the statute to them unconstitutionally vague. In addition, the defendants argue that because they were not government employees familiar with the executive branch’s classification regulations, and because the classification regulations are not explicitly incorporated into the statute, they did not have a constitutionally sufficient basis for determining who is “entitled to receive” the information. See Morison, 844 F.2d at 1074 (describing Morison’s familiarity with the classification regulations and concluding that “certainly the phrase ‘not authorized to receive it’ was well understood by the defendant.”). Although defendants’ argument is not without some force, it is ultimately unpersuasive. It is true that the difficulty in determining orally transmitted information’s classification status is highly relevant to whether the defendants knew, as is alleged, that they were transmitting NDI to one not entitled to receive it, but the defendants’ attempt to convert this difficulty into a reason for finding the statute unconstitutionally vague must ultimately fail. This is so because the statute requires the government to prove the defendants “tuillfully committed the prohibited conduct, and this “willfulness” requirement “eliminat[es] any genuine risk of holding a person ‘criminally responsible for conduct which he could not reasonably understand to be proscribed.’ ” United States v. Hsu, 364 F.3d 192, 197 (4th Cir.2004) (quoting United States v. Sun, 278 F.3d 302, 309 (4th Cir.2002)). Indeed, the Fourth Circuit has relied on this specific intent requirement in rejecting similar claims of vagueness in Morison and Truong) Thus, the government in this case must prove beyond a reasonable doubt that the defendants knew the information was NDI, i.e., that the information was closely held by the United States and that disclosure of this information might potentially harm the United States, and that the persons to whom the defendants communicated the information were not entitled under the classification regulations to receive the information. Further the government must prove beyond a reasonable doubt that the defendants communicated the information they had received from their government sources with “a bad purpose either to disobey or to disregard the law.” Morison, 844 at 1071. It follows, therefore, that if the defendants, or either of them, were truly unaware that the information they are alleged to have received and disclosed was classified, or if they were truly ignorant of the classification scheme governing who is entitled to receive the information, they cannot be held to have violated the statute. Thus, while the factual distinctions pointed out by defendants undoubtedly affect the government’s burden, they do not render the statute unconstitutionally vague. In addition to proving that the defendants committed the prohibited acts “willfully,” the statute imposes an additional and significant scienter requirement when a person is accused of transmitting “information relating to the national defense.” Thus, the statute, as-applied to these defendants also requires the government to prove that such information was communicated with “reason to believe it could be used to the injury of the United States or to the advantage of any foreign nation.” 18 U.S.C. §§ 793(d), (e). This language accompanied Congress’s amendment of the statute in 1950 adding the term “information” back into the provisions’ list of enumerated items relating to the national defense, and it is clear from both the text and the legislative history that this additional scienter requirement applies only to the communication of intangible “information,” and is intended to heighten the government’s burden when defendants are accused of communicating intangible information. As has been noted, the statute’s “willfulness” requirement obligates the government to prove that the defendants knew that disclosing the NDI could threaten the nation’s security, and that it was illegal, but it leaves open the possibility that defendants could be convicted for these acts despite some salutary motive. For example, if a person transmitted classified documents relating to the national defense to a member of the media despite knowing that such an act was a violation of the statute, he could be convicted for “willfully” committing the prohibited acts even if he viewed the disclosure as an act of patriotism. By contrast, the “reason to believe” scienter requirement that accompanies disclosures of information, requires the government to demonstrate the likelihood of defendant’s bad faith purpose to either harm the United States or to aid a foreign government. In this sense, requiring the government to prove that “the possessor has reason to believe [the information relating to the national defense] could be used to the injury of the United States or to the advantage of any foreign nation” is not duplicative of the requirement that the government prove the defendant willfully disclose information that is potentially damaging to the United States because the latter concerns only the quality of the information, whereas the former relates to the intended (or recklessly disregarded) effect of the disclosure. This conclusion is buttressed by reference to the contemporary