Emory Law Journal

Terrroism and Associations
Ashutosh Bhagwat Professor of Law and Martin Luther King Jr. Hall Research Scholar, University of California, Davis. B.A., 1986, Yale University; J.D., 1990, The University of Chicago. Thanks to Ben Strauss and Elise Weinberg for excellent research assistance. Thanks also to Deven Desai, Chris Elmendorf, Carlton Larson, and participants at the Freedom of Expression Conference at the Yale Law School for extremely helpful comments, including especially to John Inazu for serving as discussant for this Article and providing invaluable feedback.

Abstract

The domestic manifestation of the War on Terror has produced the most difficult and sustained set of controversies regarding the limits on First Amendment protections for political speech and association since the anti-Communist crusades of the Red Scare and McCarthy eras. An examination of the types of domestic terrorism prosecutions that have become common since the September 11 attacks reveals continuing, unresolved conflicts between national security needs and traditional protections for speech and (especially) associational freedoms. Yet the courts have barely begun to acknowledge, much less address, these serious issues. In the Supreme Court’s only sustained engagement with these problems, the 2010 decision in Holder v. Humanitarian Law Project, the majority largely avoided the hard questions by simply asserting that 18 U.S.C. § 2339B, the federal statute forbidding the provision of material support to foreign terrorist organizations, does not directly burden either the freedom of speech or freedom of association. Lower courts have performed even more poorly, generally rejecting powerful speech and association claims with bare assertions that “there is no First Amendment right . . . to support terrorists.”

This Article has taken as its major goal identifying and analyzing the First Amendment issues raised by the domestic War on Terror, focusing especially on the role of freedom of association in this context. Freedom of association has historically been a critical and basic First Amendment right, central to the process of democratic self-governance that the First Amendment protects. The right of association is also deeply implicated in many domestic terrorism prosecutions, since the essence of those prosecutions is an act of association, often combined with speech. Finally, the judiciary’s bare assertions that “material support” or financial contributions do not constitute association cannot be sustained given both first principles and well-developed law outside the context of terrorism. In short, in this area the courts have failed in their basic job of honestly engaging with the law.

Ultimately, however, I conclude here that there does exist a clear, textually and historically justifiable basis for limiting constitutional protections for terrorist and other violent groups. The principle derives from the textual roots of the freedom of association, which lies in the Assembly Clause of the First Amendment. The Assembly Clause, unlike the Free Speech Clause, explicitly protects only a right “peaceably to assemble,” and so excludes violent groups. This simple principle, completely missed by the courts, serves to reconcile most terrorism prosecutions with the First Amendment. It cannot, of course, resolve all issues, especially when a prosecution is based primarily on speech, not association, but it does much of the work. There also remain some difficult and complicated issues of definition and implementation, on which I provide some thoughts. But the basic argument advanced here is quite simple: the freedom of association and assembly protects only peaceable association and assembly; and terrorists are not peaceable.

Introduction

More than a decade has now passed since the attacks of September 11, 2001, fully inaugurated the Age of Terror. In the early years after the attacks, aside from the immigration sweep that followed immediately, U.S. antiterrorism policy was focused primarily on threats from abroad, including, notably, the wars in Afghanistan and (at least purportedly) Iraq. While those events raised some fascinating issues about the scope of executive authority 1See, e.g., 42 U.S.C. § 2000dd (2006) (prohibiting “cruel, inhuman, or degrading treatment or punishment” by executive officials of individuals in custody); Hamdan v. Rumsfeld, 548 U.S. 557, 568–69 (2006) (discussing the President’s issuance of a comprehensive military order governing the use of military commissions to try noncitizens in the War on Terror); Hamdi v. Rumsfeld, 542 U.S. 507, 516 (2004) (discussing whether the President has the authority to detain “enemy combatants” (internal quotation marks omitted)); Statement on Signing the Department of Defense, Emergency Supplemental Appropriations to Address Hurricanes in the Gulf of Mexico, and Pandemic Influenza Act, 2006, 2 Pub. Papers 1901 (Dec. 30, 2005) (asserting that “situations may arise, especially in wartime, in which the President must act promptly under his constitutional grants of executive power”). and about the geographic reach of the Constitution, 2See Boumediene v. Bush, 553 U.S. 723, 732 (2008) (holding that the constitutional privilege of habeas corpus reaches enemy combatants detained at Guantanamo Bay, Cuba). it was relatively rare that the individual liberties provisions of the Bill of Rights were directly implicated. 3The primary counterexample is the controversy over whether the Bush Administration’s domestic warrantless wiretapping program violated the Fourth Amendment. See Evan Tsen Lee, The Legality of the NSA Wiretapping Program, 12 Tex. J. C.L. & C.R. 1, 40–41 (2006). In subsequent years, however, the federal government initiated a series of judicial actions, including criminal prosecutions, directed at alleged terrorists and supporters of terrorism within the United States. 4See infra Part II. These actions have in turn generated a large number of constitutional disputes regarding the consistency of these legal claims with the Bill of Rights, especially the speech and association rights protected by the First Amendment. To date, the Supreme Court has had only one occasion to address these issues: its 2010 decision in Holder v. Humanitarian Law Project. 5130 S. Ct. 2705 (2010). In Humanitarian Law Project, the Court rejected a First Amendment challenge to 18 U.S.C. § 2339B(a)(1), the so-called material-support statute, which bans the provision of material support to designated foreign terrorist organizations. 6See id. at 2711; see also 18 U.S.C. § 2339B(a)(1) (2006). The Humanitarian Law Project decision, however, did very little to clarify the law regarding the interaction between First Amendment rights and antiterrorism measures; indeed, the decision, if anything, increased the already high levels of confusion and uncertainty.

Despite the lack of guidance from above, the lower federal courts have, of course, necessarily confronted and resolved many First Amendment issues in the context of terrorism prosecutions. These cases and disputes are discussed in more detail in Part II of this Article, but the bottom line is clear: the First Amendment has been irrelevant. Lower courts have not only consistently rejected First Amendment defenses, they have generally dismissed them as insubstantial. A closer look at these cases demonstrates, however, that under current law, the First Amendment claims in these cases, especially those brought under the freedom of association, are in fact quite weighty. Courts have avoided them only by contorting doctrine and, in some cases, accepting arguments that are grossly inconsistent with First Amendment law in other factual contexts. In short, the War on Terror has forced the courts to twist the First Amendment into a pretzel.

In this Article, I aim to abate some of this confusion and to build a sustainable framework within which First Amendment challenges to antiterrorism measures can be evaluated. My focus is on freedom of association, though in the course of discussing association issues, I necessarily have to consider some related free speech issues as well—indeed, one of my main points is that speech and association issues are deeply entangled in these situations, and contrary to the way courts treat them, must be considered together. 7For a more general discussion of the relationship between the speech and association rights, see Ashutosh Bhagwat, Associational Speech, 120 Yale L.J. 978 (2011). As I suggested earlier, my conclusion is that the associational claims raised in this area are far from insubstantial; indeed, under extant doctrine, many of them are probably valid. Nonetheless, I conclude that the courts have probably been correct to reject most (though by no means all) of the First Amendment challenges to antiterrorism prosecutions. The problem is that the courts are doing this for the wrong reasons, twisting doctrine to reach intellectually unsustainable conclusions. The downsides to this approach—other than rule-of-law concerns—are that first, it leads courts to reject even some legitimate claims; and second, it threatens to undermine First Amendment rights in other areas.

I conclude by arguing that instead of ignoring individual associational rights or narrowing their scope in indefensible ways, courts should focus their attention on what kinds of associations, what kinds of groups are protected by the First Amendment, and more particularly, whether certain kinds of organizations may be categorically excluded. The building blocks of such an approach can be found in the text and history of the First Amendment itself and in particular in the Assembly Clause of the First Amendment. The modern right of association, I demonstrate, is rooted in the Assembly Clause. Importantly, however, the Assembly Clause (unlike the Free Speech Clause) only protects the right “peaceably to assemble,” 8U.S. Const. amend. I. and so explicitly excludes violent groups. This fact, and the recognition that terrorist groups are, of course, not peaceable, serves to reconcile most terrorism prosecutions with the First Amendment without doing damage to either logic or legitimate rights. This Article develops these ideas more fully and demonstrates that such an approach can preserve important First Amendment principles without reaching absurd results, such as supporting a constitutional right to fund Al Qaeda.

In Part I of this Article, I discuss the Supreme Court’s Humanitarian Law Project decision. Part II presents a series of case studies from the domestic War on Terror, illustrating how the lower courts have, and have not, confronted the First Amendment issues raised by them. Part III examines the grave uncertainties these cases reveal regarding the scope and application of First Amendment rights in the context of the War on Terror. Part IV articulates a new way forward, which seeks to identify limits on the sorts of groups entitled to constitutional protection based on the language and purposes of the First Amendment. Finally, Part V examines some of the difficult boundary problems raised by any proposal to categorically restrict the scope of First Amendment protections.

I. Holder v. Humanitarian Law Project: Terrorism Trumps the First Amendment

As the discussion in the following section indicates, since the attacks of September 11, 2001, the federal courts have decided many cases, arising from a variety of contexts, touching upon the domestic War on Terror. The Supreme Court, however, has directly faced constitutional issues arising from domestic antiterrorism efforts in only one case: Holder v. Humanitarian Law Project. 9See 130 S. Ct. at 2711. The Humanitarian Law Project decision therefore provides the starting point for all modern discussions of First Amendment constraints on antiterrorism prosecutions, and it is worthy of some sustained attention.

A. The Humanitarian Law Project Decision

The Supreme Court’s decision in Holder v. Humanitarian Law Project arose out of a constitutional challenge to 18 U.S.C. § 2339B, the so-called material-support statute, which prohibits “knowingly provid[ing] material support or resources to a foreign terrorist organization.” 10See id. at 2712 & n.1. The material-support statute was originally enacted as part of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), Pub. L. No. 104-132, sec. 303(a), § 2339B(a)(1), 110 Stat. 1214, 1250. It thus predates the September 11 attacks, though the number of prosecutions under the statute appear to have increased dramatically following the attacks. See Andrew Peterson, Addressing Tomorrow’s Terrorists, 2 J. Nat’l Security L. & Pol’y 297, 301 (2008). The statute in turn defines “material support or resources” broadly, to include, inter alia, any and all property, services, training, expert advice, and personnel. 11Humanitarian Law Project, 130 S. Ct. at 2713 (quoting 18 U.S.C. § 2339A(b)(1) (2006)) (internal quotation marks omitted). A foreign terrorist organization (FTO) is defined as “an organization designated as a terrorist organization” 1218 U.S.C. § 2339B(g)(6) (2012). by the Secretary of State, pursuant to her authority under 8 U.S.C. § 1189. 13See 8 U.S.C. § 1189(a)(1) (2012). The plaintiffs in the litigation were U.S. citizens and organizations who wished to provide expert training and advice to support the nonviolent activities of two designated FTOs: the Kurdistan Workers’ Party (PKK), an organization seeking the establishment of an independent Kurdish state in Turkey, and the Liberation Tigers of Tamil Eelam (LTTE or Tamil Tigers), a now defunct organization that sought the establishment of an independent Tamil state in Sri Lanka. 14Humanitarian Law Project, 130 S. Ct. at 2713–14. The plaintiffs claimed that application of the material-support statute to their activities violated their rights to freedom of speech and freedom of association protected by the First Amendment and that the statute was void for vagueness. 15Id. at 2714. The Supreme Court, in an opinion by Chief Justice Roberts, rejected all of these claims by a 6–3 vote. 16See id. at 2712, 2719, 2730–31.

The majority’s analysis proceeded in several steps. First, the Court rejected a nonconstitutional claim that the statute should be interpreted to require that defendants possess an intent to further an FTO’s illegal activities, at least when the statute was applied to speech. 17See id. at 2717–18. The Court concluded that the statutory language did not support such a limiting construction. 18See id. The Court then rejected the void-for-vagueness challenge. 19Id. at 2719. It held that the statute’s definition of “training” and “expert advice” clearly covered the plaintiffs’ proposed activities, and, critically, it held that the statute’s definitions of “personnel” and “service” clearly excluded independent advocacy in support of an FTO, thereby providing a safe harbor so long as an individual’s activities are not directed by or coordinated with an FTO. 20Id. at 2719–22 (internal quotation marks omitted). In response to the plaintiffs’ objection that this reading in turn created fatal ambiguity about the degree of coordination or direction that would cross the line into material support, the Court simply responded that the plaintiffs had not alleged sufficient facts to properly raise this issue. 21See id. at 2722. In other words, the majority essentially ducked this, as we shall see, crucial question.

The majority then turned to the First Amendment. It began, building on its vagueness analysis, by emphasizing that the material-support statute was not a flat restriction on political speech because it permitted independent advocacy in support of FTOs. 22Id. at 2722–23. However, the Court acknowledged that the statute imposed a content-based restriction on speech, and so “more rigorous scrutiny” than intermediate scrutiny was required (though the majority opinion coyly never used the phrase “strict scrutiny”). 23See id. at 2724. The majority then quickly concluded that the government’s objective here, to combat terrorism, was sufficiently strong (presumably compelling, though in keeping with its failure to say whether it is or is not applying strict scrutiny, the Court never used that term) to meet the applicable standard of review. 24See id. (explaining that “the Government’s interest in combating terrorism is an urgent objective of the highest order”). The difficult question was one of fit—whether the statute was “necessary to further that interest.” 25See id. (quoting Opening Brief for Humanitarian Law Project at 51, Humanitarian Law Project, 130 S. Ct. 2705 (Nos. 08-148 & 09–89)) (internal quotation marks omitted). Ultimately, the Court concluded that it was necessary. 26See id. at 2728–29 (“Given the sensitive interests in national security and foreign affairs at stake . . . it was necessary to prohibit providing material support in the form of training, expert advice, personnel and services to foreign terrorist groups . . . .”). The Court pointed out that even material support for an FTO’s legal activities advances terrorism because it “frees up other resources within the organization that may be put to violent ends,” and “[i]t also importantly helps lend legitimacy to foreign terrorist groups—legitimacy that makes it easier for those groups to persist, to recruit members, and to raise funds—all of which facilitate more terrorist attacks.” 27Id. at 2725. In a similar vein, the Court argued that providing material support to FTOs “also furthers terrorism by straining the United States’ relationships with its allies and undermining cooperative efforts between nations to prevent terrorist attacks.” 28Id. at 2726. Finally, the Court granted deference to Congress and the President on the necessity of the material-support statute, given the national security and foreign relations implications of the case. 29Id. at 2727. However, the Court closed this section of the analysis with a limiting cautionary statement:

[W]e in no way suggest that a regulation of independent speech would pass constitutional muster, even if the Government were to show that such speech benefits foreign terrorist organizations. We also do not suggest that Congress could extend the same prohibition on material support at issue here to domestic organizations. 30Id. at 2730.

Having polished off the free speech claim, the majority then turned to freedom of association. Here, the Court was brief, almost dismissive. 31The Court spent a mere three paragraphs discussing the freedom of association claim. See id. at 2730–31. It rejected the claim on the simple grounds that the material-support “statute does not penalize mere association with a foreign terrorist organization,” because it does not prohibit membership in an FTO, merely the provision of material support. 32Id. at 2730.

Justice Breyer wrote a dissenting opinion, joined by Justices Ginsburg and Sotomayor. 33See id. at 2731 (Breyer, J., dissenting). From the beginning, the tone of the dissent is dramatically different from the majority. Breyer emphasized that the plaintiffs’ activities were political and so fell within the core of the First Amendment’s protections. 34Id. at 2732. He also repeatedly described the activities as implicating rights of both speech and association, rather than treating the associational claim as a poor stepchild like the majority did. 35Id. at 2732–33. Indeed, Breyer emphatically rejected the majority’s conclusion that the statute was constitutional because it prohibits only speech coordinated with an FTO, pointing out that the First Amendment, “after all, also protects the freedom of association,” and citing cases that describe the right of freedom of assembly as an independent and “cognate” right. 36Id. (citing NAACP v. Claiborne Hardware Co., 458 U.S. 886, 911 (1982); De Jonge v. Oregon, 299 U.S. 353, 364 (1937)). The dissent also pointed out that the Court’s prior precedent had clearly upheld a right to associate with groups that themselves use “unlawful means” to achieve their political goals. 37Id. at 2733 (citing Claiborne, 458 U.S. at 908; Scales v. United States, 367 U.S. 203, 229 (1961)). Finally, Justice Breyer argued that the government’s primary arguments in defense of the material-support statute—that even peaceful support to FTOs is “fungible,” and that support for FTOs can “legitimize” them—are either factually questionable or contrary to precedent. 38See id. at 2735–37 (internal quotation marks omitted). Because it found the statute, as interpreted by the government, to violate the First Amendment, the dissent would have imposed a limiting construction “criminalizing First-Amendment-protected pure speech and association only when the defendant knows or intends that those activities will assist the organization’s unlawful terrorist actions.” 39Id. at 2739–40.

B. Implications

Several significant questions are raised by the Humanitarian Law Project Court’s First Amendment analysis. First, it is striking that despite the Court’s invocation of strict scrutiny (albeit without naming it as such), it deferred to Congress and the President’s factual assertions regarding the necessity of the law. Such a deferential posture in the context of heightened scrutiny is inconsistent with most modern law 40See, e.g., Johnson v. California, 543 U.S. 499, 512–13 (2005) (noting that the Court has “refused to defer to state officials’ judgments on race in other areas where those officials traditionally exercised substantial discretion”); Landmark Commc’ns, Inc. v. Virginia, 435 U.S. 829, 844 (1978) (explaining that courts must independently review legislative actions that impinge on the First Amendment). But see Grutter v. Bollinger, 539 U.S. 306, 328 (2003) (deferring to a law school’s educational judgment about the value of diversity). and brings to mind such questionable decisions as Korematsu v. United States 41323 U.S. 214, 217–18 (1944) (determining it was within “the war power of Congress and the Executive to exclude those of Japanese ancestry from the West Coast war area”). and Dennis v. United States 42341 U.S. 494, 516–17 (1951) (upholding prosecution of Communist Party leadership as consistent with the First Amendment). in which the Court sacrificed basic civil liberties in the name of national security and deference to political and military authorities. The sweeping deference language of the Humanitarian Law Project majority opinion raises the possibility that in the Age of Terror, the Court will retreat to a similar constitutional calculus.

In other ways, however, the Humanitarian Law Project Court took a notably more speech-protective stance than it had to. Most significantly, by applying rigorous scrutiny, the Court implicitly but clearly rejected any suggestion that the speech at issue was categorically unprotected under current First Amendment doctrine, either as “incitement” under the modern Brandenburg test 43See Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (explaining that the First Amendment does “not permit a State to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action”). or as speech which is “an integral part of conduct in violation of a valid criminal statute.” 44United States v. Stevens, 130 S. Ct. 1577, 1586 (2010) (quoting New York v. Ferber, 458 U.S. 747, 761–62 (1982)) (internal quotation mark omitted). The Court also, as noted earlier, explicitly excluded independent advocacy in support of FTOs from the reach of the material-support statute and raised doubts about whether the statute could constitutionally be applied to domestic organizations, even terrorist ones. 45See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010). Ultimately, then, Humanitarian Law Project is probably best read as an important but, at least for the time being, limited holding regarding the scope of free speech in the context of the War on Terror.

But this is precisely the problem: Humanitarian Law Project is clearly a holding primarily about speech with the right of association treated merely as an afterthought. But on the facts of the case, this is quite odd. It is true that the plaintiffs themselves, to some extent, framed their case around free speech by emphasizing the training and advice they wanted to provide to FTOs; but central to the dispute (as the majority itself recognized repeatedly) was the fact that the plaintiffs wished to act in coordination with—i.e., in association with 46See id. at 2732 (Breyer, J., dissenting). —FTOs, rather than simply advocating on their behalf independently. In other words, it was the plaintiffs’ association rather than their speech that brought them within the scope of the material-support statute. Yet the majority’s analysis of association was, as we have seen, utterly cursory.

In many ways, this neglect of the associational right is typical of the Court’s recent performance and has already been extensively discussed (and criticized) in the literature. 47See generally John D. Inazu, Liberty’s Refuge: The Forgotten Freedom of Assembly 1–2 (2012) (“[The] freedom of assembly has become little more than a historical footnote in American law and political theory.”); Bhagwat, supra note 7, at 980 (explaining that the freedom of assembly and right to petition the government “have traditionally been the poor stepchildren of First American law, neglected and ignored”). The result of this neglect, however, is to create profound confusion. For example, one clear and obvious implication of the Court’s reasoning in rejecting the associational claim in Humanitarian Law Project is that the right of association protects membership in an organization but not the provision of material support. Lower courts have largely mimicked this reasoning. 48See infra Part II. The difficulty is that this conclusion is entirely inconsistent with interpretations of the associational right in other contexts. 49See infra text accompanying notes 217–24. In addition, as noted earlier, the Court completely avoided addressing the difficult question of what level of communication or direction between an individual and an FTO crosses the line into “coordination” and so brings the individual within the scope of the material-support statute. 50See Humanitarian Law Project, 130 S. Ct. at 2722. Yet these questions lie at the heart of many First Amendment claims raised in modern terrorism prosecutions. In sum, far from clarifying the scope of First Amendment rights in the context of terrorism, Humanitarian Law Project largely obfuscated or avoided the issues. As we shall now see, the net effect of Humanitarian Law Project was therefore merely to add to the confusion already rampant in the lower courts.

