Eliminating Public Disclosures of Government Information from the Reach of the Espionage Act *

Transcription

1 Eliminating Public Disclosures of Government Information from the Reach of the Espionage Act * I. Introduction A government must protect its citizens from other nations or individuals who would inflict harm upon them. One element of accomplishing this goal is the prevention of disclosures of secret information that could endanger national security. The government can do this by restricting access to information on the front end, through its extensive and complex classification system. Alternatively, in the event that such information reaches the hands of someone not authorized to receive it, 1 the government may attempt to deter the disclosure of such information by criminally prosecuting the individual for disclosing it, or just for failing to promptly return it. 2 The latter strategy poses a significant threat to a democratic society. Although the government certainly has the duty to protect the security of the nation, one of the most fundamental and important constitutional values in the United States is the right of free speech. 3 Particularly valued under the protections of the First Amendment is the discussion of issues relating to the government. 4 Public discourse about government activity can serve as a principal check on executive power in the context of national security and international affairs. 5 One could very plausibly argue that no topic of public discourse could be more important and worthy of protection than the security of the nation. Thus, any statute purporting to criminalize speech relating to * I am deeply indebted to Professor David Anderson for alerting me to this topic and providing invaluable guidance throughout the early stages of this project, and to Professor David Rabban for his interest and comments. I also thank the remarkably talented editorial board and staff members of the Texas Law Review, particularly Zack Beasley, David Bernard, Kalani Hawks, and Samantha Porphy, for their tireless editing efforts and acute insights that considerably improved this work. Finally and most importantly, I am immeasurably grateful to my family, whose love and support I will cherish always. 1. This Note is concerned with disclosures by nongovernment employees; those who work for the government are subject to a distinct provision of the Espionage Act of 1917, 18 U.S.C. 793(d) (2000) and likely also to secrecy agreements with their government employers. 2. The primary focus of this Note is the statutory authority that currently gives the government the power to prosecute such conduct: 793(e) of the Espionage Act. See 18 U.S.C. 793(e). 3. See U.S. CONST. amend. I ( Congress shall make no law... abridging the freedom of speech.... ). 4. See, e.g., Mills v. Alabama, 384 U.S. 214, 218 (1966) ( Whatever differences may exist about interpretations of the First Amendment, there is practically universal agreement that a major purpose of that Amendment was to protect the free discussion of governmental affairs. ). 5. See N.Y. Times Co. v. United States, 403 U.S. 713, 728 (1971) (per curiam) (Stewart, J., concurring) ( [T]he only effective restraint upon executive policy and power in the areas of national defense and international affairs may lie in an enlightened citizenry in an informed and critical public opinion which alone can here protect the values of democratic government. ).

2 890 Texas Law Review [Vol. 86:889 government affairs, especially those involving national security, must be carefully scrutinized. Such a statute has existed in the government s legal arsenal for over fifty years in its present form 793(e) of the Espionage Act, which provides in pertinent part: Whoever having unauthorized possession of, access to, or control over any... 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... 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... [s]hall be fined under this title or imprisoned not more than ten years, or both. 6 Notably, this provision is broad enough to allow the government to prosecute private citizens for disclosing or even retaining 7 information somehow relating to the national defense. 8 Journalists, commentators, and Supreme Court Justices have recognized that the breadth of 793(e) could allow the government to prosecute members of the media for publishing information about U.S. foreign policy and the national defense. 9 In fact, 6. Espionage Act of 1917, 18 U.S.C. 793(e) (f). The Espionage Act was last amended in 1950, adding 793 as currently written. Internal Security Act of 1950, ch. 1024, sec. 101, 793, 64 Stat. 987, (codified as amended at 18 U.S.C. 793). For an extensive survey of the legislative history surrounding the enactment of the Espionage Act, see Harold Edgar & Benno C. Schmidt, Jr., The Espionage Statutes and Publication of Defense Information, 73 COLUM. L. REV. 929 (1973). 7. This Note will not address the prohibition against the retention of national security information but rather will focus only on attempts to punish the disclosure of that information. At least one of the arguments concerning the vagueness of the prohibition against mere retention was adequately presented (and rejected) in a recent case involving 793(e). See United States v. Rosen, 445 F. Supp. 2d 602, 615 (E.D. Va. 2006) (relating the defendants argument that the prohibition against retaining information relating to the national defense becomes an absurdity in the case of orally transmitted information because one cannot avoid remembering something he learned, thereby retaining it ), amended by No. 1:05cr225, 2006 WL (E.D. Va. 2006). Much of the analysis in this Note, however, turns on the benefits of disclosing such information so that the public may have access to it. See infra subpart II(E). Because those societal benefits would not be provided if the information were merely retained and not disclosed, different considerations would apply to an analysis of the retention provision U.S.C. 793(e). 9. See, e.g., N.Y. Times, 403 U.S. at 737, (White, J., concurring) (listing 793(e) as among those sections under which he would have no difficulty in sustaining convictions against the newspaper defendants in the case if appropriate facts were shown); Edgar & Schmidt, supra note 6, at 998 (pointing out that the scattergun drafting of 793(d) and (e) caused those subsections to pose a significant threat of prosecution to journalists who obtain or publish defense information); Editorial, Secrecy and the Media, L.A. TIMES, Aug. 27, 2006, at M4 (warning that the Espionage Act could mutate into a British-style Official Secrets Act that could be used against journalists ).

