Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
Congressional research reportMay 6, 2019
Ask Donna
What actually matters in this document.
Text
Terrorism, Violent Extremism, and the
Internet: Free Speech Considerations
Victoria L. Killion
Legislative Attorney
May 6, 2019
Congressional Research Service
7-....
SUMMARY
Terrorism, Violent Extremism, and the
Internet: Free Speech Considerations
R45713
May 6, 2019
Victoria L. Killion
Recent acts of terrorism and hate crimes have prompted a renewed focus on the possible links
Legislative Attorney
-re-acte--@crs.loc.gov
between internet content and offline violence. While some have focused on the role that social
media companies play in moderating user-generated content, others have called for Congress to
For a copy of the full report,
pass laws regulating online content promoting terrorism or violence. Proposals related to
please call 7-.... or visit
government action of this nature raise significant free speech questions, including (1) the reach of
www.crs.gov.
the First Amendment’s protections when it comes to foreign nationals posting online content
from abroad; (2) the scope of so-called “unprotected” categories of speech developed long before
the advent of the internet; and (3) the judicial standards that limit how the government can craft or enforce laws to preserve
national security and prevent violence.
At the outset, it is not clear that a foreign national (i.e., a non-U.S. citizen or resident) could invoke the protections of the
First Amendment in a specific U.S. prosecution or litigation involving online speech that the foreign national posted from
abroad. The Supreme Court has never directly opined on this question. However, its decisions regarding the extraterritorial
application of other constitutional protections to foreign nationals and lower court decisions involving speech made by
foreign nationals while outside of the United States suggest that the First Amendment may not apply in that scenario. In
contrast, free speech considerations are likely to be highly relevant in evaluating the legality of (1) proposals for the U.S.
government to regulate what internet users in the United States can post, or (2) the enforcement of existing U.S. laws where
the government seeks to hold U.S. persons liable for their online speech.
Although the government typically can regulate conduct without running afoul of the First Amendment, regulations that
restrict or burden expression often do implicate free speech protections. In such circumstances, courts generally distinguish
between laws that regulate speech on the basis of its content (i.e., the topic discussed or the message expressed) and those
that do not, subjecting the former to more stringent review. A law that expressly restricts online communications or media
promoting violence or terrorism is likely to be deemed a content-based restriction on speech; whereas a law that primarily
regulates conduct could be subject to a less stringent standard of review, unless its application to speech turns on the message
expressed. Whether such laws would survive First Amendment scrutiny depends on a number of factors.
Over the past 50 years, the Supreme Court has generally extended the First Amendment’s free speech protections to speech
that advocates violence in the abstract while allowing the government to restrict or punish speech that threatens or facilitates
violence in a more specific or immediate way. The subtle distinctions that have developed over time are reflected in the
categories of speech that the court has deemed unprotected, meaning that the government generally can prohibit speech in
these areas because of its content. These include incitement to imminent lawless action, true threats, and speech integral to
criminal conduct. Although judicial decisions have helped to define the scope of some of these categories, open questions
remain as to how they apply in the context of online speech. For instance, legal scholars have questioned what it means for
speech to incite “imminent” violence when posted to social media. They have also asked how threats should be perceived
when made in the context of online forums where hyperbolic speech about violence is common.
The extent to which the government can regulate speech promoting violence or terrorism also depends on whether its law or
action satisfies the applicable level of scrutiny that the Court has developed to evaluate measures that restrict or burden
speech. In general, laws that regulate protected speech on political or ideological matters are subject to strict scrutiny, a test
that requires the government to demonstrate that its law is narrowly tailored to achieve a compelling governmental interest.
Nevertheless, in some cases, courts have concluded that the government’s national security interests justify restrictions on
protected speech, such as in 2010 when the Supreme Court upheld certain applications of a federal statute prohibiting
providing material support to U.S.-designated foreign terrorist organizations.
Congressional Research Service
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
Contents
Background ..................................................................................................................................... 4
The First Amendment and Foreign Speakers .................................................................................. 8
The Conduct Versus Speech Distinction ....................................................................................... 13
Content-Based Laws...................................................................................................................... 15
Protected and Unprotected Speech ................................................................................................ 17
(1) Brandenburg’s Incitement Standard .................................................................................. 19
(2) “True Threats” ................................................................................................................... 23
(3) Speech Integral to Criminal Conduct ................................................................................ 27
Strict Scrutiny and Overbreadth .................................................................................................... 28
Strict Scrutiny ......................................................................................................................... 28
Governmental Interests ..................................................................................................... 29
Narrow Tailoring ............................................................................................................... 30
Deference to Political Branches ........................................................................................ 31
The Overbreadth Doctrine....................................................................................................... 35
Conclusion ..................................................................................................................................... 37
Figures
Contacts
Author Contact Information .......................................................................................................... 39
Congressional Research Service
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
s the Supreme Court has observed, while the First Amendment protects the “freedom
of speech,”1 it “does not protect violence.”2 But when speech promotes violence, a
tension can form between the values of liberty and security.3 In an oft-quoted passage
from a dissenting opinion, Justice Robert Jackson argued that the problems this
tension creates are not insurmountable but must be confronted with a dose of
pragmatism: a government can temper “liberty with order,” but to treat free speech as absolute
threatens to “convert the constitutional Bill of Rights into a suicide pact.”4 While Justices of the
Court have often disagreed over when free speech rights must yield to the government’s
interests,5 when it comes to speech promoting violence, the Court has rejected an all-or-nothing
approach.6 Over the past 50 years, the Court has drawn a line between speech that advocates
violence in the abstract and speech that facilitates it in a specific way, with the former receiving
more robust constitutional protections.7 It has done so because, in the Court’s view, upholding the
First Amendment requires preserving “uninhibited, robust, and wide-open” debate on public
issues,8 even if that means allowing individuals to express ideas that are “deeply offensive to