II. Terrorism in the Lower Courts: The First Amendment in Absentia

To understand the nature of the First Amendment problems raised by modern antiterrorism cases, it is necessary to have some sense of the nature of those cases. I therefore begin this section by laying out some recent, prominent terrorism cases that raise difficult First Amendment issues. It should be noted that I focus here on cases that primarily target speech or association, as opposed to targeting the actions of individuals such as Major Nidal Malik Hasan 51Hasan is the convicted gunman in the November 2009 Fort Hood shootings. See Nidal Malik Hasan, N.Y. Times, (last visited Jan. 8, 2014); see also Manny Fernandez, Defendant in Fort Hood Shooting Case Admits Being Gunman, N.Y. Times, Aug. 7, 2013, at A14. or Faisal Shahzad 52Shahzad was convicted of attempting to detonate a car bomb in Times Square in May of 2010. See Faisal Shahzad, N.Y. Times, http://topics.nytimes.com/top/reference/timestopics/people/s/faisal_shahzad/index.html (last visited Jan. 8, 2014); see also Declan Walsh, 4 Are Acquitted by Pakistan in Times Square Bomb Plot, N.Y. Times, June 3, 2012, at A13. who engaged in, or attempted to engage in, violence, and whose prosecutions, of course, raise no serious First Amendment concerns. These cases tend to fall into one of two groups. The first is made up of antiterrorism prosecutions, generally brought pursuant to either 18 U.S.C. § 2339B, the material-support statute upheld in Humanitarian Law Project, or its sibling, 18 U.S.C. § 2339A, which prohibits the provision of material support or resources “knowing or intending that they are to be used in preparation for, or in carrying out” various violations 5318 U.S.C. § 2339A (2012). (i.e., this statute prohibits material support intended to aid actual acts of terrorism, as opposed to material support to FTOs). The second set of cases involves organizations seeking to challenge their designations as terrorist organizations.

A. The Cases

1. Material-Support Prosecutions Based on Speech

a. Tarek Mehanna

Tarek Mehanna is a native-born U.S. citizen, with a doctorate in pharmacology, who lived with his parents in the upscale suburb of Sudbury, Massachusetts. 54See Innokenty Pyetranker, Sharing Translations or Supporting Terror? An Analysis of Tarek Mehanna in the Aftermath of Holder v. Humanitarian Law Project, Nat’l Security L. Brief, Spring 2012, at 21, 21; Abby Goodnough & Liz Robbins, Massachusetts Man Is Held on Charges of Plotting Attacks, Including One at a Mall, N.Y. Times, Oct. 22, 2009, at A21. In December of 2011, Mehanna was convicted after a jury trial on multiple counts of providing material support to terrorism (as well as counts involving lying to the FBI). 55See Pyetranker, supra note 54, at 38; Abby Goodnough, U.S. Citizen Is Convicted in Plot to Support Al Qaeda, N.Y. Times, Dec. 21, 2011, at A26. Mehanna’s prosecution was based in part on his translation from Arabic, and distribution on the Internet, of jihadi literature (in particular, a text titled 39 Ways to Serve and Participate in Jihad and a video depicting Al Qaeda activities and propaganda involving Iraq). 56See Pyetranker, supra note 54, at 36 (citing Second Superseding Indictment at 1, 8, 15–16, United States v. Mehanna, No. 09-CR-10017-GAO (D. Mass. 2011), 2010 WL 2516469); see also Nikolas Abel, Note, United States vs. Mehanna, the First Amendment, and Material Support in the War on Terror, 54 B.C. L. Rev. 711, 732 (2013) (citing Government’s Proffer and Memorandum in Support of Detention at 64–66, Mehanna, No. 09-CR-10017-GAO (D. Mass 2011)). The prosecution also introduced evidence that Mehanna flew to Yemen in 2004 seeking to obtain weapons training at a terrorist training camp, but conceded that he did not succeed in locating a camp (Mehanna denied that he was seeking weapons training). 57See Abel, supra note 56, at 731 & n.169 (citing Second Superseding Indictment, supra note 56, at 5). A main thrust of the government’s case at the trial court level, however, was clearly the translation and distribution of jihadi materials. 58See id. at 732 (citing Second Superseding Indictment, supra note 56, at 8). During the late stages of the editorial process for this Article, the First Circuit issued its opinion affirming Mehanna’s conviction in full. See United States v. Mehanna, 735 F.3d 32, 41 (1st Cir. 2013). Interestingly, the appellate court’s decision to affirm Mehanna’s conviction on the terrorism counts relied entirely on his 2004 trip to Yemen. See id. at 46. Indeed ,the court came close to conceding that the government’s evidence was insufficient to demonstrate that Mehanna’s translation and speech activities were coordinated with Al Qaeda. See id. at 50–51. The court concluded that the mass of evidence that had been admitted at trial regarding the government’s “translation-as-material-support theory” was irrelevant because the jury was properly instructed, and the factual evidence regarding Mehanna’s Yemen trip was sufficient to support his conviction. Id. at 51. Whatever the merits of this analysis, it remains true that at trial, before the convicting jury, the government heavily emphasized Mehanna’s speech. Importantly, the government provided no proof that Mehanna coordinated his propaganda with any members of Al Qaeda or any other FTO. 59See id. at 739–40. Indeed, it is quite unclear whether Mehanna had any direct contact at all with Al Qaeda members. 60Id. Instead, the government’s case relied on evidence that Mehanna described himself as Al Qaeda’s “English Wing” 61Id. at 732 & n.175 (quoting Government’s Proffer and Memorandum in Support of Detention, supra note 56, at 11) (internal quotation marks omitted). and on other statements Mehanna made expressing support for Osama bin Laden and Al Qaeda. 62See Pyetranker, supra note 54, at 21–22.

Mehanna’s attorneys unsurprisingly sought to have the material-support charges against Mehanna dismissed on First Amendment grounds. 63Id. at 36–37 (citing Defendant’s Memorandum of Law in Support of His Motion to Dismiss Counts One Through Three Based on Vagueness and Overbreadth, United States v. Mehanna, No. 09-CR-10017-GAO (D. Mass. 2011), 2011 WL 3740563). The primary thrust of his argument was that the Humanitarian Law Project decision required proof of coordination with an FTO (or in the case of § 2339A, with known terrorists) and that the overt acts for which Mehanna had been indicted constituted protected, independent advocacy. 64See id. at 37 (citing Defendant’s Memorandum of Law in Support of His Motion to Dismiss, supra note 63, at 3, 5, 16–18); Abel, supra note 56 at 732–33 (citing Defendant’s Memorandum of Law in Support of His Motion to Dismiss, supra note 63, at 11–12). The government’s response was to deny that coordination was a required element of a § 2339A violation, and that in any event, “coordination” did not require direct contact between the defendant and an FTO—it was sufficient for a defendant to respond to an FTO’s call for assistance, and believe that he was assisting the organization. 65See Abel, supra note 56 at 732–33 (citing Government’s Opposition to Defendant’s Motion to Dismiss Counts One Through Three of the Second Superseding Indictment Based on Overbreadth and Vagueness at 21–23, Mehanna, No. 09-CR-10017-GAO (D. Mass. 2011), 2011 WL 3959520). The district court denied Mehanna’s motion to dismiss, and ultimately Mehanna was convicted by a jury and sentenced to seventeen-and-a-half years in prison. 66Pyetranker, supra note 54, at 38.

b. Javed Iqbal and Laleh Elahwal

Javed Iqbal is a Pakistani national who has resided in the United States since he was a teenager. 67Benjamin Weiser, A Guilty Plea in Providing Satellite TV for Hezbollah, N.Y. Times, Dec. 24, 2008, at A21. Iqbal lived on Staten Island, New York, and ran a business providing specialized satellite programming. 68See id. In 2006, Iqbal was arrested and charged with material support for terrorism. 69See id. The gravamen of the charge was that as part of his business, Iqbal retransmitted the signal of Al Manar, a television station associated with the Lebanese Shia organization, Hezbollah (sometimes spelled Hizballah). 70See Timothy Williams & William K. Rashbaum, New York Man Charged with Enabling Hezbollah Television Broadcasts, N.Y. Times, Aug. 25, 2006, . Hezbollah is a complex organization that conducts charitable activities and is a political party within Lebanon, but it also uses terrorism to achieve its political goals. 71See Nicholas Kulish, Despite U.S. Fear, Hezbollah Moves Openly in Europe, N.Y. Times, Aug. 16, 2012, at A1. It has been a designated FTO since 1997. 72Bureau of Counterterrorism, Foreign Terrorist Organizations, U.S. Dep’t St. (Sept. 28, 2012), http://www.state.gov/j/ct/rls/other/des/123085.htm. Eventually, after his First Amendment defense was denied by the trial judge on the grounds that Iqbal was being prosecuted for conduct, not speech, Iqbal pled guilty to one count of providing material support. 73See Weiser, supra note 67. In pleading guilty, Iqbal confessed to receiving money for providing services to Al Manar, a fact that the court emphasized in rejecting Iqbal’s First Amendment defense. 74Id. Iqbal was sentenced to sixty-nine months in prison. 75Press Release, U.S. Attorney’s Office, S. Dist. of N.Y., New Jersey Man Sentenced to 17 Months in Prison for Providing Material Support and Resources to Hizballah (June 23, 2009), http://www.fbi.gov/newyork/press-releases/2009/nyfo062309.htm.

Saleh Elahwal was Iqbal’s business associate. 76See id. He too eventually pled guilty to a material-support charge and was sentenced to seventeen months in prison. 77Id.

c. Jubair Ahmad

Jubair Ahmad is a young Pakistani-American (he was twenty-four at the time of his conviction) living in Woodbridge, Virginia. 78See Rachel Karas, Woodbridge Man Sentenced to 12 Years for Terrorist Propaganda Video, Post Local (Apr. 13, 2012, 6:40 PM ), http://www.washingtonpost.com/blogs/crime-scene/post/woodbridge-man-sentenced-to-12-years-for-terrorist-propaganda-video/2012/04/13/gIQACI7tFT_blog.html. When he was a teenager, before moving to the United States, Ahmad attended a training camp run by Lashkar-e-Tayyiba (LeT), a Pakistani militant group that conducts attacks against India because of the ongoing dispute between India and Pakistan over the state of Kashmir (LeT is believed to be behind the November 2008 terrorist attack on Mumbai). 79Press Release, U.S. Attorney’s Office, E. Dist. of Va., Pakistani National Living in Woodbridge Pleads Guilty to Providing Material Support to Terrorist Organization (Dec. 2, 2011), http://www.justice.gov/usao/vae/news/2011/12/20111202ahmadnr.html; see Karas, supra note 78. LeT has been a designated FTO since 2001. 80Bureau of Counterterrorism, supra note 72. In 2010, at the behest of the son of LeT’s leader, Ahmad prepared a video glorifying LeT’s activities and calling for fighters to wage jihad. 81See Karas, supra note 78. He then posted the video to YouTube. 82Id. These actions were the sole basis of his conviction for providing material support to terrorism, resulting in a twelve-year sentence. 83Id.

2. Material-Support Prosecutions Based on Money

a. Mohamad Hammoud

Mohamad Hamomud is a Lebanese citizen who came to the United States in 1992. 84United States v. Hammoud, 381 F.3d 316, 325 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005). While in the United States, he became involved in various illegal activities, including cigarette smuggling. 85See id. In addition, he assisted in raising money for Hezbollah and himself donated $3,500 to that organization. 86Id. at 326. Hammoud’s associates were also found to have provided “dual use” physical equipment to Hezbollah, but Hammoud declined to participate in these activities. Id. On the basis of these actions, Hammoud was tried and convicted of providing material support to a terrorist organization (along with a number of other offenses). 87Id.

On appeal, Hammoud argued that his material-support conviction violated his First Amendment right to freedom of association. 88See id. at 328. The en banc Fourth Circuit rejected this view. 89Id. at 329. The primary grounds for its ruling was that the material-support statute “does not prohibit mere association; it prohibits the conduct of providing material support to a designated FTO. Therefore, cases regarding mere association with an organization do not control.” 90Id. Because it found that the statute did not directly target First Amendment protected activity, the court applied the lenient O’Brien test applicable to facially neutral statutes that incidentally restrict expressive conduct and easily upheld the statute. 91Id. (citing United States v. O’Brien, 391 U.S. 367, 377 (1968)). In O’Brien, the Supreme Court upheld the conviction of David Paul O’Brien for burning his draft card during an antiwar rally, on the grounds that the statute prohibiting the burning of draft cards was not directed at speech, but was rather a neutral conduct regulation whose purpose was “unrelated to the suppression of free expression.” O’Brien, 391 U.S. at 369, 377.

b. Hossein Afshari, Roya Rahmani et al.

Between 1997 and 2001, a group of individuals, including Hossein Afshari and Roya Rahmani, is alleged to have solicited funds at the Los Angeles International Airport, purportedly on behalf of an organization named “Committee for Human Rights.” 92United States v. Afshari, 426 F.3d 1150, 1152 (9th Cir. 2005) (internal quotation marks omitted). The collected funds were then allegedly forwarded to an organization called Mujahedin-e Khalq (MEK). 93Id. The MEK is an Iranian Marxist group which was designated as an FTO in 1997 (that designation was lifted in September of 2012). 94Id. at 1152–53; Scott Shane, Iranian Dissidents Convince U.S. to Drop Terror Label, N.Y. Times, Sept. 21, 2012, http://www.nytimes.com/2012/09/22/world/middleeast/iranian-opposition-group-mek-wins-removal-from-us-terrorist-list.html?pagewanted=all&_r=0. As a consequence, Afshari, Rahmani, and their confederates were indicted for violating the material-support statute. 95Afshari, 426 F.3d at 1152. The district court dismissed the indictment on the grounds that the procedures used to designate MEK as an FTO were unconstitutional. 96See United States v. Rahmani, 209 F. Supp.2d 1045, 1059 (C.D. Cal. 2002), rev’d sub nom. United States v. Ashfari 426 F.3d 1150 (9th Cir. 2005). On appeal, the Ninth Circuit reinstated the indictment. 97See Afshari, 426 F.3d at 1153. In addition to rejecting the district court’s analysis, the Ninth Circuit also considered, and rejected, a First Amendment claim brought by the defendants. 98See id. at 1160. Specifically, the court rejected the argument that the First Amendment protected the defendants’ activities by noting that “[t]hough contributions of money given to fund speech receive some First Amendment protection, it does not follow that all contributions of money are entitled to protection as though they were speech.” 99Id. (footnote omitted) (citing McConnell v. FEC, 540 U.S. 93 (2003); Buckley v. Valeo, 424 U.S. 1 (1976)). The defendants’ contributions were not entitled to First Amendment protections because “[i]n this context, the donation of money could properly be viewed by the government as more like the donation of bombs and ammunition than speech.” 100Id. The court concluded by quoting its own earlier decision in Humanitarian Law Project v. Reno for the often-stated proposition that “[t]here is no First Amendment right to facilitate terrorism.” 101Id. at 1161 (quoting Humanitarian Law Project v. Reno, 205 F.3d 1130, 1133 (2000)) (internal quotation mark omitted).

3. Other Material-Support Prosecutions

a. The Lackawanna Six

In the spring of 2001, six Yemeni-American young men from Lackawanna, New York, a suburb of Buffalo, traveled to Afghanistan. 102For a detailed discussion of the Lackawanna Six prosecution, see Matthew Purdy & Lowell Bergman, Unclear Danger: Inside the Lackawanna Terror Case, N.Y. Times, Oct. 12, 2003, at N1, available at http://www.nytimes.com/2003/10/12/nyregion/12LACK.html?pagewanted=1. While there, they received weapons training at an Al Qaeda training camp, and at least some of them met personally with Osama bin Laden. 103See id. They then returned to the United States—all prior to the September 11 attacks later that year. 104See id. In the year following the attacks, the U.S. government received information about the “Lackawanna Six,” which raised concerns that the group might be an Al Qaeda “sleeper cell.” 105See id. Eventually, all six men pled to charges of providing material support to an FTO based on their training in Afghanistan. 106See id. At no point during the trial or investigation of the Lackawanna Six did the government present any evidence demonstrating that the group was planning to commit or intended to commit a specific act of terrorism. 107See id. The prosecution of the Lackawanna Six was hailed by the government as a major victory against terrorism and was even mentioned by President George W. Bush in his State of the Union address. 108See id.

b. Sami Al-Hussayen

Sami Al-Hussayen was a Saudi Arabian Ph.D. student in computer science at the University of Idaho in Moscow, Idaho. 109See Maureen O’Hagan, A Terrorism Case That Went Awry, Seattle Times, Nov. 22, 2004, http://seattletimes.com/html/localnews/2002097570_sami22m.html. For a detailed description of the Al-Hussayen prosecution, see Alan F. Williams, Prosecuting Website Development Under the Material Support to Terrorism Statutes: Time to Fix What’s Broken, 11 N.Y.U. J. Legis. & Pub. Pol’y 365, 367–73 (2008). He also contributed his time to run websites for a number of Islamic charities. 110See O’Hagan, supra note 109. Although none of the charities were themselves FTOs, the prosecution alleged that it was possible to link, through the sites run by Al-Hussayen, to other Internet locations where users could contribute to FTOs such as Hamas (though Al-Hussayen’s own involvement with these links is disputed). 111Compare Williams, supra note 109, at 368–71 (reciting FBI allegations that Al-Hussayen was involved in website development recruiting fighters for jihad), with O’Hagan, supra note 109 (raising doubts about Al-Hussayen’s involvement). As a consequence of these events, Al-Hussayen was arrested and charged with material support of terrorism (as well as immigration violations). 112See O’Hagan, supra note 109. After a trial, however, the jury acquitted Al-Hussayen of all the terrorism-related charges and some of the immigration charges (the jury hung on the rest). 113See id. Ultimately, Al-Hussayen agreed to be deported in exchange for the remaining charges being dropped. 114Williams, supra, note 109 at 373.

4. Groups Challenging Terrorist Designations

a. The Holy Land Foundation for Relief and Development

The Holy Land Foundation for Relief and Development was an Islamic charitable foundation that operated in the United States beginning in 1989. 115See Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 160 (D.C. Cir. 2003). The Foundation was originally named the “Occupied Land Fund” and changed its name to Holy Land Foundation in 1991. Id. It described itself as “the largest Muslim charity in the United States.” 116Id. (internal quotation marks omitted). In December of 2001, the Holy Land Foundation was designated as a “Specially Designated Global Terrorist” and all of its assets were frozen by the Department of the Treasury pursuant to an Executive Order issued by President Bush following the September 11, 2001 attacks. 117See id.; Exec. Order No. 13,224, 3 C.F.R. 786 (2002). The Executive Order was in turn issued under the authority of the International Emergency Economic Powers Act passed by Congress in 1995. 11850 U.S.C. § 1701 (2006); see 3 C.F.R. 786. The primary basis for the Holy Land Foundation’s designation was evidence indicating that the Foundation “had financial connections to” Hamas and had provided funding to Hamas as well as to charitable organizations and individuals linked to Hamas. 119Holy Land Found., 333 F.3d at 161. Hamas is, of course, the Palestinian group engaged in armed struggle against the Israeli occupation of the Occupied Territories. 120See Clyde Haberman, Israeli Tensions Rise as an Officer Is Found Slain, N.Y. Times, Dec. 16, 1992, at A1. Hamas has been designated as a Specially Designated Terrorist organization since 1995 and a Specially Designated Global Terrorist organization since 2001 121Holy Land Found., 333 F.3d at 159. (it has also been an FTO for the purposes of the material-support statute since 1997). 122Bureau of Counterterrorism, supra note 72.

The Foundation challenged its designation as a terrorist organization on a number of grounds, including a claim that “the government had violated its First Amendment rights [including freedom of association] by prohibiting it from making any humanitarian contributions by blocking its assets.” 123Holy Land Found., 333 F.3d at 164 (citing FEC v. Colo. Republican Fed. Campaign Comm., 533 U.S. 431 (2001)). The D.C. Circuit offhandedly rejected this claim, stating that the law is established that “there is no constitutional right to facilitate terrorism.” 124Id. at 164–65 (quoting Holy Land Found. for Relief & Dev. v. Ashcroft, 219 F. Supp. 2d 57, 81 (D.D.C. 2002), aff’d, 333 F.3d 156 (D.C. Cir. 2003)) (internal quotation marks omitted). The lower court had rejected the Foundation’s First Amendment claims on the same grounds, 125See id. at 164. noting as well that designating the Foundation a Specially Designated Global Terrorist organization and blocking its assets was permissible because these actions “do not prohibit membership in Hamas or endorsement of its views, and therefore do not implicate HLF’s associational rights.” 126Id. at 161 (quoting Holy Land Found., 219 F. Supp. 2d at 81) (internal quotation marks omitted). In the course of holding that “there is no First Amendment right nor any other constitutional right to support terrorists,” the D.C. Circuit cited the Ninth Circuit’s opinion in Humanitarian Law Project v. Reno. 127Id. at 166 (citing Humanitarian Law Project v. Reno, 205 F.3d 1130, 1133 (9th Cir. 2000)).

b. Islamic American Relief Agency

The Islamic American Relief Agency was, like the Holy Land Foundation, an Islamic charity, founded in this case by a Sudanese immigrant in 1985. 128Islamic Am. Relief Agency v. Gonzales, 477 F.3d 728, 730–31 (D.C. Cir. 2007). It too was designated as a Specially Designated Global Terrorist by the Bush Administration in 2004. 129See id. at 731. The factual basis for the Agency’s designation was a bit more complicated than with the Holy Land Foundation, but, in short, came down to allegations that the Agency had provided funding to another entity, which was itself acknowledged to be a terrorist organization. 130 See id. Like the Holy Land Foundation, the Islamic American Relief Agency raised a freedom of association challenge to its designation. 131See id. The D.C. Circuit predictably rejected this claim on the same grounds as before, citing its decision in Holy Land Foundation as well as the Ninth Circuit’s decision in Humanitarian Law Project. 132Id. at 736–37 (citing Holy Land Found., 333 F.3d at 166; Humanitarian Law Project, 205 F.3d at 1133). The court went on to reject the argument that the Supreme Court’s First Amendment case law required proof of a specific intent to advance the terrorist organization’s illegal ends, concluding that this requirement was limited to instances of pure association, not funding. 133Id. at 737.

B. First Amendment Conundrums

The nine prosecutions described above are, of course, merely a small sampling of the hundreds of terrorism-related prosecutions initiated in the United States since the September 11 attacks. 134See U.S. Dep’t of Justice, Nat’l Sec. Div., National Security Division Statistics on Unsealed International Terrorism and Terrorism-Related Convictions (June 6, 2012) (unpublished data chart), http://www.fas.org/irp/agency/doj/doj060612-stats.pdf (reprinting Department of Justice statistics listing 494 terrorism-related convictions obtained by the Department between September 2001 and December 2011, the most recent being the conviction of Tarek Mehanna). These particular cases were selected both because of their prominence and because their facts illustrate nicely the complex First Amendment issues that have arisen in the course of the War on Terror (albeit, these issues have been treated dismissively by the courts). In the next Part, this Article considers how First Amendment rights, especially the right of association, interact with terrorism prosecutions and how the lower courts’ treatment of these rights comports with previously well-established First Amendment principles. 135See infra Part III. We set the stage for this broader discussion by here highlighting some of the issues raised by these particular cases.