3 2008] Limiting the Reach of the Espionage Act 891 former Attorney General Alberto Gonzales admitted that the Department of Justice was considering prosecution of the New York Times for exposing the federal warrantless wiretapping program and the Washington Post for revealing the CIA s use of secret prisons in Eastern Europe to detain and interrogate persons of interest in the war on terror. 10 The fact that the language of 793(e) is broad enough to encompass these types of disclosures is unacceptable from both constitutional and policy perspectives. 11 Aside from actual prosecution, even the threat of prosecution alone is sufficient to create significant problems under First Amendment analysis. Ultimately, because 793(e) is a content-based restriction of speech, it must be the least restrictive means of achieving the compelling government interest in protecting national security and this Note will show that it does not satisfy this test. More specifically, the statute infringes on First Amendment values because its vague language does not give speakers sufficient notice as to which disclosures are protected and which can be prosecuted. This results in a chilling effect on free speech that the Supreme Court has repeatedly proscribed in its First Amendment jurisprudence. Additionally, its potential for overbroad application to disclosures that should properly be protected by the First Amendment makes it clear that 793(e) should fail the least-restrictive-means test. Finally, as a policy matter related to First Amendment values, the threat of prosecution will reduce the quality of public discourse about the government activities that are most important to citizens lives. The statute increases the power of the government to keep secrets because it serves as a weapon to threaten journalists who publish information the government may not want exposed, without regard to how dangerous the information really is. The public is thus prevented from having access to information that would foster truly meaningful debate about controversial issues of national policy and empower citizens to make fully informed decisions at the polls. This in turn leads to suboptimal decision making by officials elected by voters with suboptimal information at the polls. This Note shows why allowing 793(e) to permit prosecution of both covert and public disclosures of secret government information is unacceptable from both constitutional and policy standpoints. Specifically, the potential for overbroad application which stems from the vagueness of the 10. Geoffrey R. Stone, Scared of Scoops, N.Y. TIMES, May 8, 2006, at A21; see also Examining DOJ s Investigation of Journalists Who Publish Classified Information: Lessons from the Jack Anderson Case: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 5 (2006) [hereinafter Hearing on the Investigation of Journalists] (statement of Matthew Friedrich, Chief of Staff, Criminal Division, Department of Justice) (agreeing that it is the position of the Department of Justice today that [the Espionage Act] would... authorize the prosecution of a newspaper and a reporter ). Notably, both of these stories won Pulitzer Prizes. See Howard Kurtz, Post Wins 4 Pulitzer Prizes; 2 Go to New Orleans Paper, WASH. POST, Apr. 18, 2006, at A See infra Part II.

4 892 Texas Law Review [Vol. 86:889 statute s language makes clear that the statute is not the least restrictive means to achieve the compelling government interest in protecting national security. As written, 793(e) is unconstitutionally vague under a First Amendment analysis because its terms do not identify the specific disclosures that are prohibited by the law with sufficient precision. Correlatively, the statute is overbroad in that it allows the prosecution of disclosures that should properly be protected under the First Amendment. Specifically, by its terms, the statute can be used to prosecute both covert disclosures of secret information traditional espionage and disclosures through public outlets such as the media. The vast majority of disclosures of government secrets should only be prohibited when disclosed covertly not when disclosed in such a way as to be made available to the public at large. As will be seen, the public nature of such a disclosure increases its First Amendment value while at the same time decreasing its potential for harm to national security. Thus, removing public disclosures from the reach of 793(e) would make that statute serve as the least restrictive means of achieving the compelling government interest in preventing disclosures that would do real harm to national security. This Note proceeds as follows. Part II explains more fully why 793(e) is flawed from constitutional and policy perspectives. It begins with a comparison of three prosecutions under the Espionage Act and argues that the starkly different intuitive reactions produced by the three cases show why the present latitude afforded to the government is unacceptable. It then sets out the First Amendment doctrine under which 793(e) should be evaluated, concluding that it should receive strict scrutiny and the accompanying leastrestrictive-means analysis. After showing how one recent case has not applied the First Amendment analysis correctly when evaluating 793(e), this Part then enumerates several constitutional and policy rationales for why the statute should be amended to promote more disclosures and less government secrecy. Next, Part III briefly addresses several other possible means of limiting the reach of 793(e) but shows why each of those other methods is an inadequate solution to the problem. Part IV then sets out to show why exempting public disclosures from the reach of 793(e) is a critical and desirable first step in curing the problems with the Espionage Act. Subpart IV(A) explains how the public nature of a disclosure of secret government information both increases the First Amendment value of the disclosure and decreases its potential for harming national security. It then considers a few hypothetical disclosures to show why exempting public disclosures from prosecution strikes the correct balance between the government s and the public s interests. Finally, subpart IV(B) addresses at some length the significant and plausible objection that there are some disclosures that are so harmful that they should always be prohibited whether disclosed publicly or privately. While acknowledging these objections and the danger those disclosures may pose, this Note shows how those very dangerous disclosures