many.”9
A
1 U.S. CONST. amend. I (“Congress shall make no law . . . abridging the freedom of speech . . . .”).
2 NAACP v. Claiborne Hardware Co., 458 U.S. 886, 916 (1982).
3 See United States v. Mehanna, 735 F.3d 32, 40 (1st Cir. 2013) (“Sometimes, [the government’s efforts to combat
terrorism] require a court to patrol the fine line between vital national security concerns and forbidden encroachments
on constitutionally protected freedoms of speech and association.”).
4 See Terminiello v. Chicago, 337 U.S. 1, 37 (1949) (Jackson, J., dissenting) (“The choice is not between order and
liberty. It is between liberty with order and anarchy without either. There is 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.”);
Haig v. Agee, 453 U.S. 280, 284, 309-10 (1981) (reiterating that the Constitution “is not a suicide pact” in rejecting the
First Amendment and due process arguments of a former Central Intelligence Agency (CIA) employee who challenged
the U.S. government’s revocation of his U.S. passport for repeatedly exposing the identities of alleged undercover CIA
agents and sources abroad with “the declared purpose of obstructing intelligence operations”).
5 See, e.g., Holder v. Humanitarian Law Project, 561 U.S. 1, 41 (2010) (Breyer, Ginsburg & Sotomayor, JJ., dissenting)
(dissenting from the Court’s opinion insofar as the majority held that the government may, consistent with the First
Amendment, “prosecute the plaintiffs criminally for engaging in coordinated teaching and advocacy furthering the
designated [foreign terrorist] organizations’ lawful political objectives”); United States v. Stevens, 559 U.S. 460, 482500 (2010) (Alito, J., dissenting) (dissenting from the Court’s opinion holding that a law banning certain depictions of
animal cruelty violated the First Amendment); Dennis v. United States, 341 U.S. 494, 582-83 (1951) (Douglas, J.,
dissenting) (dissenting from the Court’s opinion affirming convictions for conspiracy to advocate the overthrow of the
U.S. government, stating, “[W]hat petitioners did was to organize people to teach and themselves teach the MarxistLeninist doctrine contained chiefly in four books . . . . But if the books themselves are not outlawed, if they can
lawfully remain on library shelves, by what reasoning does their use in a classroom become a crime?”).
6 See Claiborne Hardware Co., 458 U.S. at 928 (reasoning that when “spontaneous and emotional appeals for unity and
action in a common cause” do not “incite lawless action, they must be regarded as protected speech”); cf. Brown v.
Entm’t Merchs. Ass’n, 564 U.S. 786, 804 (2011) (asking whether violent video games “constitute a ‘well-defined and
narrowly limited clas[s] of speech, the prevention and punishment of which have never been thought to raise any
Constitutional problem,’ . . . ; and if not, whether the regulation of such works is justified by . . . a compelling state
interest” (internal citation omitted)).
7 See Brandenburg v. Ohio, 395 U.S. 444, 449 (1969) (per curiam) (distinguishing “mere advocacy” from “incitement
to imminent lawless action”). See the section entitled “Protected and Unprotected Speech” infra for a discussion of the
degree to which speech advocating violence is protected under the First Amendment.
8 See N.Y. Times Co. v. Sullivan, 376 U.S. 254, 270 (1964) (observing the “profound national commitment to the
principle that debate on public issues should be uninhibited, robust, and wide-open, and that it may well include
vehement, caustic, and sometimes unpleasantly sharp attacks on government and public officials”).
9 United States v. Eichman, 496 U.S. 310, 318 (1990) (holding that defendants’ prosecutions for violating a federal law
prohibiting flag desecration violated the First Amendment); see also Snyder v. Phelps, 562 U.S. 443, 460-61 (2011)
(“As a Nation we have chosen . . . to protect even hurtful speech on public issues to ensure that we do not stifle public
Congressional Research Service
1
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
In more recent years, some have begun to question whether and how the Court’s decisions in this
area should apply to speech online.10 The “vast democratic fora of the Internet”11 have provided
ample platforms not only for those seeking to debate issues or express controversial views, but
also for individuals and entities planning violent attacks or threatening violence online.12 Many
policymakers and commentators, including some Members of Congress, have expressed concerns
about the proliferation of social media content13 promoting terrorism and violence14 and the
influence such speech can have on other internet users.15 Some have called on Congress to restrict
or even prohibit such content,16 which raises the question of whether the First Amendment would
debate.”).
10 See, e.g., Michael J. Sherman, Brandenburg v. Twitter, 28 GEO. MASON U. CIV. RTS. L.J. 127, 135-36 (2018) (stating
that while difficult to predict, the Supreme Court “might decide that the new realities of social media require a rethinking of incitement and imminence standards established in 1969”); Lyrissa Barnett Lidsky & Linda Riedemann
Norbut, #I[gun image]U: Considering the Context of Online Threats, 106 CALIF. L. REV. 1886, 1889 (2018) (arguing
that the Supreme Court “has done relatively little to guide police, prosecutors, lower court judges, and juries seeking to
make the necessary and difficult distinctions” between constitutionally protected speech online and true threats).