judicial construction of similar language as used in related statutes. In Gorin, the leading precedent interpreting the Espionage Act at the time of § 793’s last amendment in 1950, the Supreme Court rejected a vagueness challenge to the precursor to § 794(a), in part, on the basis of the “obvious delimiting words in the statute” requiring that the defendants act with “intent or reason to believe that the information to be obtained is to be used to the injury of the United States, or to the advantage of any foreign nation.” Gorin, 312 U.S. at 27-28, 61 S.Ct. 429. According to Justice Reed, this “phrase requires those prosecuted to have acted in bad faith.” Id. at 28, 61 S.Ct. 429. This is significant because when Congress amended §§ 793(d) and (e) to reintroduce the term “information” it chose to modify the term with essentially the same delimiting words relied upon by Gorin less than ten years earlier, presumably with the intent that prosecutions under these provisions require the same proof of bad faith. The Supreme Court relied on this language when it rejected Gorin’s vagueness challenge to the phrase “information relating to the national defense,” Gorin, 312 U.S. at 28, 61 S.Ct. 429, and this added scienter requirement is yet another ground for rejecting the defendants’ vagueness challenge here. In summary, any inherent vagueness in the terms “relating to the national defense” or “entitled to receive” as used in §§ 793(d) and (e) is cured through the judicial glosses that have been added to these phrases. To the extent that oral transmission of information relating to the national defense makes it more difficult for defendants to know whether they are violating the statute, the statute is not thereby rendered unconstitutionally vague because the statute permits conviction only of those who “willfully” commit the prohibited acts and do so with bad faith. So construed, both phrases pass Fifth Amendment muster; they are not unconstitutionally vague as applied to these defendants. D. Seeking to avoid this conclusion, defendants argue that notwithstanding the clarity of the statute’s language, the application of the statute to these defendants is so novel and unprecedented that it violates the fair warning prong of the vagueness doctrine. As explained supra, the constitutionally required clarity of a statute may be provided through the gloss of judicial interpretation, and it is precisely the judicial glosses on § 793 that save the statute from defendants’ vagueness challenge. The corollary of this principle is that “due process bars courts from applying a novel construction of a criminal statute to conduct that neither the statute nor any prior judicial decision has fairly disclosed to be within its scope.” Lanier, 520 U.S. at 266, 117 S.Ct. 1219. This principle is animated by the same concern for fair warning that animates the Constitution’s prohibition of ex post facto laws. As the Supreme Court has stated: [A]n unforeseeable judicial enlargement of a criminal statute, applied retroactively, operates precisely like an ex post facto law, such as Art. I, s. 10, of the Constitution forbids---- If a state legislature is barred by the Ex Post Facto Clause from passing such a law, it must follow that a State Supreme Court is barred by the Due Process Clause from achieving precisely the same result by judicial construction. Bouie v. City of Columbia, 378 U.S. 347, 353-54, 84 S.Ct. 1697, 12 L.Ed.2d 894 (1964) (prohibiting application of criminal trespass statute to sit-in demonstrators when neither the language of the statute nor past precedent gave adequate warnings that conduct was proscribed). Thus, the test under these precedents is whether the language and application of the statute has provided defendants adequate warning that their conduct was proscribed. Section 793, as-applied here, meets this test; its language and history provided adequate warning to these defendants that the statute proscribed the alleged conduct. Defendants argue that the present prosecution represents a novel construction of the statute which they could not have anticipated because “leaks” of classified information by non-governmental persons have never been prosecuted under this statute. The statute’s plain language rebuts this argument. It is clear from this plain language that defendants’ conduct, as alleged in the superseding indictment is within the sweep of the statute. This is in sharp contrast to the statute in Bouie, which “on its face is narrow and precise,” lulling “the potential defendant into a false sense of security, giving him no reason even to suspect that conduct clearly outside the scope of the statute as written will be retroactively brought within it by an act of judicial construction.” Bouie, 378 U.S. at 352, 84 S.Ct. 1697. The same cannot fairly be said of § 793. It follows, that in contrast to the Bouie defendants, the defendants here cannot argue persuasively that the result reached here amounts to an unforeseeable judicial enlargement of § 793. And, it is useful in this regard to address defendants’ frequent use of the term “leak” in advancing their argument that there was not constitutionally adequate notice that the statute reached the alleged conduct. The term “leak,” at bottom, connotes in this