5 2008] Limiting the Reach of the Espionage Act 893 are addressed elsewhere in other statutes, making it unnecessary to provide for their prosecution under 793(e). II. The Problem with the Espionage Act A. A Three-Case Comparison A comparison of three government attempts to suppress speech related to national security illustrates the breadth of cases to which the Espionage Act can apply. In New York Times, 12 perhaps the most famous case dealing with this issue, the Supreme Court considered the government s effort to enjoin the New York Times and the Washington Post from publishing the Pentagon Papers. 13 The case represented the first time in U.S. history that the government initiated judicial proceedings against the media due to concerns over information related to national security. 14 Given the notoriety of the case, the facts will be related only briefly here. The so-called Pentagon Papers referred to a History of U.S. Decision- Making Process on Vietnam Policy, , commissioned by Secretary of Defense Robert McNamara. 15 The Pentagon Papers consisted of 7,000 pages of documents and other raw materials relating to America s activities in Vietnam, and they revealed illicit covert activities and deliberate withholding of pertinent information from the American public. 16 The New York Times published two installments of the Pentagon Papers before the Nixon Administration filed a complaint for an injunction against their publication. 17 When Judge Gurfein granted a temporary restraining order against the Times, the Washington Post began publishing its own copy of the report, and the government promptly sought to enjoin that publication as well. 18 After the district courts in both cases denied the injunctions, the cases were appealed on a remarkably expedited basis through the circuit courts and ultimately to the Supreme Court N.Y. Times Co. v. United States, 403 U.S. 713 (1971) (per curiam). 13. See id. 14. David Rudenstine, The Pentagon Papers Case: Recovering Its Meaning Twenty Years Later, 12 CARDOZO L. REV. 1869, 1869 (1991). 15. GEOFFREY R. STONE, PERILOUS TIMES: FREE SPEECH IN WARTIME 500 (2004). 16. Id. McNamara was quoted after reading the report as saying, You know,... they could hang people for what s in there. DAVID HALBERSTAM, THE BEST AND THE BRIGHTEST 633 (1972) (quoting Robert McNamara). 17. United States v. New York Times Co., 328 F. Supp. 324, 326 (S.D.N.Y. 1971), rev d, 403 U.S STONE, supra note 15, at , See Rudenstine, supra note 14, at 1903 (pointing out that in a period of fifteen days after the government filed suit, there were two evidentiary hearings at the district court level, two arguments at the circuit court level, a full briefing and oral argument at the Supreme Court level for the consolidated cases, and a public judgment issued by the Court).

6 894 Texas Law Review [Vol. 86:889 The Court s three-paragraph per curiam opinion, joined by six Justices, simply noted that the government faced an extremely high burden in seeking prior restraint and held that it had not met that burden in the case thus, it affirmed the denial of the injunctions. 20 Remarkably, though, in filing nine separate opinions, at least three Justices explicitly contemplated the possibility that the newspapers could be criminally prosecuted for publishing the Pentagon Papers under 793(e). 21 In fact, the government did initiate criminal proceedings against Daniel Ellsberg and Anthony Russo, the Rand Corporation employees who copied the documents they helped prepare and forwarded them to the newspapers; however, the district court dismissed the charges against them on grounds of prosecutorial misconduct and did not reach the merits of the case. 22 In United States v. Rosen, 23 another noteworthy and more recent case, a district court decided to allow the prosecution under 793(e) of two lobbyists for the American Israel Public Affairs Committee (AIPAC). 24 Steven Rosen, one of the defendants in the case, had been a prominent Washington lobbyist for over twenty years; the other defendant, Keith Weissman, was considered a top Middle East strategist. 25 The government alleged that Rosen and Weissman received information about American policy options in the Middle East, an FBI report about the Khobar Towers bombing in Saudi Arabia in 1996, and information about certain terrorist groups like al Qaeda. 26 They then orally relayed that information to their colleagues in AIPAC, an Israeli diplomat, and a reporter for the Washington Post. 27 Although the court s recitation of the facts in Rosen casts the events as an especially secretive and underhanded scheme, 28 in reality Rosen and 20. N.Y. Times Co., 403 U.S. at See id. at 733 (White, J., concurring) (cautioning, in an opinion joined by Justice Stewart, that just because the Government mistakenly chose to proceed by injunction does not mean that it could not successfully proceed [in a criminal proceeding under the Espionage Act] ); id. at 745 (Marshall, J., concurring) (noting that 793(e) appeared relevant to the case and the district judge s conclusion that it could not apply to newspaper publication was not... the only plausible construction that could be given ). 22. See generally Melville B. Nimmer, National Security Secrets v. Free Speech: The Issues Left Undecided in the Ellsberg Case, 26 STAN. L. REV. 311 (1974) (discussing the legal merits not reached by the judge in the case against Daniel Ellsberg) F. Supp. 2d 602 (E.D. Va. 2006), amended by No. 1:05cr225, 2006 WL (E.D. Va. 2006). 24. See id. 25. Richard B. Schmitt, Lobbyists to Stand Trial in Spy Case, L.A. TIMES, Aug. 11, 2006, at A Rosen, 445 F. Supp. 2d at Schmitt, supra note For instance, the court placed much emphasis in its factual statement on the level of classification attached to each of the relevant pieces of information that the defendants received and disclosed. See Rosen, 445 F. Supp. 2d at 608, (discussing the communication of codeword protected intelligence and classified information to the defendants). However, the classification status of the information bore only on whether the defendants were unauthorized recipients under 793(e), which was easy to show Rosen and Weissman were not employed by the government