11 Reno v. ACLU, 521 U.S. 844, 868 (1997).
12 Wash. Post v. McManus, No. PWG-18-2527, 2019 U.S. Dist. LEXIS 3073, at *50 (D. Md. Jan. 3, 2019) (“[F]or all
of its many benefits, the Internet has opened up virtually limitless opportunities for hostile foreign powers, criminals,
and terrorists to sow mischief and undermine our democratic institutions.”), appeal filed, No. 19-1132 (4th Cir. Feb. 4,
2019); see also United States v. Wheeler, 776 F.3d 736, 745 n.4 (10th Cir. 2015) (“Several attributes of the Internet
substantially amplify the fear an individual can instill via threats or incitement. Such threats have the ability to reach a
vast audience—far more than the traditional speaker or author published in a single venue. The threats may often come
cloaked in anonymity, allowing authors to make menacing statements they would never consider making to an
individual in person. And, given the prevalence and diversity of Internet fora and discussion boards, such exhortations
may often find a receptive audience of like-minded individuals—perhaps audiences more willing to do the bidding of
one urging violent action.”).
13 Social media refers to “forms of electronic communication (such as websites for social networking and
microblogging) through which users create online communities to share information, ideas, personal messages, and
other content (such as videos).” Social Media, MERRIAM-WEBSTER (2019), https://www.merriam-webster.com/
dictionary/social%20media. See generally CRS Report R45337, Social Media Adoption by Members of Congress:
Trends and Congressional Considerations 7, by Jacob R. Straus (providing examples of social media platforms and
explaining how established social media platforms like Facebook and Twitter “provide users with the ability to post
multiple types of content (e.g., video, pictures, and text),” while other platforms “generally, though not universally,
specialize in a single type of media”).
14 This report refers to the concepts of “promoting terrorism” and “promoting violence” to encompass a range of speech
(e.g., advocacy, incitement, recordings of violent crime). The use of this shorthand is not meant to equate one form of
speech to another in terms of the threat that it may pose to U.S. national security or its connection to criminal activity,
but rather to facilitate the discussion of First Amendment principles and standards that may affect the government’s
ability to regulate internet speech to prevent violence.
15 See Terrorism and Social Media: #IsBigTechDoingEnough?: Hearing Before the S. Comm. on Commerce, Sci., &
Transp., 115th Cong. 1 (2018) [hereinafter Terrorism and Social Media Hearing] (statement of Chairman John Thune)
(expressing the view that “[w]e all have a vested interest” in social media companies’ success in identifying and
removing terrorist content and that “this Committee has a significant role to play in overseeing the effectiveness of their
efforts”); see id. (statement of Ranking Member Bill Nelson) (remarking that “it is startling that today, a terrorist can be
radicalized and trained to conduct attacks all through social media” and “then a terrorist cell can activate that individual
to conduct an attack through the internet—creating in effect a terrorist drone controlled by social media”); Tarleton
Gillespie, How Social Networks Set the Limits of What We Can Say Online, WIRED (June 26, 2018, 7:00 AM),
https://www.wired.com/story/how-social-networks-set-the-limits-of-what-we-can-say-online/ (noting that “a slow
reconsideration of platform responsibility is under way” and that “[p]ublic and policy concerns around illicit content,
initially focused on sexually explicit and graphically violent images, have expanded to include hate speech, self-harm,
propaganda, and extremism”). See generally Rachel E. VanLandingham, Jailing the Twitter Bird: Social Media,
Material Support to Terrorism, and Muzzling the Modern Press, 39 CARDOZO L. REV. 1, 13 (2017) (“The link between
social media platforms and terrorism competes with privacy concerns as one of the most discussed and most
concerning, dynamics emanating from modern society’s explosive utilization of these communication technologies.”).
16 E.g., Alexander Tsesis, Opinion, Inciting Terror on the Internet Can Be Regulated. Congress Needs to Act, CNN
Congressional Research Service
2
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
allow such regulation.17 A number of legal scholars have explored these issues, and some have
proposed recommendations to Congress about how best to address these concerns in accordance
with the First Amendment.18
Although governmental efforts to combat online content promoting terrorism or violence could
take a number of forms, this report focuses on the First Amendment implications of imposing
civil or criminal liability on individual internet users (i.e., the originators of the content) rather
than the social media companies or internet service providers themselves. As such, it focuses on
the underlying First Amendment issues that are likely to be common to both forms of government
action, but does not discuss the additional considerations attendant to regulating a social media
platform, which are the subject of another CRS report.19 The report begins with some background
on the use of the internet by terrorist groups and the reported influence of online content on
certain individuals accused of committing violent attacks. It then considers the question of who
can invoke the First Amendment by analyzing whether its free speech protections apply to foreign
BUSINESS (Mar. 23, 2019, 8:31 AM), https://www.cnn.com/2019/03/23/perspectives/new-zealand-terrorism-internetregulate/index.html; see also Mark Zuckerberg, Opinion, The Internet Needs New Rules. Let’s Start in These Four
Areas, WASH. POST (Mar. 30, 2019), https://www.washingtonpost.com/opinions/mark-zuckerberg-the-internet-needsnew-rules-lets-start-in-these-four-areas/2019/03/29/9e6f0504-521a-11e9-a3f7-78b7525a8d5f_story.html (“I believe we
need a more active role for governments and regulators. . . . Regulation could set baselines for what’s prohibited and
require companies to build systems for keeping harmful content to a bare minimum.”). See generally DANIELLE KEATS
CITRON, CATO INSTITUTE, POLICY ANALYSIS: WHAT TO DO ABOUT THE EMERGING THREAT OF CENSORSHIP CREEP ON
THE INTERNET 3 (2017) (discussing indicators of “how much the debate has moved toward government oversight of
digital speech” globally).