Two immediate points jump out from the cases described above. First, in none of them (as in Humanitarian Law Project itself) was there any evidence that the defendants had knowledge of or actually helped plan, much less participated in, a specific act of terror. Thus, despite the sometimes punitive sentences imposed on these defendants, none of their prosecutions can be said to fall within the uncontroversial core of the government’s anti-terrorism efforts. Second, the activities that formed the primary, often exclusive, basis of the described prosecutions (except the Lackawanna Six)—running websites, posting material on the Internet, distributing television signals, creating and posting videos, and providing financial support—are activities that normally would receive strong First Amendment protections. Despite the lower courts’ dismissive treatment, then, the First Amendment issues raised by these prosecutions are serious and central to the cases, rather than frivolous and peripheral.

To begin with, these cases illustrate the grave ambiguities created by the Supreme Court’s Humanitarian Law Project decision. 136See supra Part I. Consider in this regard the Humanitarian Law Project Court’s conclusion that the material-support statute does not violate the First Amendment right of association because it does not bar membership, 137Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010). combined with the Court’s insistence that the statute does not directly regulate speech because it does not prevent independent advocacy in support of FTOs. 138See id. at 2722–23. But what does “membership” mean in this context; what exactly does the statute permit (and prohibit)? Tarek Mehanna was prosecuted for providing material support despite the lack of any evidence that he had direct contact with an FTO. 139See supra text accompanying notes 54–66. Sami Al-Hussayen was charged (albeit acquitted) of material-support charges based on his provision of website services to charities that were themselves not FTOs based on possible financial links between those charities and an FTO. 140See supra text accompanying notes 109–14. What then is left of “independent” advocacy, and concomitantly, what is enough to cross the line into coordination? And what exactly does it mean to say that “membership” is legal, if speech in support of an FTO, combined with any sort of “membership,” no matter how attenuated, crosses the line into material support? 141For a recent judicial discussion of these difficulties, see Hedges v. Obama, 890 F. Supp. 2d 424, 466 (S.D.N.Y. 2012), vacated, 724 F.3d 170 (2d Cir. 2013). For a similar argument that the Court’s reasoning leaves little substance to the supposedly preserved right of association, see Margaret Tarkington, Freedom of Attorney–Client Association, 2012 Utah L. Rev. 1071, 1082.

The Iqbal prosecution raises different, but also serious questions. Iqbal and Elahwal, remember, were prosecuted for retransmitting a television signal on behalf of an FTO (and in exchange for financial compensation). 142See supra text accompanying notes 67–77. There can be no doubt that their actions benefited and supported Hezbollah, of course. But that conduct also constituted speech, normally protected by the First Amendment. 143See Daphne Barak-Erez & David Scharia, Freedom of Speech, Support for Terrorism, and the Challenge of Global Constitutional Law, 2 Harv. Nat’l Security J. 1, 3 (2011). The trial judge’s rejection of Iqbal’s First Amendment defense on the grounds that he was being prosecuted for “conduct” and not speech (in part because he was paid to retransmit Al Manar) is difficult to reconcile with myriad cases protecting under the First Amendment both the expenditure of money to finance speech 144See, e.g., Citizens United v. FEC, 558 U.S. 310, 319 (2010) (holding that the Government may not prohibit independent expenditures by corporations used to fund political speech). and the republication of another’s speech for compensation. 145See, e.g., N.Y. Times Co. v. Sullivan, 376 U.S. 254, 266 (1964) (“[I]f the allegedly libelous statements would otherwise be constitutionally protected . . . , they do not forfeit that protection because they were published in the form of a paid advertisement.”). It is true that Iqbal’s conviction can probably be defended under the Supreme Court’s strict scrutiny analysis in Humanitarian Law Project, but those were not the grounds advanced by the court (the Iqbal prosecution predated Humanitarian Law Project).

Several of the other prosecutions discussed above (including the Holy Land Foundation, Islamic American Relief Agency, Hammoud, and Afshari/Rahmani cases) together raise yet another issue regarding the scope of First Amendment protections in the context of terrorism prosecutions. In all of these cases, the sole legal basis for either prosecuting the defendants under the material-support statute or designating them as a Specially Designated Global Terrorist (with the serious attendant financial consequences) was the fact that the individual or entity provided financial support to an FTO. 146See supra Part II.A. Furthermore, in each of these cases, the courts rejected a First Amendment defense on the grounds that the Constitution does not protect a right to fund terrorism because funding is not protected speech or association. 147See supra Part II.A. In this regard, the decisions seem consistent with the Supreme Court’s equally cavalier conclusion that the material-support statute did not violate freedom of association because it did not ban membership, only material support. 148See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010). In other words, both the Supreme Court and these lower courts simply assumed that money is not association. The Hammoud court built on this idea by describing the provision of material support as “conduct” and not “association.” 149See supra note 90 and accompanying text. The difficulty, as we shall see in the next Part, is that this statement is flatly inconsistent with numerous decisions in other areas of law, raising the question of whether the right of association means something different in the terrorism context than elsewhere (and if so, why). 150See infra Part III.

Finally, we come to the Lackawanna Six. In some ways, this prosecution seems the most straightforward of those described above because the actions for which the Six were prosecuted—obtaining weapons training from Al Qaeda—would not appear to be conduct conceivably protected by the First Amendment. 151See supra text accompanying notes 102–08. Indeed, insofar as some of the other prosecutions (notably of Ahmad and Mehanna) can be tied to terrorist training, that would appear to help avoid First Amendment pitfalls. But in fact, appearances can be deceptive. The difficulty is that the material-support statute does not itself prohibit obtaining weapons training. 152There is a separate statute that does prohibit obtaining military training from a terrorist group, but it was not enacted until 2004 and so could not be invoked in the Lackawanna Six prosecution. See 18 U.S.C. § 2339D (2012). For a complex set of reasons, although this statute was passed in 2004, as of 2009, only a single prosecution had been brought under it. See Robert M. Chesney, Terrorism, Criminal Prosecution, and the Preventive Detention Debate, 50 S. Tex. L. Rev. 669, 688–89 & n.53 (2009) (citing Criminal Complaint at 1, United States v. Maldonado, No. 07-cr-124 (S.D. Tex. Feb. 13, 2007)). However, there are cases currently pending, which charge violations of § 2339D. See, e.g., United States v. Medunjanin, No. 10 CR 019, 2012 WL 1514766, at *6 (E.D.N.Y. May 1, 2012); United States v. Ahmed, No. 10 Cr. 131 (PKC), 2012 WL 983545, at *1 (S.D.N.Y. Mar. 22, 2012). Of course, the weapons training imparted knowledge to the defendants, but § 2339B prohibits individuals from providing material support to an FTO, not vice-versa. 15318 U.S.C. § 2339B(a)(1) (2012). And on the facts of the Lackawanna Six litigation, there was a notable absence of proof that the defendants’ training had benefited Al Qaeda in any way, given the defendants’ failure to plan for or take any overt acts toward acts of violence. 154See supra note 107 and accompanying text. Seen through this lens, there is a strong argument that the Lackawanna Six were successfully prosecuted for providing material support simply because they associated with (joined?) Al Qaeda, a result in obvious tension with the Humanitarian Law Project Court’s saving interpretation of § 2339B as permitting simple membership.

In short, the treatment of the First Amendment in terrorism prosecutions is a mess. It is unclear what exactly the line is between protected “independent” and illegal “coordinated” advocacy in support of an FTO. It is unclear what forms of “pure” association with an FTO are protected and which cross the line into material support. And it is unclear why even speech explicitly undertaken on behalf of an FTO is unprotected, even though in other contexts such political speech is treated as lying at the core of First Amendment protections. In the next Part, this Article examines these tensions in a more systematic way and places them in the context of modern First Amendment doctrine before seeking a path out of this thicket in the last two Parts.

III. Terrorism and the First Amendment

This Article seeks to examine the relationship between terrorism and the First Amendment. Until now, it has focused on the facts and legal issues associated with a number of terrorism prosecutions. We now turn to the First Amendment. What rights does the First Amendment protect that might be implicated by the War on Terror?

As always, it is useful to begin with the text. The First Amendment states (leaving aside the Religion Clauses) that “Congress shall make no law . . . abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.” 155U.S. Const. amend. I. Of these provisions, the Speech and, to a lesser extent, the Press Clauses of the First Amendment have received by far the most judicial and scholarly attention, so we will start with them. It is, of course, well established that the First Amendment generally protects speech unless the speech falls within a well-established category of unprotected speech such as obscenity or defamation, 156United States v. Stevens, 130 S. Ct. 1577, 1584 (2010). But see James Weinstein, Participatory Democracy as the Central Value of American Free Speech Doctrine, 97 Va. L. Rev. 491, 491–92 (2011) (raising questions about the accuracy of this view). and it is equally well established that within the realm of protected speech, speech on the topic of politics or public affairs receives the highest level of protection. 157See e.g., Knox v. Serv. Emps. Int’l Union, Local 1000, 132 S. Ct. 2277, 2288 (2012) (“Our cases have often noted the close connection between our Nation’s commitment to self-government and the rights protected by the First Amendment.”); Snyder v. Phelps, 131 S. Ct. 1207, 1215 (2011) (explaining that speech on public issues has the most protection under the First Amendment); N.Y. Times Co. v. Sullivan, 376 U.S. 254, 269–70 (1964) (considering the case before it “against the background of a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open”). Finally, the Court has repeatedly emphasized that the First Amendment’s protections for political speech extend to all ideas, no matter how distasteful or in opposition to the existing social order. 158See Snyder, 131 S. Ct. at 1216–17 (quoting Rankin v. McPherson, 483 U.S. 378, 387 (1987)); see also Whitney v. California, 274 U.S. 357, 377 (1927) (Brandeis & Holmes, JJ., concurring), overruled per curiam by Brandenburg v. Ohio, 395 U.S. 444 (1969). Under these generous standards, there can be little doubt that most speech in support of terrorism constitutes political speech, presumptively entitled to the highest level of constitutional protection, and no judge or scholar has seriously questioned this conclusion.

The First Amendment’s protections are not, however, absolute. The Court has long recognized that incitement—speech that encourages others to engage in violence—constitutes unprotected speech that may constitutionally be suppressed. 159For a brief description of the evolution of the incitement doctrine, see Bhagwat, supra note 7, at 1003–05. The modern Court, however, has defined the category of incitement extremely narrowly. In its landmark 1969 decision in Brandenburg v. Ohio, the Court held that speech may be punished as incitement only if it is “directed to inciting or producing imminent lawless action and is likely to incite or produce such action.” 160395 U.S. at 447. In principle (though as we have seen not necessarily in practice), speech urging others to engage in acts of terrorism or violent jihad should be punishable only if it meets the Brandenburg standard, a very high threshold indeed.

Incitement and Brandenburg, however, are not the focus of this Article (though they are not irrelevant either). 161For a detailed discussion of the undermining of the Brandenburg test in the War on Terror, see generally Thomas Healy, Brandenburg in a Time of Terror, 84 Notre Dame L. Rev. 655 (2009). Nor indeed is free speech. Instead, this Article focuses on another, equally significant right, which has been given short shrift by the courts, the freedom of association, which is rooted not in the Speech and Press Clauses but rather in the Assembly Clause of the First Amendment. Let us start with text. The First Amendment explicitly protects freedom of peaceable assembly as an independent and coequal right to freedoms of speech and the press. 162See U.S. Const. amend. I. Moreover, as John Inazu has extensively documented, throughout most of our history until the mid-twentieth century, the freedom of assembly was treated, by courts and the public alike, as an extremely significant right, central to the American experience of self-governance. 163See Inazu, supra note 47, at 20–62 (providing a historical overview of the freedom of assembly in the United States); see also Tabatha Abu El-Haj, The Neglected Right of Assembly, 56 UCLA L. Rev. 543, 554–61 (2009) (discussing the relationship between assembly and politics in American history); Bhagwat, supra note 7, at 984–85 (discussing the treatment of the freedom of assembly by the courts in the first half of the twentieth century); Robert M. Chesney, Democratic-Republican Societies, Subversion, and the Limits of Legitimate Political Dissent in the Early Republic, 82 N.C. L. Rev. 1525, 1526, 1550–51 (2004) (highlighting the importance of the freedom of assembly); James Gray Pope, Republican Moments: The Role of Direct Popular Power in the American Constitutional Order, 139 U. Pa. L. Rev. 287, 330–32, 337–41 (1990) (discussing the origin of the freedom of assembly). During the first half of the twentieth century, in particular, the Supreme Court vigorously asserted and enforced an independent right, which it variously denoted as “assembly” and “association,” of citizens to join together in both temporary and permanent groups for a variety of purposes, including especially political ones, and also clarified that this right extended to “subversive” groups such as the Communist Party. 164See Bhagwat, supra note 7, at 983–85 (discussing, among other cases, De Jonge v. Oregon, 299 U.S. 353 (1937), in which the right to assemble with the Communist Party was protected); David Cole, Hanging with the Wrong Crowd: Of Gangs, Terrorists, and the Right of Association, 1999 Sup. Ct. Rev. 203, 216 & n.39. Furthermore, cases during this period described the right of assembly as “cognate to those of free speech and free press and . . . equally fundamental.” 165E.g., De Jonge, 299 U.S. at 364; accord Thomas v. Collins, 323 U.S. 516, 530 (1945). Beginning with the Supreme Court’s 1958 decision in NAACP v. Alabama ex rel. Patterson, 166357 U.S. 449 (1958). however, the Court gradually abandoned the independent right of assembly/association, replacing it with a truncated, nontextual right of “expressive association” that protects only the right to associate for speech-related purposes. 167The decline of the independent right of assembly/association and the rise of expressive association has been extensively recounted elsewhere, and here I aim only to briefly summarize the results of that scholarship. See Inazu, supra note 47, at 61–62, 79–84, 132–35, 141–44; Bhagwat, supra note 7, at 985–89. Indeed, John Inazu notes that the Supreme Court has not decided an Assembly Clause case since 1983. 168See Inazu, supra note 47, at 62.

Finally, it should be noted that in the modern era, the right of assembly/association logically can and should be extended to online forms of communication and association (which was the primary form of association at issue in several of the prosecutions discussed above, including Tarek Mehanna’s). 169See supra Part II.A. As John Inazu has extensively argued, in the Internet age much of the associative activities citizens engage in are online, and indeed it is becoming more and more difficult to distinguish among physical, distant offline, and online forms of assembly and association. 170See John D. Inazu, Virtual Assembly, 98 Cornell L. Rev. 1093, 1094–95 (2013). People’s online and offline interactions have come to merge and act in tandem to the point where they cannot be untangled. Furthermore, as Inazu argues, online groups and communications have played an increasingly central role in political activism and self-governance, including most obviously in both of President Obama’s presidential campaigns. 171See id. at 1109–10. Given the importance of online groups in citizens’ lives, the critical role the Internet plays in modern civic activism, and the close analogy between the traditional form of physical assembly known to the Framers and the nature of modern online forums, there is no logical reason to restrict the right of assembly and association to offline conduct.

For all of the Supreme Court’s neglect, an examination of history thus leaves no serious doubt that the First Amendment is best read to protect an independent right to form groups and otherwise associate, both physically and digitally, with fellow citizens for purposes relevant to democratic self-governance. 172For a further defense of this principle, see Cole, supra note 164, at 226–29. For the purposes of my analysis, I will assume that such an independent right of assembly/association exists 173I will henceforth describe the right as “freedom of association” because that is the dominant modern nomenclature. There is some historical support for the proposition that the term “assembly” contained in the constitutional text protected temporary gatherings, while the term “association” referred to more permanent groups, whose constitutional status was a bit less clear. See Jason Mazzone, Freedom’s Associations, 77 Wash. L. Rev. 639, 742–43 (2002). However, as the discussion in the text indicates, throughout most of our history the public and the courts used the terms assembly and association interchangeably and clearly presumed constitutional protection for permanent groups. and may be asserted by defendants in terrorism prosecutions (though as we will see, to say that the right exists does not mean that it is limitless). 174See infra Part IV. It should be noted, however, that much of the discussion that follows would not be significantly altered if the freedom of association is limited, as the modern Supreme Court seems to assume, to expressive association. This is because almost all FTOs are highly expressive in that their purpose is to advance and propagate a political and/or religious ideology, and one of the primary means they employ to do so is through speech. This statement is true of Hezbollah, Hamas, the PKK, the Tamil Tigers, the LeT, and indeed Al Qaeda itself—as illustrated by the large percentage of terrorism prosecutions that ultimately rest on speech. This is not to say that these organizations are exclusively or even primarily expressive—but neither obviously are the Boy Scouts, yet the Supreme Court has accorded full expressive association rights to that group. 175See Boy Scouts of Am. v. Dale, 530 U.S. 640, 655 (2000). If, as I argue, the freedom of association is a stand-alone right, then a fortiori it would seem to extend to citizens seeking to associate with FTOs. And if it is limited to expressive association, nevertheless the mixed expressive/non-expressive nature of FTOs is not reason enough to deny individuals the right to associate with them. If FTOs’ associational rights can be restricted, it must be on some other ground.

Now we may return to terrorism and its relationship to the right of association. Our starting point must be that invoking the word “terrorism” cannot end all analysis, any more than the word “Communism” should have in an earlier era. Indeed, in Humanitarian Law Project the Supreme Court appeared to acknowledge this point by interpreting the material-support statute to not bar simple membership in FTOs, implying that it could not do so constitutionally. 176See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2723 (2010). In fact, however, the actual contours of the Humanitarian Law Project analysis, in combination with lower court decisions such as those recounted above in Part II, 177See supra Part II. have left very little substance to this concession, in the process eviscerating the right of association itself. Freedom of association means that individuals may not be prosecuted for membership in FTOs. But what, in this context, does “membership” mean? It would appear that the answer is “almost nothing.” As Wadie Said and David Cole point out, the membership protected by Humanitarian Law Project is entirely hollow—it does not permit any acts on behalf of an FTO because that would constitute providing “personnel” in violation of § 2339B, nor does it even permit the payment of membership fees to an FTO because that would constitute material support in the form of funds. 178See Wadie E. Said, Humanitarian Law Project and the Supreme Court’s Construction of Terrorism, 2011 BYU L. Rev. 1455, 1507 & n.241 (citing David Cole, Terror Financing, Guilt by Association and the Paradigm of Prevention in the “War on Terror,” in Counterterrorism: Democracy’s Challenge 242 (Andrea Bianchi & Alexis Keller eds., 2008)). The Holy Land Foundation and Hammoud cases, discussed earlier, illustrate this point nicely. See supra Part II.A.

This point is further illustrated by the prosecution of the Lackawanna Six. Recall that in that case, the defendants were found guilty of providing material support to Al Qaeda on the basis of their attending an Al Qaeda training camp, where they received weapons training, and meeting with Osama bin Laden (all before the September 11, 2001, attacks). 179See supra note 103 and accompanying text. Given the fact, noted earlier, that the government failed to prove any benefit that the Six provided to Al Qaeda, 180See supra note 107 and accompanying text. the implication of this case seems clear: joining an FTO, meeting with its leaders, and taking actions suggesting support for the organization’s goals alone is sufficient to violate the material-support statute. But if this alone crosses the line, what exactly is the “membership” that the Humanitarian Law Project Court claimed remains protected conduct? In short, whatever the Supreme Court said in Humanitarian Law Project, in practice the right of membership in an FTO is wholly fictitious because any individual foolish enough to invoke it is likely to be found in violation of § 2339B. 181For a similar argument, see Cole, supra note 164, at 247–48.

The muddle increases when one considers Tarek Mehanna’s case. The government argued that Mehanna’s speech, with no accompanying conduct, violated the material-support statute because it was sufficiently coordinated with Al Qaeda. 182See Pyetranker, supra note 54, at 38. As noted earlier, in affirming Mehanna’s conviction, the First Circuit abandoned this theory, almost conceding its legal insufficiency. See supra note 58. But as noted earlier, that claim of coordination was not based on any strong evidence of ongoing dialogue between Mehanna and Al Qaeda leaders, but rather based on Mehanna’s stated belief that he was acting as a “wing” of Al Qaeda. 183 See Abel, supra note 56, at 732 (citing Government’s Proffer and Memorandum in Support of Detention, supra note 56, at 11). There are two puzzles here. First, the trial judge’s rejection of Mehanna’s First Amendment defense reveals an extremely odd implication of Humanitarian Law Project when we juxtapose two separate holdings in the latter case. In analyzing the free speech claim in Humanitarian Law Project, the Court concluded that the material-support statute was not a flat ban on political speech because it condemned only speech “to, under the direction of, or in coordination with” an FTO, not independent advocacy. 184Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2722–23 (2010). But in rejecting the association claim, the Court held that membership in an FTO alone is protected conduct. 185See id. at 2730. Presumably, however, active membership in an FTO would be sufficient to establish coordination, especially given the prosecution in Mehanna based on coordination with an FTO despite minimal contact between Mehanna and Al Qaeda members. 186See Pyetranker, supra note 54, at 38. Put Humanitarian Law Project and Mehanna together and under current law, one may join an FTO, and one may speak in favor of an FTO, but not both. This is bizarre. After all, what is the point of membership in a group one cannot speak in support of? Indeed, is it not the whole point of the Court’s “expressive association” jurisprudence that speech engaged in through associations is especially favored constitutionally? More broadly, there is a strong argument to be made that, for reasons both historical and theoretical, when the political rights protected by the First Amendment, such as speech and association, are exercised in tandem, they deserve special solicitude. 187This argument is laid out in detail in Bhagwat, supra note 7, at 995–1002. It should be noted in this regard that the original drafts of the Assembly Clause, including James Madison’s original proposal to Congress, 188The Complete Bill of Rights: The Drafts, Debates, Sources, and Origins 129 (Neil H. Cogan ed., 1997). the proposals that emerged from state ratifying conventions, 189Id. at 140. and the version in George Mason’s Master Draft of the Bill of Rights that provided the basis for those proposals, 190George Mason’s Master Draft of the Bill of Rights, Constitution.org., http://www.constitution.org/gmason/amd_gmas.htm (last updated Jan. 24, 2012). all described a right of the people “peaceably to assemble together and consult for their common good” or a very close approximation. This right was also inevitably linked to the right to petition the legislature for a redress of grievances. 191See The Complete Bill of Rights: The Drafts, Debates, Sources, and Origins, supra note 188, at 129, 140; George Mason’s Master Draft of the Bill of Rights, supra note 190. In other words, the right of assembly was a right to gather together with fellow citizens, to speak together, and to speak in unison to those in power. 192See John D. Inazu, The Forgotten Freedom of Assembly, 84 Tul. L. Rev. 565, 573–77 (2010). By effectively banning all speech in the context of membership, Humanitarian Law Project and Mehanna appear to stand these principles on their head.