7 2008] Limiting the Reach of the Espionage Act 895 Weismann essentially did what lobbyists typically do: they obtained information about government policy and used it to attempt to influence future policy decisions. Nevertheless, the court agreed that Rosen and Weismann could be prosecuted because their conduct was covered by the law and the statute was not unconstitutional under either the First or Fifth Amendments. 29 Despite its ultimate conclusion about the constitutionality of the statute, the court expressed a measure of discomfort with its use in cases like the one before it: The conclusion that the statute is constitutionally permissible does not reflect a judgment about whether Congress could strike a more appropriate balance between these competing interests, or whether a more carefully drawn statute could better serve both the national security and the value of public debate.... [T]he time is ripe for Congress to engage in a thorough review and revision of these provisions to ensure that they reflect both these changes, and contemporary views about the appropriate balance between our nation s security and our citizens ability to engage in public debate about the United States conduct in the society of nations. 30 The final useful case for comparison purposes is another recent Espionage Act prosecution in United States v. Squillacote. 31 The defendants in Squillacote were a husband and wife initially trained by the foreign intelligence arm of East Germany s intelligence service during the Cold War to conduct services on behalf of East Germany and other socialist countries. 32 Acting on the instructions of the East German intelligence service, Theresa Squillacote completed law school and obtained positions with the National Labor Relations Board, the House Armed Services Committee, and the Department of Defense, where she eventually became the Director of Legislative Affairs in the Office of the Undersecretary of Defense and received a security clearance with corresponding access to valuable information. 33 After the fall of the Berlin Wall, her coconspirators attempted to provide similar information to the KGB and later to the South African Communist Party. 34 Ultimately Squillacote was arrested after using her and did not have any security clearance that would permit them access to classified information. Id. at 608. The critical element related to the information itself under 793(e) had nothing to do with classification status, but rather only whether it was information relating to the national defense. See 18 U.S.C. 793(e) (2000); see also Nimmer, supra note 22, at 313 ( [C]ontrary to what may well be a popular misconception, the United States does not have an Official Secrets Act making it generally unlawful to disclose classified information. ). 29. Rosen, 445 F. Supp. 2d at Id. at F.3d 542 (4th Cir. 2000). 32. Id. at Id. at Id.

8 896 Texas Law Review [Vol. 86:889 position in the Department of Defense to obtain intelligence information and supply it to an undercover FBI agent she believed was a South African intelligence officer. 35 She and her husband were convicted of violating the Espionage Act, 36 sentenced to almost twenty-two years and seventeen-and-ahalf years, respectively, 37 and their convictions were affirmed on appeal. 38 The prosecutions in New York Times and Rosen might understandably strike a qualitatively different chord in the casual observer than the prosecution in Squillacote it might intuitively seem that the husband and wife in Squillacote were more deserving of punishment under the Espionage Act. After all, the newspapers in New York Times did no more than attempt to inform the public about government foreign policy, and the seasoned Washington advocates Rosen and Weismann probably did no more than what lobbyists typically do to influence government policy. 39 These are activities that are typically protected by the First Amendment. On the other hand, the Squillacote defendants engaged in systematic efforts to obtain positions from which they could covertly provide U.S. intelligence information to hostile foreign regimes that were potentially threatening to the United States. Little First Amendment value, if any, attaches to that type of conduct. Disturbingly, however, the law as written supplies no basis on which to accommodate this distinction. 40 Thus, to ensure that the First Amendment adequately protects and encourages the discussion of government affairs, it is imperative to make precise the reach of the statute by redefining what kinds of disclosures really ought to be prohibited. 35. Id. at Squillacote and her husband were not charged under 793(d), though their conduct would certainly have fallen under the reach of either. Rather, they were charged and convicted under, inter alia, 18 U.S.C. 794(a), (c) (2000) for transmitting national defense information directly to a foreign government. Squillacote, 221 F.3d at 552. The government probably chose this statute because it provided for up to a life sentence (or even death if certain aggravating factors were found). See 18 U.S.C. 794(a). 37. Brief for the United States in Opposition at 2, Squillacote v. United States, 532 U.S. 971 (2001) (No ), available at resp.pdf. 38. Squillacote, 221 F.3d at See Jeffrey Goldberg, Real Insiders: A Pro-Israel Lobby and an F.B.I. Sting, NEW YORKER, July 4, 2005, at 34, 40 (reporting the views of AIPAC critics who view the lobbyists actions as business as usual in Washington, where classified information is the currency of conversation with reporters and lobbyists ). 40. Specifically, in the aftermath of Rosen, commentators and journalists have expressed concerns that many of the means through which the public normally receives information about the government will be chilled by the ruling. See Recent Case, United States v. Rosen, 445 F. Supp. 2d 602 (E.D. Va. 2006), 120 HARV. L. REV. 821, 821 (2007) (noting that after Rosen, much uncertainty surrounds the government s right to criminally prosecute lobbyists, members of the press, and others who traffic in information); Schmitt, supra note 25 ( [The ruling in Rosen] could chill the ability of a broad segment of the public including lobbyists, academics and journalists to learn about the inner workings of government and expose misconduct or controversial programs of public interest. ).