17 See U.S. CONST. amend. I (“Congress shall make no law . . . abridging the freedom of speech . . . .” (emphasis
added)). Compare Eric Posner, ISIS Gives Us No Choice But to Consider Limits on Speech, SLATE (Dec. 15, 2015),
https://slate.com/news-and-politics/2015/12/isiss-online-radicalization-efforts-present-an-unprecedented-danger.html
(“We do not currently face a national emergency comparable to a world war, but anti-propaganda laws may nonetheless
be warranted because of the unique challenge posed by ISIS’s sophisticated exploitation of modern technology.”), with
David Post, Protecting the First Amendment in the Internet Age, WASH. POST: VOLOKH CONSPIRACY (Dec. 21, 2015),
https://www.washingtonpost.com/news/volokh-conspiracy/wp/2015/12/21/protecting-the-first-amendment-in-theinternet-age/ (“True, limits on speech have been ‘tolerated . . . during times of national emergency’ [referring to
Professor Eric Posner’s article], but we look back at them—at least, I look back at them—as deeply misguided,
counterproductive and often shameful.”). See generally Cecilia Kang, Can Tech Companies Silence Hate Speech?,
N.Y. TIMES, Apr. 22, 2019, at B1 (stating that “Mr. Zuckerberg’s call for action, and his lobbyists’ response [suggesting
that Zuckerberg’s op-ed was not directed at U.S. regulators], encapsulate why the United States is on an island of its
own when it comes to managing violent and racist speech online” and reporting that while “Britain, Germany,
Australia, New Zealand and India have adopted or are considering laws that require stricter content moderation by tech
platforms[,] . . . none of them need to work around free speech protections like the First Amendment in the United
States”).
18 See Alexander Tsesis, Essay: Terrorist Speech on Social Media, 70 VAND. L. REV. 651, 708 (2017) (arguing that a
“narrowly tailored, multi-pronged law should be grounded on permissible restrictions against incitement, material
support for terror, and true threats” and that “[t]hese three separate doctrines can be used to stem the growing volume
of terrorist recruitment, indoctrination, incitement, and coordination available on social media”); Posner, supra note 17
(suggesting “a law that makes it a crime to access websites that glorify, express support for, or provide encouragement
for ISIS or support recruitment by ISIS; to distribute links to those websites or videos, images, or text taken from those
websites; or to encourage people to access such websites by supplying them with links or instructions,” and proposing
changes to First Amendment standards to accommodate such a law); Daniel Hoffman, Online Terrorism Advocacy:
How AEDPA and Inchoate Crime Statutes Can Simultaneously Protect America’s Safety and Free Speech, 2 NAT’L
SEC. L.J. 200, 251 (2014) (arguing that new laws to address “online terrorism advocacy” are unnecessary because
“existing AEDPA [Antiterrorism and Effective Death Penalty Act] and inchoate crime statutes and their associated case
law” give prosecutors “the tools to explore and better evolve the legal boundaries that Congress intended based on the
threat”).
19 See CRS Report R45650, Free Speech and the Regulation of Social Media Content, by Valerie C. Brannon
(discussing potential legal barriers to regulating social media companies’ decisions to host or remove content).
Congressional Research Service
3
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
nationals20 when they post online content from abroad, for example, in circumstances such as
U.S. prosecutions where online content is introduced as evidence of a crime.21 The report then
discusses the overarching First Amendment principles that bear on what the government may
regulate under the First Amendment, including (1) the distinction between regulating conduct and
speech; (2) the presumed invalidity of content-based laws; and (3) relevant “unprotected”
categories of speech that generally can be restricted because of their content. The report next
discusses the strict scrutiny standard and the overbreadth doctrine, which impose limitations on
how the government can regulate by requiring that laws restricting speech be sufficiently tailored
and not so broad as to chill protected speech.22 Finally, the report concludes with some
considerations for Congress in evaluating the constitutionality of regulating online content
promoting terrorism or violence.
Background
According to the Federal Bureau of Investigation (FBI), the internet and, in particular, the use of
social media are among the key “factors [that] have contributed to the evolution of the terrorism
threat landscape” since the September 11, 2001, terrorist attacks.23 Certain organizations that
track or study hate crimes also cite the internet as a tool used to intimidate and harass people
because of their race, ethnicity, religion, sexual orientation, or other attributes.24 At the same time,
20 This report uses the term “foreign national” to refer to a person who is a citizen of a country other than the United
States and who is not a U.S. citizen or lawful permanent resident. Cf. 52 U.S.C. § 30121(b)(2) (defining “foreign
national” for purposes of certain federal election laws to include “an individual who is not a citizen of the United States
or a national of the United States (as defined in section 101(a)(22) of the Immigration and Nationality Act) and who is
not lawfully admitted for permanent residence, as defined by section 101(a)(20) of the Immigration and Nationality
Act”).