A second, even more significant implication of the government’s theory in the Mehanna case is that just as the right of association (through membership) that the Humanitarian Law Project Court claimed was not barred by the material-support statute turns out to be wholly illusory, so too does the right of independent advocacy on behalf of an FTO. The roots of the problem lie in Humanitarian Law Project itself. While, as noted above, the Humanitarian Law Project majority held that the material-support statute did not bar independent advocacy, in just the previous section of the opinion the Court had refused to provide any guidance whatsoever on what forms of communication or conduct might cross the line into coordination. 193See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2722 (2010). Building on this lack of clarity, in the Mehanna case, as noted above, coordination was alleged and allowed to go to the jury based on the defendant’s actions alone, taken in response to a general plea from an FTO. 194See Abel, supra note 56, at 732. This suggests that coordination/membership can be unilateral, which is certainly a rather unusual view of what association means. But more fundamentally, it makes the safe harbor of independent advocacy very dangerous indeed, because the supposedly protected speech itself can become proof of coordination. What confidence could any speaker have, then, that speech in favor of an FTO will not result in a successful prosecution?

Perhaps the underlying theory behind the Mehanna prosecution and the trial court’s approach was that speech such as Mehanna’s, which actively sought to recruit individuals to perform violent acts on behalf of Al Qaeda, crosses the line between protected abstract advocacy and unprotected incitement. This, however, makes a shambles of long-established principles of free speech law. How, for example, can this approach be reconciled with Justice Holmes’s memorable comment in his Gitlow dissent that “[e]very idea is an incitement,” 195Gitlow v. New York, 268 U.S. 652, 673 (1925) (Holmes, J., dissenting). The Gitlow dissent, arising out of the anticommunist prosecutions of the first Red Scare, is, of course, one of the foundational opinions of modern First Amendment jurisprudence. the very point of which was to reject a distinction between abstract advocacy and incitement, insisting instead on proof of actual and imminent harm? More fundamentally, this attempted distinction simply cannot be reconciled with Brandenburg, the foundational case in this area (and the direct heir to Justice Holmes’s dissent in Gitlow), which insists that the First Amendment does “not permit a State to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action.” 196Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (per curiam). The Mehanna prosecution did not appear to make any effort to prove that Mehanna’s website posed a threat of inciting imminent and likely violence, after all. 197See Abel, supra note 56, at 732. Instead, the government (successfully) sought to avoid Brandenburg altogether by recasting Mehanna’s actions as “coordination” and “material support.” 198See id. at 732–33. Such an approach could just as easily have been used to justify the now-discredited anticommunist prosecutions of the first and second Red Scares. 199See, e.g., Dennis v. United States, 341 U.S. 494, 497–98, 509–11 (1951); Whitney v. California, 274 U.S. 357, 363–64, 371–72 (1927), overruled per curiam by Brandenburg, 395 U.S. 444; Gitlow, 268 U.S. at 655–56, 664–67; Abrams v. United States, 250 U.S. 616, 617, 624 (1919). In short, the Mehanna trial court’s understanding of the Humanitarian Law Project decision threatens to undermine a well-settled line of precedent that historically and logically lies at the heart of free speech law.

Nor is Mehanna the only post–September 11 terrorism prosecution in tension with Brandenburg. Jubair Ahmad’s conviction was much like Mehanna’s in that, at heart, it was based on speech seeking to incite violent action, but there was no proof, or even hint, of likely and imminent violence in response to Ahmad’s speech. 200See Karas, supra note 78. Admittedly, in the Ahmad prosecution the proof of coordination with an FTO was much stronger than with Mehanna because Ahmad acted at the instigation and under the direction of an LeT leader. 201See id. But the “material support” he provided to LeT consisted of nothing more than inciting speech. In effect, under the theory of prosecutions such as Ahmad’s, the Brandenburg test applies only to purely unilateral speech, not speech in coordination with an FTO, making speech plus association a crime where neither, alone, could be prosecuted.

Brandenburg and the law of incitement, victims though they are of the War on Terror, 202See Healy, supra note 161, at 655; supra text accompanying note 161. are not, however, the primary focus of this Article; the focus is, rather, association. As discussed earlier, while the Supreme Court in Humanitarian Law Project purported to preserve a right to association with FTOs through “membership,” terrorism prosecutions in the lower courts (and for that matter, Humanitarian Law Project itself) have narrowed this right to the point of meaninglessness. 203See supra Parts I.B, II.A. But this is not the only way in which terrorism prosecutions conflict with, and so undermine, decisions protecting the right of association. Consider in this regard the Humanitarian Law Project Court’s argument that one of the reasons Congress could constitutionally ban even the provision of nonviolent material support to FTOs is that such support can “lend legitimacy” to those organizations, and so facilitate their illegal ends. 204Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2725 (2010). As Justice Breyer pointed out in dissent, this justification cannot really distinguish independent from coordinated advocacy in favor of an FTO, because either form of speech can act to “legitimate” an FTO. 205See id. at 2736 (Breyer, J., dissenting). More fundamentally, this reasoning also cannot be reconciled with the majority’s understanding that § 2339B does not forbid membership in an FTO. 206 See id. at 2730 (majority opinion). After all, the choice of an individual not otherwise linked to illegal activity to join an FTO surely lends substantial legitimacy to the organization. Indeed, this theory of legitimation provided one of the primary motivations for the infamous prosecution of Anita Whitney for membership in the Communist Party during the first Red Scare, 207See Ashutosh A. Bhagwat, The Story of Whitney v. California: The Power of Ideas, in Constitutional Law Stories 407, 411–12 (Michael C. Dorf ed., 2004). a prosecution and theory that was sharply criticized by Justice Brandeis in one of the most influential First Amendment opinions in the history of the Supreme Court. 208See Whitney v. California, 274 U.S. 357, 379 (1927) (Brandeis & Holmes, JJ., concurring), overruled per curiam by Brandenburg v. Ohio, 395 U.S. 444 (1969); Bhagwat, supra note 207, at 407 (quoting Vincent Blasi, The First Amendment and the Ideal of Civic Courage: The Brandeis Opinion in Whitney v. California, 29 Wm. & Mary L. Rev. 653, 668 (1988)). Moreover, as Justice Breyer also points out, the Humanitarian Law Project majority’s legitimacy argument simply cannot be reconciled with other, foundational right of association cases which uphold the right to join the Communist Party so long as one did not intend to further its violent ends, because such membership surely conferred legitimacy. 209See Humanitarian Law Project, 130 S. Ct. at 2737 (Breyer, J., dissenting) (citing United States v. Robel, 389 U.S. 258 (1967); Keyishian v. Bd. of Regents of Univ. of State of N.Y., 385 U.S. 589, 605–10 (1967); Elfbrandt v. Russell, 384 U.S. 11, 17 (1966); Scales v. United States, 367 U.S. 203, 228–30 (1961); De Jonge v. Oregon, 299 U.S. 353 (1937)). Once again, therefore, the judicial reasoning in terrorism cases is found to be in seemingly irreconcilable conflict with extant First Amendment jurisprudence, a conflict that courts do not even acknowledge, much less seek to reconcile. 210The Humanitarian Law Project majority’s only effort to respond to Justice Breyer’s argument was a cryptic footnote stating (without explanation) that because § 2339B does not ban “pure speech and association,” without more, reliance on the Communist Party cases is “unfounded.” Id. at 2723 n.4 (majority opinion) (internal quotation marks omitted).

Finally, yet another conflict between the analysis of association rights in terrorism cases and broad First Amendment doctrine concerns the treatment of financial contributions. The starting point here is to recognize that a large fraction of the terrorism cases discussed earlier, including Holy Land Foundation, Islamic American Relief Agency, Hammoud, and Afshari, were based exclusively on the contribution of money to an FTO. 211See supra Part II.A. In response to claims that such punishment violated rights of association, the standard judicial response has been that “there is no First Amendment right nor any other constitutional right to support terrorists.” 212Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 166 (D.C. Cir. 2003); accord Islamic Am. Relief Agency v. Gonzales, 477 F.3d 728, 736–37 (D.C. Cir. 2007); Humanitarian Law Project v. Reno, 205 F.3d 1130, 1133 (9th Cir. 2000). Alternatively, the Hammoud court dismissed First Amendment concerns because the material-support statute “does not prohibit mere association; it prohibits the conduct of providing material support to a designated FTO,” 213United States v. Hammoud, 381 F.3d 316, 329 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005). and similarly in Afshari, the court concluded that the donations at issue were “more like the donation of bombs and ammunition than speech.” 214United States v. Afshari, 426 F.3d 1150, 1160 (9th Cir. 2005). Moreover, these statements seem to follow the Humanitarian Law Project Court’s own reasoning in rejecting the plaintiffs’ association claim on the grounds that the material-support statute does not forbid membership in an FTO, and so does not penalize “simple association or assembly,” 215Humanitarian Law Project, 130 S. Ct. at 2730. clearly implying that the provision of material support, including financial support, is not “mere association.”

The problem is how to reconcile this reasoning with statements by the Supreme Court and lower courts in other contexts, notably campaign finance reform, that financial contributions are a form of association protected by the First Amendment. 216See, e.g., FEC v. Colo. Republican Fed. Campaign Comm., 533 U.S. 431, 440 (2001) (“Spending for political ends and contributing to political candidates both fall within the First Amendment’s protection of speech and political association.”); Wis. Right to Life State Political Action Comm. v. Barland, 664 F.3d 139, 152 (7th Cir. 2011) (quoting Colo. Republican, 533 U.S. at 440); SpeechNow.org v. FEC, 599 F.3d 686, 692 (D.C. Cir. 2010) (quoting Davis v. FEC, 554 U.S. 724, 740 n.7 (2008)); Carver v. Nixon, 72 F.3d 633, 636 (8th Cir. 1995) (quoting Buckley v. Valeo, 424 U.S. 1, 18 (1976)); Casino Ass’n of La. v. State, 2002-0265, p. 8 (La. 6/21/02); 820 So. 2d 494, 500. Indeed, the presumption that financial contributions are a form of protected association forms the bedrock of an entire line of Supreme Court decisions, starting with the seminal decision in this area, Buckley v. Valeo, 217424 U.S. at 18 (“[T]he present Act’s contribution and expenditure limitations impose direct quantity restrictions on political communication and association by persons, groups, candidates, and political parties.”). that subject laws limiting financial contribution to political candidates to close First Amendment scrutiny. 218See, e.g., Randall v. Sorrell, 548 U.S. 230, 247 (2006); Colo. Republican, 533 U.S. at 440; Nixon v. Shrink Mo. Gov’t PAC, 528 U.S. 377, 387–88 (2000). Why should it be that a contribution to a political candidate or political action group constitutes protected association, while a similar contribution to an FTO is entirely unprotected? The only hint of a judicial answer, raised in passing by the Hammoud court, is that the right of association is limited to expressive associations, and FTOs do not qualify. 219United States v. Hammoud, 381 F.3d 316, 328 n.3 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005) (“Hammoud relies in part on cases holding that a donation to a political advocacy group is a proxy for speech. Hizballah is not a political advocacy group, however.” (citation omitted)). This, however, is most unconvincing. First of all, as noted earlier, the flat assertion that FTOs are not expressive seems incomprehensible in light of the extensive political and expressive activities of most FTOs, including most especially Hezbollah (the FTO at issue in Hammoud). 220See supra notes 84–91 and accompanying text. After all, Iqbal was prosecuted for rebroadcasting Hezbollah’s television station! 221See Weiser, supra note 67. More broadly, however, for reasons discussed earlier, 222See supra notes 166–68 and accompanying text. it seems incorrect as a matter of history, text, and theory to limit the right of association to expressive associations. Instead, the Assembly Clause of the First Amendment is best understood to protect a free-standing right to associate with organizations whose activities and expression are relevant to the political process. 223See Bhagwat, supra note 7, at 990–94. Under that broad definition, at first cut FTOs surely qualify for protection because they are surely political. 224Though, as is discussed in more detail in the next Part, there are alternative grounds upon which exclusion of FTOs from First Amendment protections might be grounded. Once the expressive/nonexpressive distinction is rejected, therefore, the terrorism cases simply cannot be reconciled with the scope of freedom of association in other contexts.

Why does all of this matter? After all, surely the First Amendment does not prohibit the U.S. government from stopping its citizens from providing funds or other tangible aid to terrorist organizations, which can be used to attack U.S. citizens? Indeed, is it really plausible or realistic to expect the government to take no action against acknowledged members of FTOs until an attack is “imminent” and “likely”? Probably not. The problem is that the judiciary’s current failure to acknowledge or grapple with these contradictions at all undermines the rule of law. This is itself a dangerous tactic in a conflict that turns so much on convincing the hearts and minds of potential adversaries. But in addition, broad judicial statements dismissing seemingly substantial First Amendment concerns as frivolous threaten to undermine First Amendment rights—including, in particular, associational rights—in other contexts by providing language that “lies about like a loaded weapon” and can be drawn upon by future litigants and courts. 225Korematsu v. United States, 323 U.S. 214, 246 (1944) (Jackson, J., dissenting) (arguing that judicial statements sanctioning a racially discriminatory order could be used for harmful ends in future proceedings). Given the central role that First Amendment liberties play in the maintenance of our system of democratic government, this seems a grave risk. Better, surely, to tackle these problems directly and develop a coherent framework within which to reconcile widely held instincts regarding appropriate limits on activities that aid terrorism with the words and law of the First Amendment. It is to that task that we now turn.

IV. Peaceable Association

One clear and abiding lesson emerging from the previous discussion is that, in the course of upholding the criminal prosecutions that lie at the heart of the domestic War on Terror, courts are systematically ignoring long-standing First Amendment principles, and in particular, are tearing apart established definitions of the right of assembly and association. The watered-down associational right that emerges from the terrorism cases, a right that theoretically permits membership in an FTO (though as we noted, even that is doubtful), but clearly precludes combining membership with communications with an FTO, speech on behalf of an FTO, or any financial contribution to an FTO, has no relation to the robust right to freedom of association protected throughout most of our history. Even the truncated right of “expressive association” protected by the modern Court has been understood to protect a more robust form of membership than the terrorism cases acknowledge. 226See, e.g., Randall v. Sorrell, 548 U.S. 230, 236–37 (2006) (protecting an individual’s right to associate with a political candidate by providing financial contributions); NAACP v. Button, 371 U.S. 415, 428–29 (1963) (protecting an association’s right to provide legal representation to members); Shelton v. Tucker, 364 U.S. 479, 485–86, 490 (1960) (protecting an individual’s right not to disclose associational ties); NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 466 (1958) (protecting an association’s right to maintain its members’ anonymity). But when one takes into account the broad, free-standing right of association, rooted in the Assembly Clause, that has been recognized throughout most of our history, the damage is even more clear. For example, there seems to be little doubt that such a right of association must protect a right to pool financial resources—i.e., for individuals to contribute to associations—or the right would have little meaning because for associations to either speak or act, typically they must expend funds. The bland dismissal in the terrorism cases of a “right to support terrorists” seems irreconcilable with this fact and the history that lies behind it. 227See, e.g., Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 166 (D.C. Cir. 2003). At the same time, it does seem intuitively unlikely that the Constitution protects a right to provide Al Qaeda with funds that the organization can use to purchase explosives—after all, the Constitution is not a “suicide pact.” 228Cf. Terminiello v. Chicago, 337 U.S. 1, 37 (1949) (Jackson, J., dissenting) (“There is a danger that, if the Court does not temper its doctrinaire logic with a little practical wisdom, it will convert the constitutional Bill of Rights into a suicide pact.”). Some coherent, legal principle must be found permitting reasonable steps to combat terrorism without eviscerating the Assembly Clause.

It is the contention of this Article that a solution is in fact possible. Instead of narrowing the definition of the sorts of actions that constitute association, as courts have done in terrorism cases, 229See supra Part II.A. we should instead consider carefully what kinds of associations may claim the protections of the First Amendment. The beginnings of an answer can be found in the roots of the associational right. In particular, as discussed earlier, in the past thirty years the Court has entirely forgotten the Assembly Clause and the fact that the freedom of association is rooted directly in that constitutional text. 230See Inazu, supra note 47, at 62. If one does turn to the text of the Assembly Clause, however, there emerges an immediate limiting principle: the Assembly Clause does not protect all assemblies but only “the right of the people peaceably to assemble.” 231U.S. Const. amend. I (emphasis added). The right was conceived as one of citizens to gather together and consult on public issues but was never a right to riot. Furthermore, given the democracy-enhancing goals of the First Amendment, this restriction makes perfect sense because violence per se has no proper role in the democratic process (though as we shall discuss, abstract advocacy of violence may be different 232See infra text accompanying notes 255, 289–94. ). Indeed, in the brief legislative history of the Assembly Clause in the 1st Congress, there is an express mention of the “abuse” of the right of Assembly during the 1786 Shay’s Rebellion in Massachusetts (though the speaker, Elbridge Gerry, insisted that this was no reason to withhold a right of peaceable assembly). 233See The Complete Bill of Rights: The Drafts, Debates, Sources, and Origins, supra note 188, at 144 (referencing the speech of Elbridge Gerry). The Supreme Court, in one of its seminal Assembly Clause decisions, described the purpose of the Assembly Clause as being to ensure “that government may be responsive to the will of the people and that changes, if desired, may be obtained by peaceful means.” 234De Jonge v. Oregon, 299 U.S. 353, 365 (1937); see also Mark Tushnet, The Hughes Court and Radical Political Dissent: The Cases of Dirk De Jonge and Angelo Herndon, 28 Ga. St. U. L. Rev. 333, 358 (2012) (quoting De Jonge, 299 U.S. at 365). Thus, both the text and the historical understandings of the Assembly Clause limit its protections to peaceful gatherings.

If this restriction applies to assemblies, it must logically apply to associations (which indeed might simply be thought of as permanent assemblies). 235See supra note 173; see also Inazu, supra note 47, at 166–67. After all, the underlying purpose of the rights of temporary assembly and permanent association is the same—to enable citizen participation in governing. 236See supra notes 172–73 and accompanying text. But that participation must be peaceful. Even the insurrectionary right of assembling in convention that Akhil Amar defends is a right of the collective people to change their government peacefully, not to engage in armed rebellion. 237See Akhil Reed Amar, The Bill of Rights: Creation and Reconstruction 28, 30 (1998). Indeed, Alexis de Tocqueville himself (and others after him) argued that one of the benefits of protecting a right of association is that it provides a social “safety valve,” preventing disaffected citizens from going underground and resorting to violence by allowing them a means to jointly engage in the democratic process. 238See Cole, supra note 164, 231–32 (citing 1 Alexis de Tocqueville, Democracy in America 202–03 (Phillips Bradley ed., 1948)); cf. George Kateb, The Value of Association, in Freedom of Association 35, 37 (Amy Gutmann ed., 1998) (arguing that the act of associating is valuable in and of itself). Such an understanding is entirely inconsistent with protecting a right of association with groups engaged in, or explicitly planning for, violence. The textually rooted right of association is one of peaceable association and may properly be denied to groups outside that definition.

Moreover, the restriction to peaceable assemblies does not exclude only violence to persons; it also excludes violence to property. There can be no doubt that the Framers would have considered a riot aimed at private property no more protected than a riot seeking to harm individuals, just as today no reasonable person would protect a constitutional right to destroy, for example, a McDonald’s. This principle, however, has important implications for modern technology and, in particular, for the problem of cyberterrorism. In recent years, associations whose purpose and practice is to launch cyberattacks on ideologically disfavored websites have emerged. The most famous example of such an organization is the hackers group Anonymous, 239See Anonymous (Internet Group), N.Y. Times, http://topics.nytimes.com/top/reference/timestopics/organizations/a/anonymous_internet_group/ (last visited Jan. 9, 2014). though the Islamic group Izz ad-Din al-Qassam Cyber Fighters has also garnered attention recently. 240See Jeb Boone, Who Are the Izz ad-Din al-Qassam Cyber Fighters?, GlobalPost (Nov. 9, 2012), http://www.globalpost.com/dispatches/globalpost-blogs/the-grid/who-are-the-izz-ad-din-al-qassam-cyber-fighters; Tracy Kitten, DDoS Attacks on Banks Resume, Bank Info Security (Feb. 26, 2013), http://www.bankinfosecurity.com/ddos-attacks-on-banks-resume-a-5541. The obvious question that arises is whether such groups are protected by the right of peaceable assembly and association. The answer, I would stipulate, is fairly clearly “no.” Destruction of property takes many forms, and while in the Framers’ day the focus would, for obvious reasons, have been on physical destruction, destruction of websites and other digital property through cyber attacks is sufficiently analogous to physical attacks to warrant denial of the term “peaceable” to such conduct. In both cases, one is dealing with behavior that is clearly illegal and that effectively destroys an owner’s use of property. In a world where a vast and growing amount of property is in fact purely digital, treating destruction of such property as analogous to destruction of physical property is both logical and necessary. 241This is not to say that a simple act of patent or copyright infringement would suffice to eliminate a group’s constitutional protections. The notion that such intellectual property qualifies as “property” for these purposes or that simple infringement qualifies as destruction of property seems highly dubious, though the topic is beyond the scope of this Article.