9 2008] Limiting the Reach of the Espionage Act 897 B. The Relevant First Amendment Context The Applicable Level of Scrutiny Before proceeding further, it is important to determine precisely where disclosures of secret government information fit within the complex constellation of First Amendment law. Specifically, the applicable level of scrutiny to be applied to a law purporting to regulate the disclosure of government secrets must be determined as a threshold question before evaluating the constitutionality of the statute. The determination of what level of scrutiny applies to a statute challenged on First Amendment grounds depends on whether the law directly regulates expression or, while being directed primarily at non-speech-related conduct, it incidentally restricts some expressive activities. 41 Strict scrutiny will apply to any content-based regulation, while content-neutral regulations will enjoy a more lenient level of scrutiny. 42 Commentators have called the distinction between content-based and content-neutral regulations the central inquiry in free speech analysis because it determines the applicable level of scrutiny 43 and regulations subjected to strict scrutiny are almost never upheld. 44 Although determining whether a regulation is content based can be difficult, 45 the Supreme Court will find a regulation to be content based if it targets either a certain viewpoint or a particular subject matter. 46 If a law discriminates against a certain viewpoint or subject matter, it will be subject to strict scrutiny: The Government may... regulate the content of constitutionally protected speech in order to promote a compelling interest if it chooses the least restrictive means to further the articulated interest. 47 In the arena of content-neutral regulations, a line of cases beginning with United States v. O Brien 48 introduced an intermediate level of scrutiny to First Amendment jurisprudence under which the least-restrictive-means 41. Texas v. Johnson, 491 U.S. 397, (1989). 42. Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622, 642 (1994). 43. Erwin Chemerinsky, Content Neutrality as a Central Problem of Freedom of Speech: Problems in the Supreme Court s Application, 74 S. CAL. L. REV. 49, 51 (2000). 44. Eugene Volokh, Crime-Facilitating Speech, 57 STAN. L. REV. 1095, 1132 (2005). 45. Bartnicki v. Vopper, 532 U.S. 514, 526 (2001) (quoting Turner Broad. Sys., 512 U.S. at 642). 46. Chemerinsky, supra note 43, at 51; see also Boos v. Barry, 485 U.S. 312, 329 (1988) (holding that a statute forbidding protests against foreign governments within 500 feet of an embassy to be a content-based restriction on political speech in a public forum). 47. Sable Commc ns of Cal., Inc. v. FCC, 492 U.S. 115, 126 (1989); see also Ashcroft v. ACLU, 542 U.S. 656, 667 (2004) ( It is not an answer to say that COPA reaches some amount of materials that are harmful to minors; the question is whether it would reach more of them than less restrictive alternatives. ); Reno v. ACLU, 521 U.S. 844, 879 (1997) ( The breadth of this contentbased restriction of speech imposes an especially heavy burden on the Government to explain why a less restrictive provision would not be as effective as the CDA. ) U.S. 367 (1968).

10 898 Texas Law Review [Vol. 86:889 analysis does not apply. 49 O Brien considered a law prohibiting the destruction or mutilation of Selective Service registration cards as applied to a defendant who publicly burned his card as a sign of protest against the Vietnam War. 50 The defendant argued that the statute was unconstitutional because it abridged his freedom of speech under the First Amendment. 51 The Court stressed that the statute was directed only at the conduct of destroying a draft card; the restriction on the defendant s First Amendment rights was merely incidental. 52 In those circumstances, the Court held that such a law would survive First Amendment scrutiny if it furthered an important or substantial governmental interest and the incidental restriction on alleged First Amendment freedoms [was] no greater than [was] essential to the furtherance of that interest. 53 That test was further clarified (and relaxed) in a subsequent case: [A]n incidental burden on speech is no greater than is essential, and therefore is permissible under O Brien, so long as the neutral regulation promotes a substantial government interest that would be achieved less effectively absent the regulation. 54 Nevertheless, even conduct can sometimes be sufficiently imbued with elements of communication to implicate the First Amendment. 55 For instance, a defendant who ended a peace rally by burning an American flag was held to have engaged in expressive conduct because he intended to convey a message and that message was likely understood by those viewing it. 56 Because the regulation at issue was directed at those who desecrated the flag i.e., mistreated it in a degrading way that would likely offend others the Court noted that the regulation would only apply when the person burning the flag was doing so to communicate some particular message. 57 The regulation was thus ultimately a restriction on expressive conduct, and the O Brien test did not apply. 58 The ultimate question as to whether the regulation at issue will receive the lenient O Brien level of scrutiny is whether the asserted government interest is unrelated to expression See, e.g., Hill v. Colorado, 530 U.S. 703, 726 (2000) ( [W]hen a content-neutral regulation does not entirely foreclose any means of communication, it may satisfy the tailoring requirement even though it is not the least restrictive or least intrusive means of serving the statutory goal. ); Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 252 (1997) (O Connor, J., dissenting) ( [W]e have refrained from imposing a least-restrictive-means requirement in cases involving intermediate First Amendment scrutiny. ). 50. O Brien, 391 U.S. at Id. at Id. at Id. at United States v. Albertini, 472 U.S. 675, 689 (1985). 55. Spence v. Washington, 418 U.S. 405, 409 (1974) (per curiam). 56. Texas v. Johnson, 491 U.S. 397, (1989). 57. See id. at Id. at See id. at 407.