21 Outside of free speech concerns, other legal considerations, such as the United States’ jurisdiction over a criminal
defendant or civil litigant, the extraterritorial application of U.S. statutes, and the potential applicability of international
agreements, are beyond the scope of this report.
22 See generally Richard H. Fallon, Jr., Making Sense of Overbreadth, 100 YALE L.J. 853, 864-65 (1991) (“[S]peech
may be privileged under current [First Amendment] doctrine either because it belongs to a constitutionally protected
category, or because it merits protection as the result of a balancing test. Which type of privilege is relevant typically
depends on the type of government regulation in question. If the government endeavors to regulate speech on the basis
of content, the first and frequently dispositive question is whether the speech falls within a protected category. . . . By
contrast, when speech enjoys full First Amendment protection, the state generally may not regulate it on the basis of
content, even if the speech is harmful, unless the regulation is necessary to advance some compelling government
interest.” (footnotes omitted)).
23 See FBI, What We Investigate: Terrorism, https://www.fbi.gov/investigate/terrorism (last visited Apr. 23, 2019)
(citing the internet, the use of social media, and “homegrown violent extremists,” which the FBI defines as “globaljihad-inspired individuals who are based in the U.S.,” as “[t]hree factors [that] have contributed to the evolution of the
terrorism threat landscape”).
24 See Hate Crimes and the Rise of White Nationalism: Hearing Before the H. Comm. on the Judiciary, 116th Cong. 2
(2019) [hereinafter Hate Crimes Hearing] (statement of Kristen Clarke, President and Executive Director, Lawyers’
Committee for Civil Rights Under Law) (stating, “[T]he actions of online white supremacists are new in form but not
substance. By directing hateful threats, intimidation, and harassment online at African Americans, Latinos, immigrants,
Muslims, Jews, and other historically marginalized communities, they follow the same script as generations of white
supremacists that assaulted civil rights activists at lunch counters, defaced houses of worship, and berated children on
their way to school.”); id. at 12 (statement of Eileen Hershenov, Senior Vice President, Policy, Anti-Defamation
League) (positing that anonymous “‘imageboards,’ a type of online discussion forum originally created to share
images,” have contributed to the “toxicity on social media,” and linking these forums to “targeted [online] harassment
campaign[s]”); see also Rachel Hatzipanagos, How Online Hate Turns Into Real-Life Violence, WASH. POST (Nov. 30,
2018), https://www.washingtonpost.com/nation/2018/11/30/how-online-hate-speech-is-fueling-real-life-violence/
(reporting that “[s]everal incidents in recent years have shown that when online hate goes offline, it can be deadly”).
Congressional Research Service
4
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
some commentators have questioned the purported link between what some refer to as “hate
speech” and bias-motivated crimes25 or have expressed concern that focusing on the ideological
motivations of speakers has led to calls to criminalize protected speech divorced from any
criminal intent or action.26
While the nature or extent of the relationship between online speech and criminal conduct may be
disputed, the use of the internet by terrorist groups is well documented.27 U.S.-designated terrorist
groups28 such as the Islamic State29 (also known as ISIS or ISIL),30 Al Qaeda,31 Hamas,32 and Al
Shabaab,33 have long used social media to disseminate their ideologies and recruit new members
to their causes. The Islamic State group has used Twitter and YouTube to disseminate videos of its
25 See, e.g., Hate Crimes Hearing, supra note 24 (statement of Candace Owens, Communications Director, Turning
Point USA) (acknowledging “pockets of evil,” but opining that changes in how data are collected and how hate crime is
defined account for purported increases in hate crimes linked to white nationalism).
26 See, e.g., Kim R. Holmes, Commentary, The Origins of “Hate Speech,” THE HERITAGE FOUNDATION (Oct. 22, 2018),
https://www.heritage.org/civil-society/commentary/the-origins-hate-speech (stating, “There are very serious problems
with the concept of hate speech. For one thing, it fails to distinguish between legitimate political content, which is
protected by the Constitution, and explicit intentions to commit violence, which are not.”).
27 See, e.g., U.N. OFFICE ON DRUGS & CRIME, THE USE OF THE INTERNET FOR TERRORIST PURPOSES 3 (2012) (identifying
“six sometimes overlapping categories” for the purpose of classifying “the means by which the Internet is often utilized
to promote and support acts of terrorism,” which are “propaganda (including recruitment, radicalization and incitement
to terrorism); financing; training; planning (including through secret communication and open-source information);
execution; and cyberattacks); FBI, supra note 23 (noting with regard to social media use that “ISIS, in particular,
encourages sympathizers to carry out simple attacks where they are located against targets”); Terrorism and Social
Media Hearing, supra note 15, at 1 (statement of Clint Watts, Robert A. Fox Fellow, Foreign Policy Research Institute)
(stating, “Ten years ago, it was al Qaeda in Iraq videos on YouTube. A few years later, al Shabaab’s deadly rampages
played out on Twitter. Shortly after, Facebook groups and Twitter feeds brought the Islamic State to the world’s
attention and into the homes of new recruits, before they scurried off to other social media applications like
Telegram.”).