Difficult boundary questions of course remain. For example, there is an argument to be made for extending the concept of violence to property to conduct, such as larceny and embezzlement, which though literally peaceful, has the practical effect of denying the owner his or her property. Indeed, it is tempting to extend the exclusion to all groups that interfere with property rights, or even engage in illegal activities, on the theory that such groups play no proper part in the democratic process. This further step, however, would be unjustified and dangerous because such expansions of the range of groups excluded from the definition of “peaceable” assemblies and associations could easily eviscerate the underlying right. The impact of such an expansion, for example, would be to deny constitutional protection to any protest group that regularly engages in civil disobedience because such disobedience is by definition illegal, and it generally interferes with property rights (consider lunch counter sit-ins). It would also permit hostile governments to single out unpopular groups for suppression based on inadvertent or exaggerated violations of minor laws. Finally, it is important to bear in mind that even if the First Amendment protects association with a group that violates the law in nonviolent ways, it does not protect the underlying illegal conduct. Associating with a group that engages in trespass may not be subject to punishment, but trespass remains so.

The insight that the Assembly Clause and association right do not protect all groups equally seems to be largely missing from current jurisprudence. To be fair, the Supreme Court in Humanitarian Law Project made a gesture in this direction. In setting out the limits of its reasoning in upholding the material-support statute, the Court stated that it did “not suggest that Congress could extend the same prohibition on material support at issue here to domestic organizations.” 242Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010). Perhaps that is the better path out of this conundrum—a rule restricting the associational right to domestic groups, which would permit prosecution for association with foreign entities, rather than my proposed restriction to peaceable groups. After all, FTOs are by definition foreign, and so such a rule would cabin the key terrorism cases without threatening domestic liberties. Moreover, there is precedent supporting the proposition that the protections of the Bill of Rights do not extend to foreigners outside the territory of the United States, 243See, e.g., United States v. Verdugo-Urquidez, 494 U.S. 259, 261 (1990) (holding that the Fourth Amendment does not apply to the search and seizure of property owned by a nonresident alien and located in a foreign country); Al Maqaleh v. Gates, 605 F.3d 84, 99 (D.C. Cir. 2010) (holding that “the jurisdiction of the courts to afford the right to habeas relief and the protection of the Suspension Clause does not extend to aliens held in Executive detention in the Bagram detention facility in the Afghan theater of war”); Kiyemba v. Obama, 555 F.3d 1022, 1026 (D.C. Cir. 2009), vacated, 559 U.S. 131 (2010) (noting that the “due process clause does not apply to aliens without property or presence in the sovereign territory of the United States”); United States v. Vilches-Navarette, 523 F.3d 1, 13 (1st Cir. 2008) (explaining that the Fourth Amendment does not apply when the United States acts against aliens in international waters). lending further support to this thesis.

Ultimately, however, the restriction the Humanitarian Law Project majority points toward is unsatisfying. To understand why, it is useful to go back to first principles and consider why the First Amendment protects associational freedom (or for that matter, speech, the press, and petitioning as well). There exists an extensive body of scholarship demonstrating the close, historical links between the freedoms of speech, the press, assembly, and petitioning, and the practice of democratic self-governance. 244See, e.g., Abu El-Haj, supra note 163, at 547, 554–55, 586–89; Bhagwat, supra note 7, at 991–93; Inazu, supra note 192, at 571–77; Mazzone, supra note 173, at 647, 729–30; see also Amar, supra note 237, at 28–30 (arguing that the Assembly Clause protects a right of the sovereign people to assemble in convention and change their government). Indeed, the significance of the associational right in the United States and its close link to our system of democracy has been recognized by observers since de Tocqueville published Democracy in America. 245See, e.g., 2 Alexis de Tocqueville, Democracy in America 513 (J.P. Mayer ed., George Lawrence trans., Anchor Books 1969) (1840). The Supreme Court has similarly been quite explicit in tying the right of association to democratic politics. 246See Buckley v. Valeo, 424 U.S. 1, 15 (1976) (quoting Kusper v. Pontikes, 414 U.S. 51, 56, 57 (1973)); Cousins v. Wigoda, 419 U.S. 477, 487 (1975). This is not necessarily to say that the only function of First Amendment liberties, including freedom of association, is to further self-governance; but there can be little doubt that this is their primary function.

At first cut, the close connection between association and self-governance would seem to support the Humanitarian Law Project Court’s distinction between domestic and foreign groups. After all, democracy is about self-governance, and in this case “self” means the sovereign people of the United States, not foreigners outside of this country. 247I leave aside the more difficult question of foreign citizens residing in the United States. For an argument that self-governance principles support extending the protections of the Bill of Rights to this group, see Ashutosh Bhagwat, The Myth of Rights: The Purposes and Limits of Constitutional Rights 264–66 (2010). FTOs such as Hamas or Al Qaeda surely have no legitimate role to play in our system of democracy. Ultimately, however, this argument goes too far. While it seems relatively clear that purely foreign organizations, whether FTOs or otherwise, have no claim to associational rights under the First Amendment, the terrorism cases do not involve that situation—they involve attempts by U.S. citizens and residents to associate with foreign groups. 248See supra Part II.A. Perhaps at the time of the Constitution’s framing, such associations would have had limited relevance to democratic politics, but in the globalized world in which we live, a world in which the United States is the dominant international power and has an almost infinite variety of commitments and ties abroad, this position is implausible. Surely, as part of their roles as active citizen–participants in our system of government, U.S. citizens must have a protected right to associate and consult with foreign groups such as Amnesty International, Médecins Sans Frontières, Greenpeace, Oxfam, and for that matter, the Communist Internationale (in its time). To restrict the associational right to purely domestic groups would be crippling to U.S. citizens’ ability to meaningfully participate in debating and influencing American policy, a result which cannot be reconciled with the purposes of the First Amendment—which is no doubt why the Supreme Court’s own cases recognize the right of a U.S. citizen to receive communications (in particular, communist propaganda communications) from abroad. 249See Lamont v. Postmaster Gen., 381 U.S. 301, 306–07 (1965); see also Aptheker v. Sec’y of State, 378 U.S. 500, 505 (1964) (striking down, on right-to-travel grounds, a statute denying passports to all members of registered Communist organizations).

The problem with the distinction between foreign and domestic groups drawn by the Humanitarian Law Project Court, the error in its analysis, is that the Court’s reasoning is rooted in neither the text nor the purposes of the First Amendment. One suspects the reason for this is that the Court, in its modern embrace of the ahistorical right of “expressive association,” has lost track of the Assembly Clause. To reiterate, however, when one does look at the text of that provision, it becomes clear that the Constitution simply does not protect any right to assemble, or associate, with organizations whose primary activities and goals are violent. It should be noted that to lose constitutional protection, it must be the group that is violent; it is not enough that some members of the group engage in violence. 250See NAACP v. Claiborne Hardware Co., 458 U.S. 886, 920 (1982) (“Civil liability may not be imposed merely because an individual belonged to a group, some members of which committed acts of violence.”). But violent groups are simply out of the First Amendment. And this remains true no matter how ideological or political the group’s motivations happen to be. Al Qaeda is not a peaceable association, any more than the Mafia, and there is therefore no right to associate with it. 251I do not consider here (because they are beyond the scope of this Article) the difficult questions of proof that arise in implementing this restriction, including in particular the extent to which entities may be designated as violent through an administrative process to which the courts then defer in subsequent legal proceedings. See United States v. Hammoud, 381 F.3d 316, 331 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005); Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 162–63 (D.C. Cir. 2003). For a description and critique of the administrative process for designating FTOs, see Sahar F. Aziz, Note, The Laws on Providing Material Support to Terrorist Organizations: The Erosion of Constitutional Rights or a Legitimate Tool for Preventing Terrorism?, 9 Tex. J. C.L. & C.R. 45, 91–92 (2003). Importantly, moreover, this reasoning applies fully to domestic associations, no less than foreign ones. This is an important point because terrorism is not, of course, a wholly foreign phenomenon (nor is it, obviously, an exclusively Islamic one). The material-support statute targets foreign terrorist organizations only, 252See 18 U.S.C. § 2339B (2012). but its companion statute, 18 U.S.C. § 2339A, is not so limited, 253See id. § 2339A. and the broader social interest in preventing terrorism encompasses wholly domestic terrorism no less than terrorism inspired from abroad. After all, aside from the September 11 attacks, the most deadly terrorist attack on U.S. soil was a purely domestic event: the bombing of the Oklahoma City federal office building on April 19, 1995, by Timothy McVeigh. 254See United States v. McVeigh, 153 F.3d 1166, 1166 (10th Cir. 1998) (upholding conviction of bomber Timothy McVeigh). Because the exclusion of non-peaceable associations from First Amendment protections does not turn on the ideological character of the group, it means that no constitutional protection is due to any predominantly violent domestic organizations, whether they be right-wing militias, radical animal-rights groups, or wholly domestic jihadi groups. Al Qaeda is merely a special case illustrating this general principle.

It is important to note that under this reasoning, what becomes unprotected is the act of associating with a violent group. The exclusion of violent groups discussed above has no application to the freedom of speech, since the Free Speech Clause does not limit its protections to “peaceable” speech. What this means is that, as the Humanitarian Law Project Court says, purely independent advocacy, even of terrorism, may not be punished unless the Brandenburg standard can be met. 255See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2733 (2010) (Breyer, J., dissenting). But association with violent groups may be punished regardless of whether it is accompanied by speech. Furthermore, one’s speech may be used as evidence of such association, just as speech, such as confessions, may be used as evidence of any other crime. Of course, speech alone, even speech supportive of an organization, cannot in itself be equated with association or the free speech right would be eviscerated. But nor does such speech immunize otherwise unprotected association. The key is whether an individual has joined or is otherwise working together, in coordination, with a violent group. If the answer is no, the individual’s speech is protected, subject to the Brandenburg test; but if the answer is yes, the individual may be prosecuted without more.

One important implication of this analysis is that because the distinction between independent and coordinated speech (or for that matter conduct) becomes central to defining the scope of constitutional protections, it is essential to clearly define that line. As described earlier, the Supreme Court in Humanitarian Law Project was notably unhelpful on this point; 256See id. at 2722 (majority opinion). nor have the lower courts provided much clarity in terrorism cases. 257See supra Part II.A. There is, however, another area of law that implicates associational rights and where the independent/coordinated line is critical, and so might provide some guidance in terrorism cases. This is in the area of campaign finance reform. Under current law, individuals (including political candidates) and corporations have an essentially unlimited First Amendment right to spend unlimited amounts of money to fund independent speech on political topics, including supporting or opposing political candidates. 258See Buckley v. Valeo, 424 U.S. 1, 39, 44–45, 47–48 (1976); Citizens United v. FEC, 558 U.S. 310, 319 (2010). In contrast, when expenditure for speech is made in coordination with a political candidate, it is treated as a contribution rather than an independent expenditure and as such is subject to greater regulation. 259See 11 C.F.R. § 109.21(b) (2012); Buckley, 424 U.S. at 46–47. Thus the distinction between independent and coordinated actions is central to the enforcement of campaign finance regulations, and courts have developed an extensive jurisprudence on the subject. 260See, e.g., Long Beach Area Chamber of Commerce v. City of Long Beach, 603 F.3d 684, 687 (9th Cir. 2010); Clifton v. FEC, 114 F.3d 1309, 1311 (1st Cir. 1997); United States v. Goland, 959 F.2d 1449, 1452 (9th Cir. 1992); Nader v. FEC, 823 F. Supp. 2d 53, 59 (D.D.C. 2011); Cao v. FEC, 688 F. Supp. 2d 498, 514–15 (E.D. La. 2010); FEC v. Christian Coal., 52 F. Supp. 2d 45, 92 (D.D.C. 1999); Colo. Educ. Ass’n v. Rutt, 184 P.3d 65, 74–75 (Colo. 2008). Moreover, in 2007, the Federal Election Commission enacted a regulation setting forth a complex and detailed multifactor test for identifying coordinated expenditures. 261See 11 C.F.R. § 109.21(a). Admittedly, the courts have not developed any crystal-clear rules regarding when communication crosses the line into coordination, and the FEC regulation similarly leaves room for interpretation; but, in combination the cases and regulations do identify the relevant factors and provide a good deal of guidance regarding when the coordination line is crossed. I do not intend to canvass the campaign finance cases in detail here as drawing the line between independent and coordinated actions is beyond the scope of this Article. The point, however, is that the line has been successfully drawn elsewhere, and there is no reason to think that courts cannot develop reasonably clear rules in the terrorism context as well. 262For a discussion of the coordination concept, arguing that the campaign finance definition does not translate well into the terrorism area, see generally Catherine A. Hardee, The Coordination Conundrum, 49 Willamette L. Rev. 189 (2012).

Another, perhaps troubling, implication of restricting associational freedoms to peaceable associations is that the right of association has, in some respects, a significantly narrower scope than the rights of free speech and the press. In fact, this distinction is quite consistent with extant law. Although since 1969 the Brandenburg test has prohibited punishment of speech advocating violence unless it can be proven that the speech is linked to imminent and likely violence, 263See Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (per curiam) (allowing states to “forbid or proscribe advocacy . . . directed to inciting or producing imminent lawless action and . . . likely to incite or produce such action”). the standard for association is much less demanding. Under the leading case addressing association and violence (the 1961 decision in Scales v. United States 264367 U.S. 203 (1961). ), membership in an organization with violent goals may be punished, consistent with the First Amendment, so long as the prosecuted individual’s membership is “active and purposive membership, purposive that is as to the organization’s criminal ends.” 265Id. at 209, 229–30. No proof of actual, imminent, or likely violence is required. 266See id. at 229. For a similar argument, see Williams, supra note 109, at 391–92. It must be acknowledged that in Brandenburg itself, the Court briefly addressed freedom of assembly at the end of its per curiam opinion and suggested in a footnote that prosecution for assembly must satisfy the same requirements as prosecution of speech. Brandenburg, 395 U.S. at 449 n.4. Brandenburg, however, has not been understood to overrule Scales, as indicated by the fact that in Humanitarian Law Project the Court approvingly cites and describes Scales without any hint that it was inconsistent with Brandenburg. See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2718 (2010).

Moreover, there is some sense to a legal regime that is more suspicious of, and less willing to protect, groups that are directed to illegal and dangerous conduct as opposed to individuals acting on their own. Groups are powerful, and groups are dangerous. It is precisely because groups are powerful, and concomitantly, that individuals can only effectively participate in democratic governance as members of groups, that we protect the right of assembly and association in the first place. 267See Abu El-Haj, supra note 163, at 554–61; Bhagwat, supra note 7, at 993, 997. But the flip side of this principle is that just as groups can be powerful, so can they also be dangerous. The occasional Unabomber notwithstanding, the reality is that individuals pose a systematically lesser threat to the public than groups, and this is especially so in the context of First Amendment liberties. A single individual calling out for violence to an undifferentiated audience rarely creates a very high risk of violence, thus making the stringent protections of the Brandenburg test socially acceptable. But a group of individuals working together and encouraging each other to take up arms is quite a different matter. Such a group is more likely to form effective plans because groups can bring together a mixture of skills. And such a group is more likely to carry those plans through because an individual acting alone will often suffer doubts, but in a group setting such doubts can be assuaged. This insight regarding the dangerousness of groups underlies the law of conspiracy and provided the basis for the Supreme Court’s decision in 1927 to affirm Anita Whitney’s prosecution for criminal syndicalism in one of the foundational cases analyzing associational rights. 268See Whitney v. California, 274 U.S. 357, 372 (1927) (“That . . . united and joint action involves even greater danger to the public peace and security than the isolated utterances and acts of individuals, is clear. We cannot hold that, as here applied, the Act is an unreasonable or arbitrary exercise of the police power of the State, unwarrantably infringing any right of free speech, assembly or association, or that those persons are protected from punishment by the due process clause who abuse such rights by joining and furthering an organization thus menacing the peace and welfare of the State.”), overruled per curiam by Brandenburg v. Ohio, 395 U.S. 444 (1969).

Once it is recognized that while the First Amendment provides broad protection to violent speech, it provides no protection to violent groups, many of the terrorism cases become much easier to reconcile with constitutional requirements. Most obviously, the many prosecutions based purely on financial contributions to an FTO, including Holy Land Foundation, Islamic American Relief Agency, Hammoud, and Afshari, become trivial. After all, even if a financial contribution qualifies as an act of association, as it surely should, there is simply no right to associate in any way with a violent group, making the act of association unprotected.

Similarly trivial are the prosecutions, such as those of the Lackawanna Six and others brought pursuant to 18 U.S.C. § 2339D, for receiving weapons training from an FTO. 269See supra notes 152–54 and accompanying text. In discussing the Lackawanna Six prosecution, I argued that one perplexing aspect of the prosecution was why the act of receiving weapons training from Al Qaeda provided any benefits to Al Qaeda as required by the material-support statute, § 2339B. 270See supra Part II.A.3. That, however, is merely a statutory argument, not a First Amendment one (though it should be noted that when the Lackawanna Six traveled to Afghanistan in 2001, § 2339D, the military training statute, had not yet been enacted). 271See supra note 152 and accompanying text. From the perspective of the First Amendment, not only is association with an FTO not protected, but in addition only peaceable forms of association and assembly are protected. A riot instigated by a previously peaceable group is no more a protected form of assembly than is a riot instigated by the Ku Klux Klan. But of course, the act of receiving military training is by no stretch of the imagination peaceable. Indeed, it would appear to be the polar opposite. The First Amendment therefore poses no barrier to prosecutions for such activities. It should be noted, moreover, that this reasoning applies fully to domestic organizations, and indeed does not appear to be limited to violent domestic organizations. If weapons training is not a form of peaceable association or assembly when engaged in with an FTO, it is no more so with a domestic militia group, or, for that matter, with the Boy Scouts. There is simply no First Amendment bar to prosecuting such behavior (whether there is a Second Amendment bar is a of course another question).

Javel Iqbal’s prosecution can also probably be justified under this understanding of associational rights. Iqbal, recall, was prosecuted because he was paid by an FTO, Hezbollah, to rebroadcast the organization’s television station, Al Manar. 272See supra Part II.A.1; see also Press Release, U.S. Attorney’s Office, S. Dist. of N.Y., supra note 75. While Iqbal’s and his associate Elahwal’s prosecutions were rooted in speech acts, it was critical to the rejection of Iqbal’s First Amendment defense that he had been paid by Hezbollah 273See Weiser, supra note 67. —that the speech was not independent. That financial relationship constitutes a form of association unprotected by the First Amendment and so permissibly subject to prosecution, even if the same speech, engaged in independently, might well have been protected under Brandenburg.

Jubair Ahmad’s prosecution follows a similar pattern. His recruiting video for the LeT would probably have constituted protected speech if produced independently, but it was not. It is true that Ahmad, unlike Iqbal, was not paid by an FTO for his work, but he did act under the direction of, and for the direct benefit of, a terrorist leader. 274See Karas, supra note 78. That seems more than enough association to justify prosecution, even if again the speech alone is not.

Finally, a focus on peaceable associations also throws useful light on the prosecution of Sami Al-Hussayen. Al-Hussayen was, of course, acquitted of the key terrorism charges against him, 275See O’Hagan, supra note 109. making the First Amendment issues somewhat moot. However, what the above analysis suggests is that if it can be demonstrated that an individual is intentionally creating mechanisms, including Internet links, that enable others to provide funding to an FTO, that should suffice to permit prosecution under the material-support statute and eliminate any First Amendment defense. Just as funding is association, so too is fund-raising—indeed, most of the funding cases, including Holy Land Foundation, Islamic American Relief Agency, Hammoud, and Afshari, could easily be recharacterized as fund-raising cases, but the fact of association would surely not be affected. And if fund-raising in the direct form of collecting and forwarding money is unprotected, so surely is the act of channeling funds without taking possession oneself (so long as done knowingly). These are all essentially equivalent actions, with similar associational implications—after all, surely no one would question that creating a website to help raise funds for a protected association would itself be conduct protected by the First Amendment. 276Cf. Watchtower Bible & Tract Soc’y of N.Y., Inc. v. Vill. of Straton, 536 U.S. 150, 160–64 (2002) (finding that an ordinance that required a permit to engage in door-to-door advocacy violated the First Amendment because it would have placed a burden on some speech). When that fund-raising is for an unprotected group, however, First Amendment protections disappear.

This is not to say, of course, that recognizing the lack of protection for violent groups solves all First Amendment problems in the context of terrorism. In particular, Tarek Mehanna’s prosecution remains problematic. He was prosecuted primarily for his speech without strong evidence that he was acting in association with an FTO. 277See Abel, supra note 56, at 739–40. As noted earlier, the First Circuit abandoned this theory in affirming Mehanna’s conviction, see United States v. Mehanna, 735 F.3d 32, 50–51 (1st Cir. 2013), and concluded (rather unconvincingly) that permitting the evidence supporting the government’s “translation-as-material-support theory” to go the jury did not prejudice Mehanna. See id. at 59–64. But it remains true that the primary theory on which Mehanna was prosecuted in the trial court was based on his independent speech, and the trial court judge permitted this theory to go to the jury, despite the lack of evidence of coordination. Id. at 47. Moreover, though the district court instructed the jury that only coordinated speech violated the material-support statute, it failed to directly define coordination, see id. at 48–49, leaving open the possibility (which the First Circuit essentially ignored) that the jury convicted based on the theory of unilateral coordination that the government advanced. But under the analysis set forth above, the essence of a prosecutable crime, and the avoidance of the stringent Brandenburg standard, is association with a violent group. Absent evidence of such association, the prosecution boils down to one directed at pure, independent speech. And that is the realm of Brandenburg. There are admittedly complications that might justify the prosecution—in particular, there were suggestions by the government that Mehanna might have been in contact with Al Qaeda leadership, which would create the requisite association. 278See Abel, supra note 56, at 734. But on the whole, the government’s prosecution of Mehanna without satisfying the Brandenburg standard for incitement poses severe First Amendment problems.

V. Dual-Function, Complex, and Evolving Associations

There is one last issue that requires attention. Under the analysis proposed here, constitutional protection (or lack thereof) turns critically on the goals and activities of a group. To implement a group-based rule, however, it is necessary to grapple with the complexities surrounding the nature and potential structures of different groups. Up to this point, the analysis in this Article has presumed that there is a clear demarcation between unprotected “violent” associations and protected “peaceable” ones. And, indeed, in many cases that line does seem quite clear. A group such as Al Qaeda, or for that matter the Mafia, is surely not a peaceable association under any definition because its primary function and operations are violent. Nor, as noted earlier, are groups such as Al Qaeda’s religious or political motivations relevant because the First Amendment limits its protections to peaceable assemblies, not peaceable, nonpolitical assemblies. 279See U.S. Const. amend. I. The problem, however, is that not all groups are as easily classified as Al Qaeda because not all groups are so single-minded or monolithic. In particular, groups can have multiple goals, they can have complex structures, and they can change over time. While a full investigation of all of the possible permutations of group structure is impossible in this space, some preliminary consideration of the implications of the complexity of groups seems in order.