11 2008] Limiting the Reach of the Espionage Act 899 Subsection 793(e) of the Espionage Act should receive strict scrutiny. First, disclosures of such information constitute speech under the First Amendment, as opposed to the conduct regulated by the O Brien line of cases. The Supreme Court has previously held that disclosures of secret information, made with the intention that such information be made public, constitute pure speech under the First Amendment. 60 In Bartnicki v. Vopper, 61 the Court considered whether a radio station could be prosecuted for publicizing a telephone conversation that was illegally intercepted by an unknown third party and anonymously delivered to the station. 62 Although the Court held that the statute prohibiting any further dissemination of an illegally intercepted conversation 63 was content neutral (because the content of the intercepted conversations did not matter for purposes of the statute), it quoted with approval the lower court s statement that [i]f the acts of disclosing and publishing information do not constitute speech, it is hard to imagine what does fall within that category, as distinct from the category of expressive conduct. 64 Additionally, 793(e) discriminates against such speech on the basis of its subject matter. The relevant part of the statute for present purposes prohibits the disclosure of information relating to the national defense ; 65 thus, the content of the disclosure is the principal determinant of whether the disclosure can be prosecuted. An individual who discloses the security codes to gain entry into the Pentagon has almost certainly violated the statute, while the same person s disclosure of the kind of car the Secretary of Defense 60. Bartnicki v. Vopper, 532 U.S. 514, (2001) U.S. 514 (2001). 62. Id. at The Bartnicki Court noted that the case before it did not present the still-open question of whether a newspaper that unlawfully obtained information could be punished not only for the illegal acquisition but also for the disclosure of that information. Id. at 528. After ten years of litigation between two members of the House of Representatives, the D.C. Circuit recently rejected in a part of the dissenting opinion that commanded a majority in a 5 4 en banc decision the proposition that the First Amendment did not protect disclosures of information that the discloser knew was illegally obtained by someone else. Boehner v. McDermott, 484 F.3d 573, 586 (D.C. Cir. 2007) (en banc) (Sentelle, J., dissenting, writing for the majority as to Part I). An even more recent First Circuit case came to the same conclusion. See Jean v. Mass. State Police, 492 F.3d 24, 33 (1st Cir. 2007). It bears emphasis that the act of unlawfully acquiring national security information (e.g., stealing documents from the Pentagon) is conceptually distinct from the act of disclosing that information. See Branzburg v. Hayes, 408 U.S. 665, 691 (1972) ( Although stealing documents or private wiretapping could provide newsworthy information, neither reporter nor source is immune from conviction for such conduct, whatever the impact on the flow of news. ). This Note, like the Supreme Court, will defer the question of whether the initial unlawful acquisition of the information renders void any First Amendment protection of the disclosure of that information to another day. It will thus be assumed herein that the information has not been obtained unlawfully by the discloser of the information. 64. Bartnicki, 532 U.S. at 527 (alteration in original) (quoting Bartnicki v. Vopper, 200 F.3d 109, 120 (3d Cir. 1999), aff d, 532 U.S. 514) U.S.C. 793(e) (2000).

12 900 Texas Law Review [Vol. 86:889 drives likely has not regardless of how secret each bit of information initially might have been. The sole fact that the content of the former disclosure relates in some way to the national defense a particular subject matter brings it under the reach of the statute. Accordingly, because 793(e) is a content-based regulation of speech, strict scrutiny should apply in the evaluation of its constitutionality under the First Amendment. The application of strict scrutiny to a content-based regulation of speech gives rise to a heavy presumption that regulation is impermissible. 66 Any statute purporting to regulate speech protected by the First Amendment must use the least restrictive means of doing so, in furtherance of a compelling government interest. 67 Most importantly, the application of strict scrutiny, even when confronted with the compelling interest of preserving national security, does not permit a pure balancing of the competing interests in each particular case. Instead, the balancing must be done in advance without regard to the particular circumstances of each case to ensure that the law constitutes the least restrictive means of restricting speech. 68 Chief Justice Warren s refusal to engage in such ad hoc balancing warrants quoting at length: It has been suggested that this case should be decided by balancing the governmental interests... against the First Amendment rights asserted by the appellee. This we decline to do. We recognize that both interests are substantial, but we deem it inappropriate for this Court to label one as being more important or more substantial than the other. Our inquiry is more circumscribed. Faced with a clear conflict between a federal statute enacted in the interests of national security and an individual s exercise of his First Amendment rights, we have confined our analysis to whether Congress has adopted a constitutional means in achieving its concededly legitimate legislative goal.... We have ruled only that the Constitution requires that the conflict between congressional power and individual rights be accommodated by legislation drawn more narrowly to avoid the conflict Richard H. Fallon, Jr., Making Sense of Overbreadth, 100 YALE L.J. 853, 865 (1991). 67. See supra note 47 and accompanying text. 68. Professor Nimmer also advocated for a balancing-in-advance approach, without regard to the particular circumstances of a case, while rejecting any ad hoc balancing in the First Amendment context. See Melville B. Nimmer, The Right to Speak from Times to Time: First Amendment Theory Applied to Libel and Misapplied to Privacy, 56 CAL. L. REV. 935, 939 (1968) (asserting that an ad hoc balancing approach is particularly pernicious where speech is concerned because it tends to deter all but the most courageous (not necessarily the most rational) from entering the market place of ideas ). This Note proposes the most acceptable method of balancing in advance the competing interests behind 793(e) in a way that satisfies the least-restrictive-means analysis. 69. United States v. Robel, 389 U.S. 258, 268 n.20 (1967).