28 See generally Foreign Terrorist Organizations, U.S. DEPARTMENT OF STATE, https://www.state.gov/j/ct/rls/other/
des/123085.htm (last visited Apr. 24, 2019); CRS In Focus IF10613, Foreign Terrorist Organization (FTO), by John
W. Rollins.
29 See generally CRS Report R43612, The Islamic State and U.S. Policy, by Christopher M. Blanchard and Carla E.
Humud.
30 See FBI, supra note 23 (“ISIS, in particular, [uses social media to] encourage[] sympathizers to carry out simple
attacks where they are located against targets—in particular, soft targets—or to travel to ISIS-held territory in Iraq and
Syria and join its ranks as foreign fighters.”); Antonia Ward, ISIS’s Use of Social Media Still Poses a Threat to Stability
in the Middle East and Africa, THERANDBLOG (Dec. 11, 2018), https://www.rand.org/blog/2018/12/isiss-use-of-socialmedia-still-poses-a-threat-to-stability.html (“Despite territorial losses and repression of civilian internet access, ISIS
will likely continue to seek to leverage individuals’ increased [information and communication technologies] usage in
Africa and the Middle East to attract new followers.”).
31 See Ward, supra note 30 (“Al Qaeda in the Arabian Peninsula (AQAP) has used social media and online propaganda
for more than a decade, launching its English language digital magazine, which inspired the Boston Marathon bombers,
in 2010.”); Scott Shane, The Lessons of Anwar al-Awlaki, N.Y. TIMES MAGAZINE (Aug. 27, 2015),
http://www.nytimes.com/2015/08/30/magazine/the-lessons-of-anwar-al-awlaki.html (discussing the “digital legacy” of
Anwar al-Awlaki, an American citizen who became closely involved with the Al Qaeda affiliate in Yemen).
32 See Sheera Frenkel & Ben Hubbard, After Social Media Bans, Militant Groups Found Ways to Remain, N.Y. TIMES
(Apr. 19, 2019), https://www.nytimes.com/2019/04/19/technology/terrorist-groups-social-media.html (reporting that
“Hamas and Hezbollah, in particular, have evolved by getting their supporters to publish images and videos that deliver
their message—but that do not set off the alarm bells of the social media platforms”).
33 See Ken Menkhaus, Al-Shabaab and Social Media: A Double-Edged Sword, 20 BROWN J. WORLD AFF. 309, 309
(2014) (“Al-Shabaab . . . has distinguished itself as one of the most sophisticated—and, in more recent times,
conflicted—jihadi users of communication technology. In its early years, it effectively used Internet chat rooms,
websites, and YouTube videos to recruit and fundraise internationally. Its real-time tweets during the group’s terrorist
attack on Nairobi’s Westgate Mall in September 2013 became almost as big a media story as the attack itself.”).
Congressional Research Service
5
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
fighters executing prisoners and to claim credit for attacks around the world.34 Al Shabaab used
Twitter to claim credit for the 2013 attack on the Westgate Shopping Mall in Nairobi, Kenya, and
to distribute information about the attack while it unfolded.35 News outlets are also beginning to
examine how the alleged perpetrator of the March 2019 terrorist attacks on two mosques in New
Zealand may have used social media to announce and promote his actions.36
In addition, the internet reportedly has played a key role in certain individuals’ personal
“journey[s] to terrorism” or violent extremism37—a process often referred to as “radicalization.”38
For example, the person convicted of killing nine black parishioners in a South Carolina church in
2015 was said to have “self-radicalized” online, adopting a “white supremacy extremist ideology,
including a belief in the need to use violence to achieve white supremacy.”39 In 2015, it was
34 Rick Gladstone & Vindu Goel, ISIS Is Adept on Twitter, Study Finds, N.Y. TIMES (Mar. 5, 2015),
https://www.nytimes.com/2015/03/06/world/middleeast/isis-is-skilled-on-twitter-using-thousands-of-accounts-studysays.html; Mark Townsend & Toby Helm, Jihad in a Social Media Age: How Can the West Win an Online War, THE
GUARDIAN (Aug. 23, 2014), https://www.theguardian.com/world/2014/aug/23/jihad-social-media-age-west-win-onlinewar; see also Greg Myre & Camila Domonoske, What Does It Mean When ISIS Claims Responsibility for an Attack?,
NPR (May 24, 2017), https://www.npr.org/sections/thetwo-way/2017/05/24/529685951/what-does-it-mean-when-isisclaims-responsibility-for-an-attack (explaining that one way ISIS claims credit for attacks is through its news agency
posting stories on social media and the agency’s website that tie an attack to ISIS).
35 Peter Bergen, Are Mass Murderers Using Twitter as a Tool?, CNN (Sept. 27, 2013, 5:28 PM), https://www.cnn.com/
2013/09/26/opinion/bergen-twitter-terrorism/index.html; Mark Gollom, Kenya Attack: Why Al-Shabaab Live-Tweeted
the Assault, CBC/RADIO-CANADA (Sept. 24, 2013), https://www.cbc.ca/news/world/kenya-attack-why-al-shabaab-livetweeted-the-assault-1.1865566.
36 E.g., Terrorism in New Zealand: White Supremacy, Gun Laws and the Role of Social Media, THE INTELLIGENCE
(Mar. 18, 2019), https://www.economist.com/podcasts/2019/03/18/terrorism-in-new-zealand-white-supremacy-gunlaws-and-the-role-of-social-media; Ian Bogost, Social Media Are a Mass Shooter’s Best Friend, THE ATLANTIC (Mar.