A. Dual-Function Associations

First, let us consider the problem of dual-function associations. What should the constitutional status be of groups that pursue both unprotected, violent ends, and protected, nonviolent ends? The answer to this question turns out to be extremely difficult, but critical. The reason it is critical is that dual-function groups, which are characterized by a combination of violent and nonviolent goals and methods, are ubiquitous. For example, the two FTOs at issue in the Humanitarian Law Project litigation, the PKK and the Tamil Tigers, both concededly “engage[d] in political and humanitarian activities” in addition to terrorist acts. 280Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2713 (2010). Similarly, as noted earlier, the FTO Hezbollah is an established political party in Lebanon and supports a vast array of nonviolent political and social activities alongside its violence. 281See Kulish, supra note 71. And even Hamas, the Palestinian group that is widely accepted to be a terrorist group, necessarily has substantial nonviolent activities, given that since 2007 it has been in effective political control of the Gaza Strip. 282See Steven Erlanger, Hamas Seizes Broad Control in Gaza Strip, N.Y. Times, June 14, 2007, http://www.nytimes.com/2007/06/14/world/middleeast/14mideast.html?_r=0; Hamas, N.Y. Times, http://topics.nytimes.com/top/reference/timestopics/organizations/h/hamas/index.html (last visited Jan. 9, 2014). The question that arises is whether a constitutional right exists to associate with such dual-function groups, so long as the association itself does not touch upon the group’s violent activities. Or put differently, do a group’s nonviolent activities partially immunize it from the constitutional consequences of its violent activities?

The answer given by the Humanitarian Law Project Court to this question was quite clear: No. The Humanitarian Law Project majority emphasized that any material support given to an FTO, even material support directed at a dual-function FTO’s nonviolent activities, nonetheless can end up advancing violence because material support is fungible. 283 See Humanitarian Law Project, 130 S. Ct. at 2724–27. And the Court specifically rejected the view that only financial support is fungible. 284See id. at 2725. It must be acknowledged that some of the arguments advanced by the Court to defend this position, such as that material support lends “legitimacy” to FTOs, 285Id. or that material support to an FTO might strain relations between the United States and its allies, 286See id. at 2726–27. seem forced because they fail to meaningfully distinguish association with an FTO from speech supporting an FTO. Ultimately, however, there is some force to Humanitarian Law Project’s basic conclusion, even if not to its reasoning. A group that systematically engages in violence, which by definition all FTOs do, is a violent group. The fact that a group is both violent and nonviolent does not mitigate that reality. And the associational right simply does not extend to violent groups. When an individual chooses to associate with a dual-function FTO such as Hezbollah, he or she is associating with that group, not some firewalled, nonviolent element of the group. Given the famous leakiness of firewalls and the impossible complexity of ensuring that different parts of groups do not support each other (especially when the relevant group operates abroad and illegally), 287See id. at 2727–29. it seems entirely reasonable, as a constitutional matter, for the government to define the act of associating with an FTO as associating with the entire FTO, the group, as opposed to with parts of it. For the same reasons, the Humanitarian Law Project Court appears to have been justified, both as a constitutional and statutory matter, to decline to interpret the material-support statute to require proof of a defendant’s intent to further the FTO’s illegal activities because, regardless of one’s intent, the fact of an unprotected association remains. 288Id. at 2717. For violent groups that have not been publicly designated as FTOs, it probably makes sense to require that an individual have knowledge of the group’s violence to avoid ensnaring the innocent; but intent is irrelevant.

So, the First Amendment is probably best read not to protect dual-function groups. An important clarification must, however, be emphasized here. The definition of violent, unprotected assemblies and associations includes all groups that engage in, or are actively planning, violence. It does not cover groups that merely advocate violence in abstract terms. That is the learning of the Court’s foundational association decisions such as De Jonge v. Oregon 289 See 299 U.S. 353, 364–65 (1937). and Scales v. United States, 290See 367 U.S. 203, 229–30 (1961). and it also underlies Brandenburg 291See Brandenburg v. Ohio, 395 U.S. 444, 447–48 (1969) (per curiam). and that case’s predecessor, Yates v. United States. 292See 354 U.S. 298, 318 (1957) (construing the Smith Act to forbid only “advocacy directed at promoting unlawful action,” not “advocacy of abstract doctrine”). An otherwise peaceable group does not become violent simply because its members discuss the permissibility, or even advisability, of violence under certain circumstances. To hold otherwise would be to severely restrict the protections of the First Amendment, eliminating many groups outside of a narrowly defined mainstream. This was the approach taken by the Court, and by the country, during the Red Scare and McCarthy eras of the 1920s and 1950s, 293See, e.g., Dennis v. United States, 341 U.S. 494, 497–98, 509–11 (1951); Whitney v. California, 274 U.S. 357, 363–64, 371–72 (1927), overruled per curiam by Brandenburg, 395 U.S. at 449; Gitlow v. New York, 268 U.S. 652, 655–56, 664–67 (1925); Abrams v. United States, 250 U.S. 616, 617, 624 (1919). but it is surely one the nation has regretted. Moreover, given the broad tools available to the government to target groups that pose any sort of tangible threat of violence, it also seems quite unnecessary in the battle against terrorism. Under this approach, a domestic terrorist cell planning violence is completely unprotected no matter the extent of its nonviolent activities. But at the same time, a group that only discusses or even abstractly condones violence is fully protected, whether it be a gun rights organization advocating an anti–federal government reading of the Second Amendment, a group endorsing the virtues of jihad, or the Communist Party. 294One question that arises is how to treat an act of association by an individual who believes that the group she has joined is nonviolent, but which is in fact a dual-function group. To grant no protection for such behavior could be quite chilling because it would discourage association with dissident groups absent a very high degree of confidence that the group is nonviolent. On the other hand, the Humanitarian Law Project Court was certainly correct to conclude that the government may legitimately bar association by individuals who do not intend to further an FTO’s violent ends, given the fungibility of money and other tangible resources. Perhaps the solution here is a mens rea requirement focused on knowledge, but not intent—association may be punished if an individual is aware, or should have been aware, of a group’s violent nature even if the individual does not specifically intend to support violence.

This limitation on the definition of violent associations to groups that engage in or actively plan violence is, for the reasons noted above, essential if the exception is not to swallow the general rule protecting associational freedoms. It must be acknowledged, however, that this principle will not always be easy to apply. The full question of the procedures, and strength of proof, that should be required before a group can be definitively designated an FTO is, as noted earlier, extremely complex and beyond the scope of this Article. 295See supra note 251. It may well be, however, that the distinction between foreign and domestic groups relied upon by the Humanitarian Law Project Court 296See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010). might be of use here. While, as noted earlier, the complete exclusion of foreign associations from the protections of the First Amendment cannot be reconciled with the Amendment’s purpose to advance democratic self-governance, there can be no doubt that associations composed exclusively or primarily of citizens are more relevant to, and more directly connected with, the democratic process than primarily foreign organizations. Indeed, the Supreme Court’s own case law recognizes this point by upholding Congress’s plenary power to exclude aliens, even in the face of an admittedly colorable First Amendment claim, 297See Kleindienst v. Mandel, 408 U.S. 753, 764–70 (1972). and by upholding the denial of an individual’s request for a passport to travel to Cuba, despite the impact of the denial on the individual’s ability to gather information about Cuba. 298See Zemel v. Rusk, 381 U.S. 1, 16–17 (1965). More broadly, the very existence of government policies, such as trade embargoes, and the Humanitarian Law Project Court’s undoubtedly correct recognition of the traditional deference given to the political branches on issues touching on foreign policy, 299See Humanitarian Law Project, 130 S. Ct. at 2727. clearly indicate that the government’s practical power to constitutionally restrict association with foreigners, while not unlimited, is greater than its purely domestic authority.

There are thus good reasons to treat association with foreign groups as constitutionally protected but not quite as strongly protected as purely domestic associations. At the same time, it seems equally obvious that as the orientation of a group turns more violent, its relevance to democratic self-governance becomes more marginal. It may be, therefore, that with respect to foreign groups, a somewhat lower quantum of evidence may be required to strip a group of protection on the grounds that it is violent than with a domestic group. Or, put perhaps more positively, it may well be that the Constitution is best read to require strong, positive proof that a domestic group has engaged in, or is actively planning to engage in, violence before a court will place it outside the bounds of the freedom of association, even if a similarly situated foreign group may be denied protection more easily. Of course, even under such an approach there will be intermediate instances of groups that possess a mixed domestic and foreign membership, but those will be relatively rare (especially because we are concerned here only with potentially violent groups) and can probably be dealt with on a case-by-case basis.

B. Complex and Evolving Associations

Having multiple goals, some violent and some nonviolent, is not the only complication one runs into in considering how to delineate the scope of protected and unprotected groups under the First Amendment. Another set of problems arises when one considers the possibilities of complex organizational structures and of change within groups. Regarding structure, the obvious problem is that not all groups are unitary. Instead, many groups have more complicated structures. For example, some groups might be made up of local chapters as well as a national organization (examples include the Boy Scouts and the Sierra Club), or might be a loose federation of affiliated groups (examples include the Ku Klux Klan 300See About the Ku Klux Klan, Anti-Defamation League, http://archive.adl.org/learn/ext_us/kkk/default.asp (last visited Jan. 9, 2014). and the Tea Party movement). The obvious question raised by such entities is whether they should be considered a single association or assembly for First Amendment purposes, or as multiple groups. The answer has some significance because if the organization is treated as a single group, then systematic violence by any significant number of local chapters would taint the entire organization, denying it constitutional protection. Treating such organizations as multiple groups, however, would permit some local groups to retain protection by distancing themselves from the violence of their compatriots.

Despite the importance of this question, however, it seems clear that a single, one-size-fits-all answer to the dilemma is impossible because of the enormous variation among groups. For many groups, moreover, the answer does not matter. Obviously, groups like the Boy Scouts and the Sierra Club will remain fully protected regardless of whether they are one or multiple groups because none of their components has violence as a significant goal. Similarly, it does not matter whether the various Al Qaeda affiliates, such as Al Qaeda in the Islamic Mahgred (AQIM), Al Qaeda in Iraq (AQI), and Al Qaeda in the Arabian Peninsula (AQAP), 301Bureau of Counterterrorism, supra note 72. are treated as separate groups or as part of an overarching Al Qaeda organization because all of the component groups as well as the parent group are dedicated to violence and so are unprotected. Clearly, however, there will be instances where a national network is largely peaceful but some local affiliates of that group commit themselves to violence. How should such groups be analyzed?

The best that can be said here is that the answer must depend on the amount of integration, coordination, and central control the national (or international) network demonstrates. Organizations that demonstrate coherent, centrally formulated policy positions, and a commitment to discipline local chapters that deviate from them, clearly qualify as unitary groups. As the level of centralized control or cohesion declines, it becomes more likely that local affiliates can claim independent associational status. An important factor in all of this may well be the understandings of the group members—to what extent does their behavior evince a feeling of membership in, and commitment to, a larger organization as opposed to a smaller, more intimate subgroup. Finally, it may well be reasonable to insist, if a larger organization or a substantial number of its affiliates have demonstrated a commitment to violence, that any local group that wishes to retain protection must explicitly repudiate and break with those elements. These are, of course, general principles only, and their application to specific groups or organizations will depend on the nature of that group or groups. It is also clear that some judgment, and some close cases, will be inevitable. But that is of course true of the scope of all constitutional rights.

A related problem to the phenomenon of network groups is the problem of affiliation. Consider in this regard Sinn Fein, the Irish political party that allegedly shared close links with the Provisional Irish Republican Army (Provisional IRA). 302See Gerry Adams and Sinn Fein, PBS, http://www.pbs.org/wgbh/pages/frontline/shows/ira/conflict/gasf.html (last visited Jan. 9, 2014). Until it repudiated violence following the Good Friday Accords, 303See IRA Statement, Guardian (July 28, 2005), http://www.guardian.co.uk/politics/2005/jul/28/northernireland.devolution. the Provisional IRA surely constituted an unprotected, violent association because it had engaged in terrorism. But what of Sinn Fein (assuming that links between Sinn Fein and the Provisional IRA could be proven, a point that Sinn Fein, it should be noted, has always denied albeit not especially convincingly)? Again, no simple answer could respond to infinite factual variations, but at a minimum, it seems clear that if a purportedly peaceful organization such as Sinn Fein has substantial leadership overlap with a violent group and coordinates with such a group in establishing policy and strategy, it forfeits its right to constitutional protection. Such a taint by association may seem unfair, but it is the only way to prevent violent, unprotected groups from creating sanitized affiliates, which in turn can strengthen the violent group by engaging in such activities as fund-raising and recruitment. Such a possibility would simply blow too large a hole in the state’s legitimate efforts to control and suppress violence and violent groups.

In addition to organizational complexity, the Provisional IRA (and perhaps Sinn Fein) also nicely illustrates another problem with group definition—the possibility of evolution and change. As noted above, during the Troubles, there can be no doubt that the Provisional IRA was a violent, unprotected association, membership in which could clearly be punished consistent with the First Amendment. It is also true, however, that in the wake of the Good Friday accords, the leadership of the Provisional IRA has permanently renounced violence 304See id. and also appears to have decommissioned its weapons. 305IRA ‘Has Destroyed All Its Arms,BBC News (Sept. 26, 2005), http://news.bbc.co.uk/2/hi/uk_news/northern_ireland/4283444.stm. Can a formerly violent group repudiate violence and regain constitutional protections? This turns out to be a profoundly difficult question.

The problem lies in balancing a group’s (and its members’) interest in abandoning violence and returning to the political fold against the danger that a group’s purported conversion is not genuine, but rather an opportunistic attempt to regroup and perhaps rearm. Given the very strong governmental interest in preventing a violent group from reinvigorating itself during a phony abandonment of violence, it seems reasonable for courts to insist that before a group is considered to have moved beyond violence, there must be a full and unequivocal renunciation of violence by the group’s leadership and by most of its membership, expulsion of members who continue to plan violence, good evidence that the group is not maintaining weapons or other tools of violence, and most importantly, the passage of time to confirm the genuineness of the renunciation.

On this view, a group such as the Provisional IRA may fairly be said, at this point in time, to have reformed itself and so recovered constitutional protections even though this could not have been said immediately after either the Good Friday Accords or even the Provisional IRA leadership’s renunciation of armed struggle in 2005. 306See IRA Statement, supra note 303. Similarly, the signing of the Oslo Accords by the Palestinian Liberation Organization (PLO) and its subsequent recognition of the State of Israel 307Letter from Yasser Arafat, Chairman, The Palestine Liberation Org., to Yitzhak Rabin, Prime Minister of Isr. (Sept. 9, 1993), http://www.incore.ulst.ac.uk/services/cds/agreements/pdf/is4.pdf. probably was not sufficient in itself to cleanse the group, but the passage of time and the group’s continuing renunciation of violence might at this point suffice. 308This is not to say, of course, that individual members of these organizations are immune from punishment for past crimes.

A final, difficult example of an evolving group is the Ku Klux Klan. The question of constitutional protections for the Klan first came to the Supreme Court in 1928 in the New York ex rel. Bryant v. Zimmerman case. 309See 278 U.S. 63 (1928). Thanks to John Inazu for pointing me to this issue, which he discusses in more detail in Inazu, supra note 47, at 50–51. At issue in Bryant was the constitutionality of a New York statute that required membership organizations having twenty or more members, and requiring an oath as a condition of membership, to disclose their members and officers to the New York secretary of state, and criminalized knowing membership in an organization that had not complied with this requirement. 310Bryant, 278 U.S. at 65–66. Bryant was charged and convicted of being a member of the Klan, knowing that it had not complied with the statute. 311Id. at 71. He challenged the constitutionality of the statute under the Fourteenth Amendment 312Id. at 65. (the First Amendment was not explicitly evoked, presumably because the incorporation doctrine was quite undeveloped in the 1920s). The Court, by an 8–1 vote, rejected this argument, with both Justices Holmes and Brandeis joining the majority 313See id. at 73, 77. despite their many influential dissents in other free speech and assembly cases during this era. Critical to the Court’s reasoning was the fact that the Klan was well-known to be a violent organization that acted against racial and religious minorities, and so was distinguishable from organizations such as labor unions, which were exempt from the statute. 314See id. at 75–77. Indeed, the Klan’s violence was precisely the grounds upon which the Court distinguished Bryant in NAACP v. Alabama ex rel. Patterson, 315357 U.S. 449, 465 (1958). the pathbreaking freedom of association decision recognizing the NAACP’s right to maintain the secrecy of its membership lists. 316Id. at 462–63.Bryant and Patterson thus stand together as a strong, judicial endorsement of the principle that violent and nonviolent organizations are distinguishable for constitutional purposes, with the former entitled to fewer constitutional protections. They also appear to establish that the Klan is such a violent organization.

But then the plot thickens. Forty-one years after Bryant, in 1969, the Klan was back before the Court in the famous Brandenburg decision. 317Brandenburg v. Ohio, 395 U.S. 444, 444–45 (1969). This time, however, the Court reversed the prosecution of a Klan leader on First Amendment grounds, invoking the Assembly Clause as well as the Free Speech Clause. 318Id. at 449 & n.4. What had changed? The Court’s laconic per curiam opinion in Brandenburg does not say (indeed, it does not even reference Bryant). As John Inazu points out, presumably one explanation might well be the Klan’s substantially reduced size, influence, and responsibility for violence by 1969. 319Inazu, supra note 47, at 51. In fact, however, this explanation is troubling. Klan groups were, after all, responsible for serious violence against civil rights protestors during the 1950s and 1960s, not all that long before Brandenburg. Furthermore, unlike the Provisional IRA, the Klan had not, to my knowledge, repudiated violence as of 1969. As such, it seems likely that if the State of Ohio had wanted to simply proscribe the Klan in 1969 as a terrorist association, it could have done so consistent with the First Amendment. This is not to say that the result in Brandenburg was wrong. Insofar as Brandenburg’s conviction was based on his speech, as opposed to his membership in the Klan, the First Amendment clearly protected him. Furthermore, the Criminal Syndicalism statute that Brandenburg was convicted under, which proscribed assembling with a group that simply advocated violence, 320See Brandenburg, 395 U.S. at 444–45. was clearly overbroad. So Brandenburg’s conviction certainly violated the First Amendment. But in 1969, the Ku Klux Klan was not entitled to constitutional rights of assembly or association.

Of course, over half a century has elapsed since Brandenburg, and more since the Civil Rights Era. What of the Klan today? That is a most difficult question. On the one hand, the significance of the Klan and its connection with actual acts of violence has ebbed greatly over the years as the ideas with which the Klan is associated have become marginalized. This suggests that the Klan has “evolved” sufficiently to no longer qualify as a group that engages in violence (whatever it preaches). On the other hand, the Klan has historically been associated with grotesque violence and terrorism. 321See Virginia v. Black, 538 U.S. 343, 352–57 (2003). To overcome that history, it would seem that a clear repudiation of violence would be necessary. This is complicated by the fact that the modern Klan is not a single organization; rather, numerous—sometimes competing—groups claim the title and mantle. Given this fact, the best answer would seem to be that a current group claiming to be the Ku Klux Klan is entitled to constitutional protection only if it has no recent history of violence and has clearly repudiated the Klan’s past violence.

In short, groups are complex and changing, and so any constitutional doctrine granting protection, or denial of protection, for groups must wrestle with difficult boundary questions. This problem, however, is not unique to violent groups; it is raised by any legal rule that treats groups, qua groups, as special. And ultimately, while the boundary questions are difficult, they are not, as this discussion has hopefully demonstrated, unsolvable.

Conclusion

The freedom of assembly and association protected by the First Amendment is a fundamental constitutional liberty that plays a critical role in our system of democratic self-governance; it therefore cannot be allowed to weaken or narrow. Terrorism poses the most significant security threat the United States has faced since the end of the Cold War and perhaps the most significant domestic threat since World War II; the government must therefore possess the tools to effectively combat that threat. The questions this Article explores are whether the courts have successfully reconciled those two statements, and more broadly, whether they can be reconciled. As to the first question, it is quite clear that the courts have not identified any legal principles regarding the role of the First Amendment in the domestic War on Terror and, indeed, have barely made an effort in that direction. This Article has demonstrated, however, that the answer to the second question is at least a qualified yes. There does exist an understanding of the right of assembly and association, rooted in the text and history of the First Amendment, which retains a robust set of constitutional protections within their proper scope, while permitting prosecution of dangerous groups.

Behind these debates, however, looms a much broader question. The truth of the matter is that in the thirty years following the Supreme Court’s pathbreaking reformulation of free speech law in its 1969 Brandenburg decision, First Amendment liberties were vigorously protected without much controversy because that was an era of relative geopolitical stability. In the wake of the September 11 attacks, however, the United States finds itself once again facing serious foreign and domestic threats. In this atmosphere, do either the courts or the nation as a whole have the will to continue to vigorously enforce the First Amendment in the face of perceived violence or subversion? The argument presented here supports the view that we absolutely can and should retain a strong First Amendment, at least in part, because the First Amendment, properly understood, does not force us into absurd choices. What legal analysis cannot answer, however, is whether we have the will to do so.

Footnotes

Professor of Law and Martin Luther King Jr. Hall Research Scholar, University of California, Davis. B.A., 1986, Yale University; J.D., 1990, The University of Chicago. Thanks to Ben Strauss and Elise Weinberg for excellent research assistance. Thanks also to Deven Desai, Chris Elmendorf, Carlton Larson, and participants at the Freedom of Expression Conference at the Yale Law School for extremely helpful comments, including especially to John Inazu for serving as discussant for this Article and providing invaluable feedback.