13 2008] Limiting the Reach of the Espionage Act 901 C. The Relevant First Amendment Context Vagueness and Overbreadth The proper application of strict scrutiny is especially important because it informs two concepts deeply ingrained in First Amendment jurisprudence: vagueness and overbreadth. The vagueness inquiry asks whether ordinary people can understand in advance whether their speech will be prohibited by a statute, 70 while the overbreadth doctrine inquires whether the statute s terms could criminalize speech properly protected under the First Amendment and thus deter other speakers from exercising their protected rights. 71 Conceptually, the two doctrines are distinct: an ordinance prohibiting any depiction of a naked person would be overbroad but not vague, while another ordinance prohibiting any unconstitutional depiction of a naked person would not be overbroad but would most likely be vague. However, the Supreme Court and various commentators have recognized that in the First Amendment context, these doctrines are highly interrelated. The Court has historically regarded the doctrines of vagueness and overbreadth as logically related and similar doctrines. 72 Professor Fallon has argued that vagueness is best viewed as a subpart of the overbreadth doctrine because they stem from the same basic rationale and one often gives rise to the other. 73 It is easy to imagine why this is the case: a vague statute may cause a conscientious individual to believe that certain speech, which should be protected, would fall under the reach of the statute and that suspicion may very well be correct on a reasonable reading of the statute, making it overbroad in its potential application. Similarly, a statute that would be overbroad if applied literally such as a statute forbidding any First Amendment activities inside an airport 74 would suffer from vagueness problems if held not to apply literally. In any event, the vagueness and overbreadth inquiries aid courts in determining whether a statute is the least restrictive means of achieving a compelling government interest. For this reason, and most importantly for purposes of the analysis of 793(e), the vagueness and overbreadth doctrines appear to enjoy a somewhat privileged status in First Amendment jurisprudence. For instance, outside the First Amendment context, a party whose conduct clearly falls under the reach of a statute cannot mount a facial challenge of vagueness or overbreadth against that statute. 75 On the other hand, 70. Kolender v. Lawson, 461 U.S. 352, 357 (1983). 71. Bd. of Airport Comm rs v. Jews for Jesus, Inc., 482 U.S. 569, 574 (1987). 72. Kolender, 461 U.S. at 358 n Fallon, supra note 66, at Jews for Jesus, 482 U.S. at Fallon, supra note 66, at 862, 904.

14 902 Texas Law Review [Vol. 86:889 Under the First Amendment overbreadth doctrine, an individual whose own speech or conduct may be prohibited is permitted to challenge a statute on its face because it also threatens others not before the court those who desire to engage in legally protected expression but who may refrain from doing so rather than risk prosecution or undertake to have the law declared partially invalid. 76 Although it is not as clearly established, at least some authority suggests that the same rule applies for First Amendment vagueness challenges. 77 This special rule has been said to give parties the ability to use First Amendment overbreadth doctrine as a sword, not just a shield, that should be available to attack statutes that appear inimical to First Amendment liberties. 78 Furthermore, language from some of the Court s opinions implies that it applies stricter vagueness and overbreadth analyses to speech-regulating laws than it does in other contexts. For instance, the Court in Reno v. ACLU 79 implied that vagueness problems were more easily found in a statute under a First Amendment challenge than under a Fifth Amendment challenge. 80 Similarly, in NAACP v. Button, 81 the Court contrasted the typical Fifth Amendment due process standards with the different (and impliedly stricter) standard under the First Amendment: The objectionable quality of vagueness and overbreadth does not depend upon absence of fair notice to a criminally accused or upon unchanneled delegation of legislative powers, but upon the danger of 76. Jews for Jesus, 482 U.S. at 574 (internal quotation marks omitted). To invalidate the statute on its face, the overbreadth must be substantial. Id. A statute will be substantially overbroad if its illegitimate applications are too numerous judged in relation to the statute s plainly legitimate sweep, and... no constitutionally adequate narrowing construction suggests itself. Fallon, supra note 66, at 863 (quoting Osborne v. Ohio, 495 U.S. 103, 112 (1990)). 77. Compare Coates v. City of Cincinnati, 402 U.S. 611, (1971) (White, J., dissenting) (noting that in a First Amendment challenge, [a]lthough a statute may [not be] vague... as applied to the conduct charged against a particular defendant, he is permitted to raise its vagueness... as applied to others ), with Broadrick v. Oklahoma, 413 U.S. 601, 608 (1973) ( [E]ven if the outermost boundaries... may be imprecise, any such uncertainty has little relevance here, where appellants conduct falls squarely within the hard core of the statute s proscriptions and appellants concede as much. ). The mixed message from the Court regarding this type of challenge on vagueness grounds has not gone unnoticed by commentators. See RICHARD H. FALLON, JR., DANIEL J. MELTZER & DAVID L. SHAPIRO, HART AND WECHSLER S THE FEDERAL COURTS AND THE FEDERAL SYSTEM 197 (5th ed. 2003) ( Even in the First Amendment context, it seems indisputable that the Supreme Court has sent a mixed message about whether and when parties may challenge a statute that clearly applies to their conduct on the ground that the statute would be vague as applied to others. ). 78. Fallon, supra note 66, at 872; see also City of Lakewood v. Plain Dealer Publ g Co., 486 U.S. 750, (1988) (allowing a newspaper that had not yet applied for a permit to bring a facial challenge to a city ordinance requiring permits for locating newspaper vending racks on public property) U.S. 844 (1997). 80. See id. at 870 ( Regardless of whether [the statute] is so vague that it violates the Fifth Amendment, the many ambiguities concerning the scope of its coverage render it problematic for purposes of the First Amendment. ) U.S. 415 (1963).