15, 2019), https://www.theatlantic.com/technology/archive/2019/03/how-terrorism-new-zealand-spread-socialmedia/585040/.
37 This report refers to terrorism and violent extremism as forces associated with the broader public policy debate over
the use of social media. Because terrorism and violent extremism can take many forms, this report does not adopt a
single definition of those terms. See generally 18 U.S.C. § 2332b(g)(5) (defining a federal crime of terrorism in relation
to a host of underlying criminal offenses); U.N. Office on Drugs & Crime, University Module Series on CounterTerrorism: “Radicalization” and “Violent Extremism,” THE DOHA DECLARATION: PROMOTING A CULTURE OF
LAWFULNESS (July 2018), https://www.unodc.org/e4j/en/terrorism/module-2/key-issues/radicalization-violentextremism.html (“As with the concept of ‘terrorism,’ there is no universally agreed definition of the term ‘violent
extremism’ . . . . There are, however, a number of definitions which have been developed at the national, regional and
international levels.”).
38 See Mark Townsend & Ian Traynor, Norway Attacks: How Far Right Views Created Anders Behring Breivik, THE
GUARDIAN (July 30, 2011), https://www.theguardian.com/world/2011/jul/30/norway-attacks-anders-behring-breivik
(“The fact that [the perpetrator of the July 22, 2011 attacks in Norway] chose the internet to disseminate his ideology is
important. His journey to terrorism was forged within a network of blogs where violence is glorified and
multiculturalism despised, along with those who embrace it.”); Rich Lord, How Robert Bowers Went from
Conservative to White Nationalist, PITTSBURGH POST-GAZETTE (Nov. 10, 2018), https://www.post-gazette.com/news/
crime-courts/2018/11/10/Robert-Bowers-extremism-Tree-of-Life-massacre-shooting-pittsburgh-Gab-Warroom/stories/
201811080165 (stating, with regard to the person accused of killing 11 worshipers at a Pennsylvania synagogue in
2018, that accounts from his “coworkers of two decades ago, and an analysis of his social media posts in the weeks
prior to the massacre, suggest that staunch conservatism metastasized into white nationalism,” and suggesting that the
internet may have contributed to this progression); see also United States v. Wright, 285 F. Supp. 3d 443, 448 (D.
Mass. 2018) (remarking that “[f]or the first time I, as a presiding officer, as a citizen, came to understand what it meant
to be ‘radicalized.’ A couple of clicks on the computer, the dark web, and all this material is there” and surmising that
there are “going to be other young men, restless, on the web—ISIS will be gone—looking for a cause” (quoting Court’s
Remarks, Sentencing Hr’g Tr. at 53:4-55:5, United States v. Wright, No. 15-cr-10153 (D. Mass. Dec. 19, 2017), ECF
No. 417)).
39 Mark Berman, Prosecutors Say Dylann Roof “Self-Radicalized” Online, Wrote Another Manifesto in Jail, WASH.
POST (Aug. 22, 2016), https://www.washingtonpost.com/news/post-nation/wp/2016/08/22/prosecutors-say-accused-
Congressional Research Service
6
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
reported that “the digital legacy” of Anwar al Awlaki—a U.S. citizen who became closely
involved with Al Qaeda’s affiliate in Yemen and was targeted and killed by a U.S. drone strike
there—influenced the ideologies of certain individuals accused or convicted of terrorist activities,
including the Boston Marathon bombers.40
Speech advocating violence and terrorism is prohibited by the terms of service of Facebook,
Twitter, and certain other social media outlets.41 Such prohibitions are permissible under current
judicial interpretations of First Amendment law because these platforms are operated by private
actors, and the First Amendment constrains only state (i.e., government) action.42 Although the
more established sites reportedly have increased their efforts to disable accounts that are
associated with terrorist groups or remove content promoting terrorism or violence,43 it is not
charleston-church-gunman-self-radicalized-online/ (internal quotation marks omitted) (quoting from court filings by the
prosecution); see also Rachel Kaadzi Ghansah, A Most American Terrorist: The Making of Dylann Roof, GQ (Aug. 21,
2017), https://www.gq.com/story/dylann-roof-making-of-an-american-terrorist (suggesting that the internet and
message boards that this individual frequented played a role in his indoctrination).
40 Shane, supra note 31. Awlaki left behind numerous videos on YouTube. Id. According to a news report summarizing
the range of topics covered in these videos, while some of Awlaki’s videos expounded upon topics such as respect for
the holy month of Ramadan or the relationship between Islam and Jesus Christ, other videos depicted Awlaki exhorting
his followers never to trust a non-Muslim; that the United States is waging war on Islam; and, in a video entitled “Call
to Jihad,” that Muslims have a religious duty to kill Americans. Id.