1See, e.g., 42 U.S.C. § 2000dd (2006) (prohibiting “cruel, inhuman, or degrading treatment or punishment” by executive officials of individuals in custody); Hamdan v. Rumsfeld, 548 U.S. 557, 568–69 (2006) (discussing the President’s issuance of a comprehensive military order governing the use of military commissions to try noncitizens in the War on Terror); Hamdi v. Rumsfeld, 542 U.S. 507, 516 (2004) (discussing whether the President has the authority to detain “enemy combatants” (internal quotation marks omitted)); Statement on Signing the Department of Defense, Emergency Supplemental Appropriations to Address Hurricanes in the Gulf of Mexico, and Pandemic Influenza Act, 2006, 2 Pub. Papers 1901 (Dec. 30, 2005) (asserting that “situations may arise, especially in wartime, in which the President must act promptly under his constitutional grants of executive power”).

2See Boumediene v. Bush, 553 U.S. 723, 732 (2008) (holding that the constitutional privilege of habeas corpus reaches enemy combatants detained at Guantanamo Bay, Cuba).

3The primary counterexample is the controversy over whether the Bush Administration’s domestic warrantless wiretapping program violated the Fourth Amendment. See Evan Tsen Lee, The Legality of the NSA Wiretapping Program, 12 Tex. J. C.L. & C.R. 1, 40–41 (2006).

4See infra Part II.

5130 S. Ct. 2705 (2010).

6See id. at 2711; see also 18 U.S.C. § 2339B(a)(1) (2006).

7For a more general discussion of the relationship between the speech and association rights, see Ashutosh Bhagwat, Associational Speech, 120 Yale L.J. 978 (2011).

8U.S. Const. amend. I.

9See 130 S. Ct. at 2711.

10See id. at 2712 & n.1. The material-support statute was originally enacted as part of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), Pub. L. No. 104-132, sec. 303(a), § 2339B(a)(1), 110 Stat. 1214, 1250. It thus predates the September 11 attacks, though the number of prosecutions under the statute appear to have increased dramatically following the attacks. See Andrew Peterson, Addressing Tomorrow’s Terrorists, 2 J. Nat’l Security L. & Pol’y 297, 301 (2008).

11Humanitarian Law Project, 130 S. Ct. at 2713 (quoting 18 U.S.C. § 2339A(b)(1) (2006)) (internal quotation marks omitted).

1218 U.S.C. § 2339B(g)(6) (2012).

13See 8 U.S.C. § 1189(a)(1) (2012).

14Humanitarian Law Project, 130 S. Ct. at 2713–14.

15Id. at 2714.

16See id. at 2712, 2719, 2730–31.

17See id. at 2717–18.

18See id.

19Id. at 2719.

20Id. at 2719–22 (internal quotation marks omitted).

21See id. at 2722.

22Id. at 2722–23.

23See id. at 2724.

24See id. (explaining that “the Government’s interest in combating terrorism is an urgent objective of the highest order”).

25See id. (quoting Opening Brief for Humanitarian Law Project at 51, Humanitarian Law Project, 130 S. Ct. 2705 (Nos. 08-148 & 09–89)) (internal quotation marks omitted).

26See id. at 2728–29 (“Given the sensitive interests in national security and foreign affairs at stake . . . it was necessary to prohibit providing material support in the form of training, expert advice, personnel and services to foreign terrorist groups . . . .”).

27Id. at 2725.

28Id. at 2726.

29Id. at 2727.

30Id. at 2730.

31The Court spent a mere three paragraphs discussing the freedom of association claim. See id. at 2730–31.

32Id. at 2730.

33See id. at 2731 (Breyer, J., dissenting).

34Id. at 2732.

35Id. at 2732–33.

36Id. (citing NAACP v. Claiborne Hardware Co., 458 U.S. 886, 911 (1982); De Jonge v. Oregon, 299 U.S. 353, 364 (1937)).

37Id. at 2733 (citing Claiborne, 458 U.S. at 908; Scales v. United States, 367 U.S. 203, 229 (1961)).

38See id. at 2735–37 (internal quotation marks omitted).

39Id. at 2739–40.

40See, e.g., Johnson v. California, 543 U.S. 499, 512–13 (2005) (noting that the Court has “refused to defer to state officials’ judgments on race in other areas where those officials traditionally exercised substantial discretion”); Landmark Commc’ns, Inc. v. Virginia, 435 U.S. 829, 844 (1978) (explaining that courts must independently review legislative actions that impinge on the First Amendment). But see Grutter v. Bollinger, 539 U.S. 306, 328 (2003) (deferring to a law school’s educational judgment about the value of diversity).

41323 U.S. 214, 217–18 (1944) (determining it was within “the war power of Congress and the Executive to exclude those of Japanese ancestry from the West Coast war area”).

42341 U.S. 494, 516–17 (1951) (upholding prosecution of Communist Party leadership as consistent with the First Amendment).

43See Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (explaining that the First Amendment does “not permit a State to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action”).

44United States v. Stevens, 130 S. Ct. 1577, 1586 (2010) (quoting New York v. Ferber, 458 U.S. 747, 761–62 (1982)) (internal quotation mark omitted).

45See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010).

46See id. at 2732 (Breyer, J., dissenting).

47See generally John D. Inazu, Liberty’s Refuge: The Forgotten Freedom of Assembly 1–2 (2012) (“[The] freedom of assembly has become little more than a historical footnote in American law and political theory.”); Bhagwat, supra note 7, at 980 (explaining that the freedom of assembly and right to petition the government “have traditionally been the poor stepchildren of First American law, neglected and ignored”).

48See infra Part II.

49See infra text accompanying notes 217–24.

50See Humanitarian Law Project, 130 S. Ct. at 2722.

51Hasan is the convicted gunman in the November 2009 Fort Hood shootings. See Nidal Malik Hasan, N.Y. Times, (last visited Jan. 8, 2014); see also Manny Fernandez, Defendant in Fort Hood Shooting Case Admits Being Gunman, N.Y. Times, Aug. 7, 2013, at A14.

52Shahzad was convicted of attempting to detonate a car bomb in Times Square in May of 2010. See Faisal Shahzad, N.Y. Times, http://topics.nytimes.com/top/reference/timestopics/people/s/faisal_shahzad/index.html (last visited Jan. 8, 2014); see also Declan Walsh, 4 Are Acquitted by Pakistan in Times Square Bomb Plot, N.Y. Times, June 3, 2012, at A13.

5318 U.S.C. § 2339A (2012).

54See Innokenty Pyetranker, Sharing Translations or Supporting Terror? An Analysis of Tarek Mehanna in the Aftermath of Holder v. Humanitarian Law Project, Nat’l Security L. Brief, Spring 2012, at 21, 21; Abby Goodnough & Liz Robbins, Massachusetts Man Is Held on Charges of Plotting Attacks, Including One at a Mall, N.Y. Times, Oct. 22, 2009, at A21.

55See Pyetranker, supra note 54, at 38; Abby Goodnough, U.S. Citizen Is Convicted in Plot to Support Al Qaeda, N.Y. Times, Dec. 21, 2011, at A26.

56See Pyetranker, supra note 54, at 36 (citing Second Superseding Indictment at 1, 8, 15–16, United States v. Mehanna, No. 09-CR-10017-GAO (D. Mass. 2011), 2010 WL 2516469); see also Nikolas Abel, Note, United States vs. Mehanna, the First Amendment, and Material Support in the War on Terror, 54 B.C. L. Rev. 711, 732 (2013) (citing Government’s Proffer and Memorandum in Support of Detention at 64–66, Mehanna, No. 09-CR-10017-GAO (D. Mass 2011)).

57See Abel, supra note 56, at 731 & n.169 (citing Second Superseding Indictment, supra note 56, at 5).

58See id. at 732 (citing Second Superseding Indictment, supra note 56, at 8). During the late stages of the editorial process for this Article, the First Circuit issued its opinion affirming Mehanna’s conviction in full. See United States v. Mehanna, 735 F.3d 32, 41 (1st Cir. 2013). Interestingly, the appellate court’s decision to affirm Mehanna’s conviction on the terrorism counts relied entirely on his 2004 trip to Yemen. See id. at 46. Indeed ,the court came close to conceding that the government’s evidence was insufficient to demonstrate that Mehanna’s translation and speech activities were coordinated with Al Qaeda. See id. at 50–51. The court concluded that the mass of evidence that had been admitted at trial regarding the government’s “translation-as-material-support theory” was irrelevant because the jury was properly instructed, and the factual evidence regarding Mehanna’s Yemen trip was sufficient to support his conviction. Id. at 51. Whatever the merits of this analysis, it remains true that at trial, before the convicting jury, the government heavily emphasized Mehanna’s speech.

59See id. at 739–40.

60Id.

61Id. at 732 & n.175 (quoting Government’s Proffer and Memorandum in Support of Detention, supra note 56, at 11) (internal quotation marks omitted).

62See Pyetranker, supra note 54, at 21–22.

63Id. at 36–37 (citing Defendant’s Memorandum of Law in Support of His Motion to Dismiss Counts One Through Three Based on Vagueness and Overbreadth, United States v. Mehanna, No. 09-CR-10017-GAO (D. Mass. 2011), 2011 WL 3740563).

64See id. at 37 (citing Defendant’s Memorandum of Law in Support of His Motion to Dismiss, supra note 63, at 3, 5, 16–18); Abel, supra note 56 at 732–33 (citing Defendant’s Memorandum of Law in Support of His Motion to Dismiss, supra note 63, at 11–12).

65See Abel, supra note 56 at 732–33 (citing Government’s Opposition to Defendant’s Motion to Dismiss Counts One Through Three of the Second Superseding Indictment Based on Overbreadth and Vagueness at 21–23, Mehanna, No. 09-CR-10017-GAO (D. Mass. 2011), 2011 WL 3959520).

66Pyetranker, supra note 54, at 38.

67Benjamin Weiser, A Guilty Plea in Providing Satellite TV for Hezbollah, N.Y. Times, Dec. 24, 2008, at A21.

68See id.

69See id.

70See Timothy Williams & William K. Rashbaum, New York Man Charged with Enabling Hezbollah Television Broadcasts, N.Y. Times, Aug. 25, 2006, .

71See Nicholas Kulish, Despite U.S. Fear, Hezbollah Moves Openly in Europe, N.Y. Times, Aug. 16, 2012, at A1.

72Bureau of Counterterrorism, Foreign Terrorist Organizations, U.S. Dep’t St. (Sept. 28, 2012), http://www.state.gov/j/ct/rls/other/des/123085.htm.

73See Weiser, supra note 67.

74Id.

75Press Release, U.S. Attorney’s Office, S. Dist. of N.Y., New Jersey Man Sentenced to 17 Months in Prison for Providing Material Support and Resources to Hizballah (June 23, 2009), http://www.fbi.gov/newyork/press-releases/2009/nyfo062309.htm.

76See id.

77Id.

78See Rachel Karas, Woodbridge Man Sentenced to 12 Years for Terrorist Propaganda Video, Post Local (Apr. 13, 2012, 6:40 PM ), http://www.washingtonpost.com/blogs/crime-scene/post/woodbridge-man-sentenced-to-12-years-for-terrorist-propaganda-video/2012/04/13/gIQACI7tFT_blog.html.

79Press Release, U.S. Attorney’s Office, E. Dist. of Va., Pakistani National Living in Woodbridge Pleads Guilty to Providing Material Support to Terrorist Organization (Dec. 2, 2011), http://www.justice.gov/usao/vae/news/2011/12/20111202ahmadnr.html; see Karas, supra note 78.

80Bureau of Counterterrorism, supra note 72.

81See Karas, supra note 78.

82Id.

83Id.

84United States v. Hammoud, 381 F.3d 316, 325 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005).

85See id.

86Id. at 326. Hammoud’s associates were also found to have provided “dual use” physical equipment to Hezbollah, but Hammoud declined to participate in these activities. Id.

87Id.

88See id. at 328.

89Id. at 329.

90Id.

91Id. (citing United States v. O’Brien, 391 U.S. 367, 377 (1968)). In O’Brien, the Supreme Court upheld the conviction of David Paul O’Brien for burning his draft card during an antiwar rally, on the grounds that the statute prohibiting the burning of draft cards was not directed at speech, but was rather a neutral conduct regulation whose purpose was “unrelated to the suppression of free expression.” O’Brien, 391 U.S. at 369, 377.

92United States v. Afshari, 426 F.3d 1150, 1152 (9th Cir. 2005) (internal quotation marks omitted).

93Id.

94Id. at 1152–53; Scott Shane, Iranian Dissidents Convince U.S. to Drop Terror Label, N.Y. Times, Sept. 21, 2012, http://www.nytimes.com/2012/09/22/world/middleeast/iranian-opposition-group-mek-wins-removal-from-us-terrorist-list.html?pagewanted=all&_r=0.

95Afshari, 426 F.3d at 1152.

96See United States v. Rahmani, 209 F. Supp.2d 1045, 1059 (C.D. Cal. 2002), rev’d sub nom. United States v. Ashfari 426 F.3d 1150 (9th Cir. 2005).

97See Afshari, 426 F.3d at 1153.

98See id. at 1160.

99Id. (footnote omitted) (citing McConnell v. FEC, 540 U.S. 93 (2003); Buckley v. Valeo, 424 U.S. 1 (1976)).

100Id.

101Id. at 1161 (quoting Humanitarian Law Project v. Reno, 205 F.3d 1130, 1133 (2000)) (internal quotation mark omitted).

102For a detailed discussion of the Lackawanna Six prosecution, see Matthew Purdy & Lowell Bergman, Unclear Danger: Inside the Lackawanna Terror Case, N.Y. Times, Oct. 12, 2003, at N1, available at http://www.nytimes.com/2003/10/12/nyregion/12LACK.html?pagewanted=1.

103See id.

104See id.

105See id.

106See id.

107See id.

108See id.

109See Maureen O’Hagan, A Terrorism Case That Went Awry, Seattle Times, Nov. 22, 2004, http://seattletimes.com/html/localnews/2002097570_sami22m.html. For a detailed description of the Al-Hussayen prosecution, see Alan F. Williams, Prosecuting Website Development Under the Material Support to Terrorism Statutes: Time to Fix What’s Broken, 11 N.Y.U. J. Legis. & Pub. Pol’y 365, 367–73 (2008).

110See O’Hagan, supra note 109.

111Compare Williams, supra note 109, at 368–71 (reciting FBI allegations that Al-Hussayen was involved in website development recruiting fighters for jihad), with O’Hagan, supra note 109 (raising doubts about Al-Hussayen’s involvement).

112See O’Hagan, supra note 109.

113See id.

114Williams, supra, note 109 at 373.

115See Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 160 (D.C. Cir. 2003). The Foundation was originally named the “Occupied Land Fund” and changed its name to Holy Land Foundation in 1991. Id.

116Id. (internal quotation marks omitted).

117See id.; Exec. Order No. 13,224, 3 C.F.R. 786 (2002).

11850 U.S.C. § 1701 (2006); see 3 C.F.R. 786.

119Holy Land Found., 333 F.3d at 161.

120See Clyde Haberman, Israeli Tensions Rise as an Officer Is Found Slain, N.Y. Times, Dec. 16, 1992, at A1.

121Holy Land Found., 333 F.3d at 159.

122Bureau of Counterterrorism, supra note 72.

123Holy Land Found., 333 F.3d at 164 (citing FEC v. Colo. Republican Fed. Campaign Comm., 533 U.S. 431 (2001)).

124Id. at 164–65 (quoting Holy Land Found. for Relief & Dev. v. Ashcroft, 219 F. Supp. 2d 57, 81 (D.D.C. 2002), aff’d, 333 F.3d 156 (D.C. Cir. 2003)) (internal quotation marks omitted).

125See id. at 164.

126Id. at 161 (quoting Holy Land Found., 219 F. Supp. 2d at 81) (internal quotation marks omitted).

127Id. at 166 (citing Humanitarian Law Project v. Reno, 205 F.3d 1130, 1133 (9th Cir. 2000)).

128Islamic Am. Relief Agency v. Gonzales, 477 F.3d 728, 730–31 (D.C. Cir. 2007).

129See id. at 731.

130 See id.

131See id.

132Id. at 736–37 (citing Holy Land Found., 333 F.3d at 166; Humanitarian Law Project, 205 F.3d at 1133).

133Id. at 737.

134See U.S. Dep’t of Justice, Nat’l Sec. Div., National Security Division Statistics on Unsealed International Terrorism and Terrorism-Related Convictions (June 6, 2012) (unpublished data chart), http://www.fas.org/irp/agency/doj/doj060612-stats.pdf (reprinting Department of Justice statistics listing 494 terrorism-related convictions obtained by the Department between September 2001 and December 2011, the most recent being the conviction of Tarek Mehanna).

135See infra Part III.

136See supra Part I.

137Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010).

138See id. at 2722–23.

139See supra text accompanying notes 54–66.

140See supra text accompanying notes 109–14.

141For a recent judicial discussion of these difficulties, see Hedges v. Obama, 890 F. Supp. 2d 424, 466 (S.D.N.Y. 2012), vacated, 724 F.3d 170 (2d Cir. 2013). For a similar argument that the Court’s reasoning leaves little substance to the supposedly preserved right of association, see Margaret Tarkington, Freedom of Attorney–Client Association, 2012 Utah L. Rev. 1071, 1082.

142See supra text accompanying notes 67–77.

143See Daphne Barak-Erez & David Scharia, Freedom of Speech, Support for Terrorism, and the Challenge of Global Constitutional Law, 2 Harv. Nat’l Security J. 1, 3 (2011).

144See, e.g., Citizens United v. FEC, 558 U.S. 310, 319 (2010) (holding that the Government may not prohibit independent expenditures by corporations used to fund political speech).

145See, e.g., N.Y. Times Co. v. Sullivan, 376 U.S. 254, 266 (1964) (“[I]f the allegedly libelous statements would otherwise be constitutionally protected . . . , they do not forfeit that protection because they were published in the form of a paid advertisement.”).

146See supra Part II.A.

147See supra Part II.A.

148See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010).

149See supra note 90 and accompanying text.

150See infra Part III.

151See supra text accompanying notes 102–08.

152There is a separate statute that does prohibit obtaining military training from a terrorist group, but it was not enacted until 2004 and so could not be invoked in the Lackawanna Six prosecution. See 18 U.S.C. § 2339D (2012). For a complex set of reasons, although this statute was passed in 2004, as of 2009, only a single prosecution had been brought under it. See Robert M. Chesney, Terrorism, Criminal Prosecution, and the Preventive Detention Debate, 50 S. Tex. L. Rev. 669, 688–89 & n.53 (2009) (citing Criminal Complaint at 1, United States v. Maldonado, No. 07-cr-124 (S.D. Tex. Feb. 13, 2007)). However, there are cases currently pending, which charge violations of § 2339D. See, e.g., United States v. Medunjanin, No. 10 CR 019, 2012 WL 1514766, at *6 (E.D.N.Y. May 1, 2012); United States v. Ahmed, No. 10 Cr. 131 (PKC), 2012 WL 983545, at *1 (S.D.N.Y. Mar. 22, 2012).

15318 U.S.C. § 2339B(a)(1) (2012).

154See supra note 107 and accompanying text.

155U.S. Const. amend. I.

156United States v. Stevens, 130 S. Ct. 1577, 1584 (2010). But see James Weinstein, Participatory Democracy as the Central Value of American Free Speech Doctrine, 97 Va. L. Rev. 491, 491–92 (2011) (raising questions about the accuracy of this view).

157See e.g., Knox v. Serv. Emps. Int’l Union, Local 1000, 132 S. Ct. 2277, 2288 (2012) (“Our cases have often noted the close connection between our Nation’s commitment to self-government and the rights protected by the First Amendment.”); Snyder v. Phelps, 131 S. Ct. 1207, 1215 (2011) (explaining that speech on public issues has the most protection under the First Amendment); N.Y. Times Co. v. Sullivan, 376 U.S. 254, 269–70 (1964) (considering the case before it “against the background of a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open”).

158See Snyder, 131 S. Ct. at 1216–17 (quoting Rankin v. McPherson, 483 U.S. 378, 387 (1987)); see also Whitney v. California, 274 U.S. 357, 377 (1927) (Brandeis & Holmes, JJ., concurring), overruled per curiam by Brandenburg v. Ohio, 395 U.S. 444 (1969).

159For a brief description of the evolution of the incitement doctrine, see Bhagwat, supra note 7, at 1003–05.

160395 U.S. at 447.

161For a detailed discussion of the undermining of the Brandenburg test in the War on Terror, see generally Thomas Healy, Brandenburg in a Time of Terror, 84 Notre Dame L. Rev. 655 (2009).

162See U.S. Const. amend. I.

163See Inazu, supra note 47, at 20–62 (providing a historical overview of the freedom of assembly in the United States); see also Tabatha Abu El-Haj, The Neglected Right of Assembly, 56 UCLA L. Rev. 543, 554–61 (2009) (discussing the relationship between assembly and politics in American history); Bhagwat, supra note 7, at 984–85 (discussing the treatment of the freedom of assembly by the courts in the first half of the twentieth century); Robert M. Chesney, Democratic-Republican Societies, Subversion, and the Limits of Legitimate Political Dissent in the Early Republic, 82 N.C. L. Rev. 1525, 1526, 1550–51 (2004) (highlighting the importance of the freedom of assembly); James Gray Pope, Republican Moments: The Role of Direct Popular Power in the American Constitutional Order, 139 U. Pa. L. Rev. 287, 330–32, 337–41 (1990) (discussing the origin of the freedom of assembly).

164See Bhagwat, supra note 7, at 983–85 (discussing, among other cases, De Jonge v. Oregon, 299 U.S. 353 (1937), in which the right to assemble with the Communist Party was protected); David Cole, Hanging with the Wrong Crowd: Of Gangs, Terrorists, and the Right of Association, 1999 Sup. Ct. Rev. 203, 216 & n.39.

165E.g., De Jonge, 299 U.S. at 364; accord Thomas v. Collins, 323 U.S. 516, 530 (1945).

166357 U.S. 449 (1958).

167The decline of the independent right of assembly/association and the rise of expressive association has been extensively recounted elsewhere, and here I aim only to briefly summarize the results of that scholarship. See Inazu, supra note 47, at 61–62, 79–84, 132–35, 141–44; Bhagwat, supra note 7, at 985–89.

168See Inazu, supra note 47, at 62.

169See supra Part II.A.

170See John D. Inazu, Virtual Assembly, 98 Cornell L. Rev. 1093, 1094–95 (2013).

171See id. at 1109–10.

172For a further defense of this principle, see Cole, supra note 164, at 226–29.