15 2008] Limiting the Reach of the Espionage Act 903 tolerating, in the area of First Amendment freedoms, the existence of a penal statute susceptible of sweeping and improper application. 82 Other cases and commentators have recognized similar principles. 83 The essence of both doctrines stems from the chilling effect that each infirmity could exert on speech that should properly be protected under the First Amendment. 84 With these general First Amendment principles in mind, this Note now turns to the specific analysis in Rosen to illustrate the problems with 793(e). D. An Inappropriately Permissive Analysis of a Vague and Overbroad Statute As written, 793(e) infringes on First Amendment values because it is impermissibly vague and overbroad. Because the statute suffers from those infirmities, causing it to cover a wide range of disclosures that should be protected by the First Amendment, (e) is not the least restrictive means of achieving the compelling government interest of protecting national security. Although the Rosen court did an admirable job of attempting to reconcile the remarkable imprecision of 793(e) with constitutional standards under the First Amendment, its ultimate conclusion was unsatisfactory from both constitutional and policy perspectives. The court made two critical mistakes in its analysis of 793(e): first, instead of applying true strict scrutiny to the content-based regulation on speech, it engaged in the kind of ad hoc balancing test that Chief Justice Warren held impermissible in United States v. Robel 86 without subjecting the statute to a least-restrictive-means analysis; and second, it failed to properly apply the specialized vagueness and overbreadth inquiries required in First Amendment jurisprudence. 87 While acknowledging that the case did implicate First Amendment 82. Id. at See, e.g., Smith v. Goguen, 415 U.S. 566, 573 (1974) ( Where a statute s literal scope, unaided by a narrowing state court interpretation, is capable of reaching expression sheltered by the First Amendment, the [vagueness] doctrine demands a greater degree of specificity than in other contexts. ); Grayned v. City of Rockford, 408 U.S. 104, 109 (1972) ( [W]here a vague statute abut[s] upon sensitive areas of basic First Amendment freedoms, it operates to inhibit the exercise of [those] freedoms. (internal quotation marks omitted)); Scull v. Virginia ex rel. Comm. on Law Reform & Racial Activities, 359 U.S. 344, 353 (1959) ( Certainty is all the more essential when vagueness might induce individuals to forego their rights of speech.... ); Recent Case, supra note 40, at 826 (discussing the higher level of scrutiny for vagueness that is required when the First Amendment is implicated ). 84. Reno, 521 U.S. at 872, (vagueness); accord New York v. Ferber, 458 U.S. 747, 772 n.27 (1982) (overbreadth). 85. See infra subpart II(E). 86. See 389 U.S. 258 (1967). 87. See Recent Case, supra note 40, at 826.

16 904 Texas Law Review [Vol. 86:889 concerns, 88 the court did not conduct a least-restrictive-means analysis or give meaningful weight to the possibility of the statute s chilling effect on protected speech in its lengthy analysis of the defendants free-speech challenges. Instead, drawing on two cases Schenck v. United States 89 and Dennis v. United States, 90 from the World War I and World War II eras, respectively that scholars agree have been greatly reduced in their influence if not implicitly overruled by subsequent Supreme Court decisions, 91 the court engaged in an ad hoc balancing test in which it simply weighed the government s interest in preserving national-defense secrets against the individual defendants interest in disclosing the information in that particular case. 92 To justify its holding that the statute was not unconstitutionally overbroad, the court emphasized the judicial limitations that it would place on the application of the statute. 93 The judicial limitations in Rosen, however, essentially changed the entire meaning of a relatively clear statute, and they still did not make 793(e) the least restrictive means of furthering the government interest in protecting national security. A more exacting scrutiny applied to 793(e), using the least-restrictivemeans analysis, should have yielded the opposite result. Although the government s interest in protecting national security is certainly compelling, the vagueness and overbreadth problems of 793(e) reveal that it is not the least restrictive means of addressing that interest. In testing for vagueness or overbreadth under the First Amendment, courts should determine whether the statute as worded could have a chilling effect on otherwise-protected speech. 94 For instance, in Reno v. ACLU, the Supreme Court considered a provision of the Communications Decency Act that would criminally prohibit the transmission of indecent material to minors over the Internet or the placement of patently offensive material on the Internet in any way that would make it available to minors. 95 The Court held that the statute was unconstitutionally vague under the First Amendment, largely because the 88. United States v. Rosen, 445 F. Supp. 2d 602, (E.D. Va. 2006), amended by No. 1:05cr225, 2006 WL (E.D. Va. 2006) U.S. 47 (1919) U.S. 494 (1951). 91. See, e.g., Wilson R. Huhn, Assessing the Constitutionality of Laws that Are Both Content- Based and Content-Neutral: The Emerging Constitutional Calculus, 79 IND. L.J. 801, 856 (2004) (asserting that the clear and present danger test of Schenck and Dennis has been supplanted by Brandenburg v. Ohio, 395 U.S. 444 (1969)). Furthermore, Schenck and Dennis concerned regulating speech intended to advocate unlawful and subversive conduct a different context and inquiry than those involved in Espionage Act cases. See Dennis, 341 U.S. at (affirming a conviction of individuals found to have conspired to organize the Communist Party to advocate the violent overthrow of the U.S. government); Schenck, 249 U.S. at 49, (affirming a conviction for disseminating a pamphlet seeking to incite insubordination in, and obstruct recruitment efforts of, the U.S. military during World War I). 92. Rosen, 445 F. Supp. 2d at Id. at Reno v. ACLU, 521 U.S. 844, (1997). 95. Id. at