173I will henceforth describe the right as “freedom of association” because that is the dominant modern nomenclature. There is some historical support for the proposition that the term “assembly” contained in the constitutional text protected temporary gatherings, while the term “association” referred to more permanent groups, whose constitutional status was a bit less clear. See Jason Mazzone, Freedom’s Associations, 77 Wash. L. Rev. 639, 742–43 (2002). However, as the discussion in the text indicates, throughout most of our history the public and the courts used the terms assembly and association interchangeably and clearly presumed constitutional protection for permanent groups.

174See infra Part IV.

175See Boy Scouts of Am. v. Dale, 530 U.S. 640, 655 (2000).

176See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2723 (2010).

177See supra Part II.

178See Wadie E. Said, Humanitarian Law Project and the Supreme Court’s Construction of Terrorism, 2011 BYU L. Rev. 1455, 1507 & n.241 (citing David Cole, Terror Financing, Guilt by Association and the Paradigm of Prevention in the “War on Terror,” in Counterterrorism: Democracy’s Challenge 242 (Andrea Bianchi & Alexis Keller eds., 2008)). The Holy Land Foundation and Hammoud cases, discussed earlier, illustrate this point nicely. See supra Part II.A.

179See supra note 103 and accompanying text.

180See supra note 107 and accompanying text.

181For a similar argument, see Cole, supra note 164, at 247–48.

182See Pyetranker, supra note 54, at 38. As noted earlier, in affirming Mehanna’s conviction, the First Circuit abandoned this theory, almost conceding its legal insufficiency. See supra note 58.

183 See Abel, supra note 56, at 732 (citing Government’s Proffer and Memorandum in Support of Detention, supra note 56, at 11).

184Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2722–23 (2010).

185See id. at 2730.

186See Pyetranker, supra note 54, at 38.

187This argument is laid out in detail in Bhagwat, supra note 7, at 995–1002.

188The Complete Bill of Rights: The Drafts, Debates, Sources, and Origins 129 (Neil H. Cogan ed., 1997).

189Id. at 140.

190George Mason’s Master Draft of the Bill of Rights, Constitution.org., http://www.constitution.org/gmason/amd_gmas.htm (last updated Jan. 24, 2012).

191See The Complete Bill of Rights: The Drafts, Debates, Sources, and Origins, supra note 188, at 129, 140; George Mason’s Master Draft of the Bill of Rights, supra note 190.

192See John D. Inazu, The Forgotten Freedom of Assembly, 84 Tul. L. Rev. 565, 573–77 (2010).

193See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2722 (2010).

194See Abel, supra note 56, at 732.

195Gitlow v. New York, 268 U.S. 652, 673 (1925) (Holmes, J., dissenting). The Gitlow dissent, arising out of the anticommunist prosecutions of the first Red Scare, is, of course, one of the foundational opinions of modern First Amendment jurisprudence.

196Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (per curiam).

197See Abel, supra note 56, at 732.

198See id. at 732–33.

199See, e.g., Dennis v. United States, 341 U.S. 494, 497–98, 509–11 (1951); Whitney v. California, 274 U.S. 357, 363–64, 371–72 (1927), overruled per curiam by Brandenburg, 395 U.S. 444; Gitlow, 268 U.S. at 655–56, 664–67; Abrams v. United States, 250 U.S. 616, 617, 624 (1919).

200See Karas, supra note 78.

201See id.

202See Healy, supra note 161, at 655; supra text accompanying note 161.

203See supra Parts I.B, II.A.

204Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2725 (2010).

205See id. at 2736 (Breyer, J., dissenting).

206 See id. at 2730 (majority opinion).

207See Ashutosh A. Bhagwat, The Story of Whitney v. California: The Power of Ideas, in Constitutional Law Stories 407, 411–12 (Michael C. Dorf ed., 2004).

208See Whitney v. California, 274 U.S. 357, 379 (1927) (Brandeis & Holmes, JJ., concurring), overruled per curiam by Brandenburg v. Ohio, 395 U.S. 444 (1969); Bhagwat, supra note 207, at 407 (quoting Vincent Blasi, The First Amendment and the Ideal of Civic Courage: The Brandeis Opinion in Whitney v. California, 29 Wm. & Mary L. Rev. 653, 668 (1988)).

209See Humanitarian Law Project, 130 S. Ct. at 2737 (Breyer, J., dissenting) (citing United States v. Robel, 389 U.S. 258 (1967); Keyishian v. Bd. of Regents of Univ. of State of N.Y., 385 U.S. 589, 605–10 (1967); Elfbrandt v. Russell, 384 U.S. 11, 17 (1966); Scales v. United States, 367 U.S. 203, 228–30 (1961); De Jonge v. Oregon, 299 U.S. 353 (1937)).

210The Humanitarian Law Project majority’s only effort to respond to Justice Breyer’s argument was a cryptic footnote stating (without explanation) that because § 2339B does not ban “pure speech and association,” without more, reliance on the Communist Party cases is “unfounded.” Id. at 2723 n.4 (majority opinion) (internal quotation marks omitted).

211See supra Part II.A.

212Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 166 (D.C. Cir. 2003); accord Islamic Am. Relief Agency v. Gonzales, 477 F.3d 728, 736–37 (D.C. Cir. 2007); Humanitarian Law Project v. Reno, 205 F.3d 1130, 1133 (9th Cir. 2000).

213United States v. Hammoud, 381 F.3d 316, 329 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005).

214United States v. Afshari, 426 F.3d 1150, 1160 (9th Cir. 2005).

215Humanitarian Law Project, 130 S. Ct. at 2730.

216See, e.g., FEC v. Colo. Republican Fed. Campaign Comm., 533 U.S. 431, 440 (2001) (“Spending for political ends and contributing to political candidates both fall within the First Amendment’s protection of speech and political association.”); Wis. Right to Life State Political Action Comm. v. Barland, 664 F.3d 139, 152 (7th Cir. 2011) (quoting Colo. Republican, 533 U.S. at 440); SpeechNow.org v. FEC, 599 F.3d 686, 692 (D.C. Cir. 2010) (quoting Davis v. FEC, 554 U.S. 724, 740 n.7 (2008)); Carver v. Nixon, 72 F.3d 633, 636 (8th Cir. 1995) (quoting Buckley v. Valeo, 424 U.S. 1, 18 (1976)); Casino Ass’n of La. v. State, 2002-0265, p. 8 (La. 6/21/02); 820 So. 2d 494, 500.

217424 U.S. at 18 (“[T]he present Act’s contribution and expenditure limitations impose direct quantity restrictions on political communication and association by persons, groups, candidates, and political parties.”).

218See, e.g., Randall v. Sorrell, 548 U.S. 230, 247 (2006); Colo. Republican, 533 U.S. at 440; Nixon v. Shrink Mo. Gov’t PAC, 528 U.S. 377, 387–88 (2000).

219United States v. Hammoud, 381 F.3d 316, 328 n.3 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005) (“Hammoud relies in part on cases holding that a donation to a political advocacy group is a proxy for speech. Hizballah is not a political advocacy group, however.” (citation omitted)).

220See supra notes 84–91 and accompanying text.

221See Weiser, supra note 67.

222See supra notes 166–68 and accompanying text.

223See Bhagwat, supra note 7, at 990–94.

224Though, as is discussed in more detail in the next Part, there are alternative grounds upon which exclusion of FTOs from First Amendment protections might be grounded.

225Korematsu v. United States, 323 U.S. 214, 246 (1944) (Jackson, J., dissenting) (arguing that judicial statements sanctioning a racially discriminatory order could be used for harmful ends in future proceedings).

226See, e.g., Randall v. Sorrell, 548 U.S. 230, 236–37 (2006) (protecting an individual’s right to associate with a political candidate by providing financial contributions); NAACP v. Button, 371 U.S. 415, 428–29 (1963) (protecting an association’s right to provide legal representation to members); Shelton v. Tucker, 364 U.S. 479, 485–86, 490 (1960) (protecting an individual’s right not to disclose associational ties); NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 466 (1958) (protecting an association’s right to maintain its members’ anonymity).

227See, e.g., Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 166 (D.C. Cir. 2003).

228Cf. Terminiello v. Chicago, 337 U.S. 1, 37 (1949) (Jackson, J., dissenting) (“There is a danger that, if the Court does not temper its doctrinaire logic with a little practical wisdom, it will convert the constitutional Bill of Rights into a suicide pact.”).

229See supra Part II.A.

230See Inazu, supra note 47, at 62.

231U.S. Const. amend. I (emphasis added).

232See infra text accompanying notes 255, 289–94.

233See The Complete Bill of Rights: The Drafts, Debates, Sources, and Origins, supra note 188, at 144 (referencing the speech of Elbridge Gerry).

234De Jonge v. Oregon, 299 U.S. 353, 365 (1937); see also Mark Tushnet, The Hughes Court and Radical Political Dissent: The Cases of Dirk De Jonge and Angelo Herndon, 28 Ga. St. U. L. Rev. 333, 358 (2012) (quoting De Jonge, 299 U.S. at 365).

235See supra note 173; see also Inazu, supra note 47, at 166–67.

236See supra notes 172–73 and accompanying text.

237See Akhil Reed Amar, The Bill of Rights: Creation and Reconstruction 28, 30 (1998).

238See Cole, supra note 164, 231–32 (citing 1 Alexis de Tocqueville, Democracy in America 202–03 (Phillips Bradley ed., 1948)); cf. George Kateb, The Value of Association, in Freedom of Association 35, 37 (Amy Gutmann ed., 1998) (arguing that the act of associating is valuable in and of itself).

239See Anonymous (Internet Group), N.Y. Times, http://topics.nytimes.com/top/reference/timestopics/organizations/a/anonymous_internet_group/ (last visited Jan. 9, 2014).

240See Jeb Boone, Who Are the Izz ad-Din al-Qassam Cyber Fighters?, GlobalPost (Nov. 9, 2012), http://www.globalpost.com/dispatches/globalpost-blogs/the-grid/who-are-the-izz-ad-din-al-qassam-cyber-fighters; Tracy Kitten, DDoS Attacks on Banks Resume, Bank Info Security (Feb. 26, 2013), http://www.bankinfosecurity.com/ddos-attacks-on-banks-resume-a-5541.

241This is not to say that a simple act of patent or copyright infringement would suffice to eliminate a group’s constitutional protections. The notion that such intellectual property qualifies as “property” for these purposes or that simple infringement qualifies as destruction of property seems highly dubious, though the topic is beyond the scope of this Article.

242Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010).

243See, e.g., United States v. Verdugo-Urquidez, 494 U.S. 259, 261 (1990) (holding that the Fourth Amendment does not apply to the search and seizure of property owned by a nonresident alien and located in a foreign country); Al Maqaleh v. Gates, 605 F.3d 84, 99 (D.C. Cir. 2010) (holding that “the jurisdiction of the courts to afford the right to habeas relief and the protection of the Suspension Clause does not extend to aliens held in Executive detention in the Bagram detention facility in the Afghan theater of war”); Kiyemba v. Obama, 555 F.3d 1022, 1026 (D.C. Cir. 2009), vacated, 559 U.S. 131 (2010) (noting that the “due process clause does not apply to aliens without property or presence in the sovereign territory of the United States”); United States v. Vilches-Navarette, 523 F.3d 1, 13 (1st Cir. 2008) (explaining that the Fourth Amendment does not apply when the United States acts against aliens in international waters).

244See, e.g., Abu El-Haj, supra note 163, at 547, 554–55, 586–89; Bhagwat, supra note 7, at 991–93; Inazu, supra note 192, at 571–77; Mazzone, supra note 173, at 647, 729–30; see also Amar, supra note 237, at 28–30 (arguing that the Assembly Clause protects a right of the sovereign people to assemble in convention and change their government).

245See, e.g., 2 Alexis de Tocqueville, Democracy in America 513 (J.P. Mayer ed., George Lawrence trans., Anchor Books 1969) (1840).

246See Buckley v. Valeo, 424 U.S. 1, 15 (1976) (quoting Kusper v. Pontikes, 414 U.S. 51, 56, 57 (1973)); Cousins v. Wigoda, 419 U.S. 477, 487 (1975).

247I leave aside the more difficult question of foreign citizens residing in the United States. For an argument that self-governance principles support extending the protections of the Bill of Rights to this group, see Ashutosh Bhagwat, The Myth of Rights: The Purposes and Limits of Constitutional Rights 264–66 (2010).

248See supra Part II.A.

249See Lamont v. Postmaster Gen., 381 U.S. 301, 306–07 (1965); see also Aptheker v. Sec’y of State, 378 U.S. 500, 505 (1964) (striking down, on right-to-travel grounds, a statute denying passports to all members of registered Communist organizations).

250See NAACP v. Claiborne Hardware Co., 458 U.S. 886, 920 (1982) (“Civil liability may not be imposed merely because an individual belonged to a group, some members of which committed acts of violence.”).

251I do not consider here (because they are beyond the scope of this Article) the difficult questions of proof that arise in implementing this restriction, including in particular the extent to which entities may be designated as violent through an administrative process to which the courts then defer in subsequent legal proceedings. See United States v. Hammoud, 381 F.3d 316, 331 (4th Cir. 2004) (en banc), vacated, 543 U.S. 1097 (2005); Holy Land Found. for Relief & Dev. v. Ashcroft, 333 F.3d 156, 162–63 (D.C. Cir. 2003). For a description and critique of the administrative process for designating FTOs, see Sahar F. Aziz, Note, The Laws on Providing Material Support to Terrorist Organizations: The Erosion of Constitutional Rights or a Legitimate Tool for Preventing Terrorism?, 9 Tex. J. C.L. & C.R. 45, 91–92 (2003).

252See 18 U.S.C. § 2339B (2012).

253See id. § 2339A.

254See United States v. McVeigh, 153 F.3d 1166, 1166 (10th Cir. 1998) (upholding conviction of bomber Timothy McVeigh).

255See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2733 (2010) (Breyer, J., dissenting).

256See id. at 2722 (majority opinion).

257See supra Part II.A.

258See Buckley v. Valeo, 424 U.S. 1, 39, 44–45, 47–48 (1976); Citizens United v. FEC, 558 U.S. 310, 319 (2010).

259See 11 C.F.R. § 109.21(b) (2012); Buckley, 424 U.S. at 46–47.

260See, e.g., Long Beach Area Chamber of Commerce v. City of Long Beach, 603 F.3d 684, 687 (9th Cir. 2010); Clifton v. FEC, 114 F.3d 1309, 1311 (1st Cir. 1997); United States v. Goland, 959 F.2d 1449, 1452 (9th Cir. 1992); Nader v. FEC, 823 F. Supp. 2d 53, 59 (D.D.C. 2011); Cao v. FEC, 688 F. Supp. 2d 498, 514–15 (E.D. La. 2010); FEC v. Christian Coal., 52 F. Supp. 2d 45, 92 (D.D.C. 1999); Colo. Educ. Ass’n v. Rutt, 184 P.3d 65, 74–75 (Colo. 2008).

261See 11 C.F.R. § 109.21(a).

262For a discussion of the coordination concept, arguing that the campaign finance definition does not translate well into the terrorism area, see generally Catherine A. Hardee, The Coordination Conundrum, 49 Willamette L. Rev. 189 (2012).

263See Brandenburg v. Ohio, 395 U.S. 444, 447 (1969) (per curiam) (allowing states to “forbid or proscribe advocacy . . . directed to inciting or producing imminent lawless action and . . . likely to incite or produce such action”).

264367 U.S. 203 (1961).

265Id. at 209, 229–30.

266See id. at 229. For a similar argument, see Williams, supra note 109, at 391–92. It must be acknowledged that in Brandenburg itself, the Court briefly addressed freedom of assembly at the end of its per curiam opinion and suggested in a footnote that prosecution for assembly must satisfy the same requirements as prosecution of speech. Brandenburg, 395 U.S. at 449 n.4. Brandenburg, however, has not been understood to overrule Scales, as indicated by the fact that in Humanitarian Law Project the Court approvingly cites and describes Scales without any hint that it was inconsistent with Brandenburg. See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2718 (2010).

267See Abu El-Haj, supra note 163, at 554–61; Bhagwat, supra note 7, at 993, 997.

268See Whitney v. California, 274 U.S. 357, 372 (1927) (“That . . . united and joint action involves even greater danger to the public peace and security than the isolated utterances and acts of individuals, is clear. We cannot hold that, as here applied, the Act is an unreasonable or arbitrary exercise of the police power of the State, unwarrantably infringing any right of free speech, assembly or association, or that those persons are protected from punishment by the due process clause who abuse such rights by joining and furthering an organization thus menacing the peace and welfare of the State.”), overruled per curiam by Brandenburg v. Ohio, 395 U.S. 444 (1969).

269See supra notes 152–54 and accompanying text.

270See supra Part II.A.3.

271See supra note 152 and accompanying text.

272See supra Part II.A.1; see also Press Release, U.S. Attorney’s Office, S. Dist. of N.Y., supra note 75.

273See Weiser, supra note 67.

274See Karas, supra note 78.

275See O’Hagan, supra note 109.

276Cf. Watchtower Bible & Tract Soc’y of N.Y., Inc. v. Vill. of Straton, 536 U.S. 150, 160–64 (2002) (finding that an ordinance that required a permit to engage in door-to-door advocacy violated the First Amendment because it would have placed a burden on some speech).

277See Abel, supra note 56, at 739–40. As noted earlier, the First Circuit abandoned this theory in affirming Mehanna’s conviction, see United States v. Mehanna, 735 F.3d 32, 50–51 (1st Cir. 2013), and concluded (rather unconvincingly) that permitting the evidence supporting the government’s “translation-as-material-support theory” to go the jury did not prejudice Mehanna. See id. at 59–64. But it remains true that the primary theory on which Mehanna was prosecuted in the trial court was based on his independent speech, and the trial court judge permitted this theory to go to the jury, despite the lack of evidence of coordination. Id. at 47. Moreover, though the district court instructed the jury that only coordinated speech violated the material-support statute, it failed to directly define coordination, see id. at 48–49, leaving open the possibility (which the First Circuit essentially ignored) that the jury convicted based on the theory of unilateral coordination that the government advanced.

278See Abel, supra note 56, at 734.

279See U.S. Const. amend. I.

280Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2713 (2010).

281See Kulish, supra note 71.

282See Steven Erlanger, Hamas Seizes Broad Control in Gaza Strip, N.Y. Times, June 14, 2007, http://www.nytimes.com/2007/06/14/world/middleeast/14mideast.html?_r=0; Hamas, N.Y. Times, http://topics.nytimes.com/top/reference/timestopics/organizations/h/hamas/index.html (last visited Jan. 9, 2014).

283 See Humanitarian Law Project, 130 S. Ct. at 2724–27.

284See id. at 2725.

285Id.

286See id. at 2726–27.

287See id. at 2727–29.

288Id. at 2717.

289 See 299 U.S. 353, 364–65 (1937).

290See 367 U.S. 203, 229–30 (1961).

291See Brandenburg v. Ohio, 395 U.S. 444, 447–48 (1969) (per curiam).

292See 354 U.S. 298, 318 (1957) (construing the Smith Act to forbid only “advocacy directed at promoting unlawful action,” not “advocacy of abstract doctrine”).

293See, e.g., Dennis v. United States, 341 U.S. 494, 497–98, 509–11 (1951); Whitney v. California, 274 U.S. 357, 363–64, 371–72 (1927), overruled per curiam by Brandenburg, 395 U.S. at 449; Gitlow v. New York, 268 U.S. 652, 655–56, 664–67 (1925); Abrams v. United States, 250 U.S. 616, 617, 624 (1919).

294One question that arises is how to treat an act of association by an individual who believes that the group she has joined is nonviolent, but which is in fact a dual-function group. To grant no protection for such behavior could be quite chilling because it would discourage association with dissident groups absent a very high degree of confidence that the group is nonviolent. On the other hand, the Humanitarian Law Project Court was certainly correct to conclude that the government may legitimately bar association by individuals who do not intend to further an FTO’s violent ends, given the fungibility of money and other tangible resources. Perhaps the solution here is a mens rea requirement focused on knowledge, but not intent—association may be punished if an individual is aware, or should have been aware, of a group’s violent nature even if the individual does not specifically intend to support violence.

295See supra note 251.

296See Holder v. Humanitarian Law Project, 130 S. Ct. 2705, 2730 (2010).

297See Kleindienst v. Mandel, 408 U.S. 753, 764–70 (1972).

298See Zemel v. Rusk, 381 U.S. 1, 16–17 (1965).

299See Humanitarian Law Project, 130 S. Ct. at 2727.

300See About the Ku Klux Klan, Anti-Defamation League, http://archive.adl.org/learn/ext_us/kkk/default.asp (last visited Jan. 9, 2014).

301Bureau of Counterterrorism, supra note 72.

302See Gerry Adams and Sinn Fein, PBS, http://www.pbs.org/wgbh/pages/frontline/shows/ira/conflict/gasf.html (last visited Jan. 9, 2014).

303See IRA Statement, Guardian (July 28, 2005), http://www.guardian.co.uk/politics/2005/jul/28/northernireland.devolution.

304See id.

305IRA ‘Has Destroyed All Its Arms,BBC News (Sept. 26, 2005), http://news.bbc.co.uk/2/hi/uk_news/northern_ireland/4283444.stm.

306See IRA Statement, supra note 303.

307Letter from Yasser Arafat, Chairman, The Palestine Liberation Org., to Yitzhak Rabin, Prime Minister of Isr. (Sept. 9, 1993), http://www.incore.ulst.ac.uk/services/cds/agreements/pdf/is4.pdf.

308This is not to say, of course, that individual members of these organizations are immune from punishment for past crimes.

309See 278 U.S. 63 (1928). Thanks to John Inazu for pointing me to this issue, which he discusses in more detail in Inazu, supra note 47, at 50–51.

310Bryant, 278 U.S. at 65–66.

311Id. at 71.

312Id. at 65.

313See id. at 73, 77.

314See id. at 75–77.

315357 U.S. 449, 465 (1958).

316Id. at 462–63.

317Brandenburg v. Ohio, 395 U.S. 444, 444–45 (1969).

318Id. at 449 & n.4.

319Inazu, supra note 47, at 51.

320See Brandenburg, 395 U.S. at 444–45.

321See Virginia v. Black, 538 U.S. 343, 352–57 (2003).