41 See, e.g., Community Standards: Violence and Criminal Behavior, FACEBOOK, https://www.facebook.com/
communitystandards/violence_criminal_behavior (last visited Apr. 24, 2019) (stating that Facebook “remove[s] content
that expresses support or praise for groups, leaders, or individuals involved” in, among other things, “terrorist activity”
and asking users not to post, among other categories, “[c]redible statements of intent to commit violence against any
person, groups of people, or place (city or smaller)” or “[c]alls for violence or statements advocating violence against
[certain enumerated] targets (identified by name, title, image, or other reference),” including public individuals, groups
of people, or places, if the threat is credible); The Twitter Rules, TWITTER, https://help.twitter.com/en/rules-andpolicies/twitter-rules (last visited Apr. 25, 2019) (prohibiting, among other things, “specific threats of violence or
wish[ing] for the serious physical harm, death, or disease of an individual or group of people,” including “threatening
or promoting terrorism” as well as some forms of “graphic violence,” and outlining the “enforcement actions” Twitter
may take for failure to abide by the rules, including requiring the user to remove prohibited content in order to create
new posts and temporarily or permanently suspending accounts); Community Guidelines, YOUTUBE,
https://www.youtube.com/yt/about/policies/#community-guidelines (last visited Apr. 25, 2019) (stating, in a section
called “Violent Criminal Organizations” within “Additional policies” that “[c]ontent intended to praise, promote, or aid
violent criminal organizations is not allowed on YouTube,” and asking users not to post, among other things, content
“produced by violent criminal or terrorist organizations,” “praising or justifying violent acts carried out by violent
criminal or terrorist organizations,” or “aimed at recruiting new members to violent criminal or terrorist
organizations”). Such restrictions on user content were not always the norm across social media. See generally
VanLandingham, supra note 15, at 18-19 (“From 2009 through 2015, Twitter stated in its Terms of Service that ‘we do
not actively monitor and will not censor user content, except in limited circumstances described below.’ However, in
2015 Twitter followed Facebook’s lead and suddenly (and dramatically) ratcheted up its policies toward offensive
speech by explicitly banning ‘excessively violent media.’ Additionally, in April 2015, the company also prohibited
‘threatening or promoting terrorism,’ . . . .” (footnotes and citations omitted)).
42 See CRS Report R45650, supra note 19, at 5-9 (discussing the state action requirement and explaining that although
the Supreme Court has not resolved the question, “lower courts have uniformly concluded that the First Amendment
does not prevent social media providers from restricting users’ ability to post content on their networks”).
43 See Terrorism and Social Media Hearing, supra note 15, at 2 (statement of Monika Bickert, Head of Product Policy
and Counterterrorism, Facebook) (“Our proactive efforts—specifically, the use of artificial intelligence (AI) and other
automation—have become increasingly central to keeping [terrorist] content off of Facebook. We currently focus our
most cutting-edge techniques on combating terrorist content about ISIS, Al Qaeda, and their affiliates, and we are
working to expand to other terrorist organizations.”); Ali Breland, Facebook, Twitter and YouTube to Testify on Capitol
Hill About Terrorism and Social Media, THE HILL (Jan. 9, 2018, 5:36 PM), https://thehill.com/policy/technology/
368184-facebook-twitter-and-google-to-testify-on-capitol-hill-about-terrorism-and (reporting that Facebook, Twitter,
and YouTube have acted “to curb terrorists’ use of their platforms”).
Congressional Research Service
7
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
clear how comprehensive or successful these efforts have been.44 Moreover, users banned from
one platform may move to another online forum that does not have the same restrictions,
sometimes finding a community of like-minded individuals who reinforce or escalate their violent
rhetoric (sometimes referred to as “echo chambers”).45
As previously noted, this report focuses on the First Amendment considerations relevant to
government regulation of online content promoting terrorism or violence. Because of the global
reach of many online platforms,46 this report begins with the threshold question of the First
Amendment’s reach, and in particular, whether it applies to foreign nationals posting online
content from outside of the United States.
The First Amendment and Foreign Speakers
While the First Amendment may extend to U.S. citizens speaking abroad or foreign nationals
speaking within the United States under some circumstances,47 the Supreme Court has not
directly opined on whether the First Amendment applies to online content that a foreign national
posts while located outside of the United States.48 Nevertheless, the Court’s decisions involving
44 See Nitasha Tiku, Tech Platforms Treat White Nationalism Different from Islamic Terrorism, WIRED (Mar. 20, 2019,
8:00 AM), https://www.wired.com/story/why-tech-platforms-dont-treat-all-terrorism-same/ (reporting on “concerns
that Big Tech expends more effort to curb the spread of terrorist content from high-profile foreign groups, while
applying fewer resources and less urgency toward terrorist content from white supremacists”); Robinson Meyer, The 3
Questions Mark Zuckerberg Hasn’t Answered, THE ATLANTIC (Apr. 11, 2018), https://www.theatlantic.com/
technology/archive/2018/04/3-questions-mark-zuckerberg-hasnt-answered/557720/ (asking how Facebook’s artificial
intelligence applications could accurately identify terrorist propaganda if they “can’t even flag an ad for a new
apartment,” referring to difficulties Facebook reportedly had in detecting housing advertisements that discriminate by
race); Terrorism and Social Media Hearing, supra note 15, at 3 (statement of Clint Watts, Robert A. Fox Fellow,
Foreign Policy Research Institute) (positing that “[s]ocial media companies continue to get beat in part because they
rely too heavily on technologists and technical detection to catch bad actors”); id. at 3 (statement of Monika Bickert,
Head of Product Policy and Counterterrorism, Facebook) (“When we disable terrorist accounts, those account owners
may try to create new accounts using different identities. We have become faster at using technology to detect new fake
accounts created by repeat offenders, or recidivists.”).
45 See Hate Crimes Hearing, supra note 24, at 11 (statement of Eileen Hershenov, Senior Vice President, Policy, AntiDefamation League) (“[A]s we have recently reported, mainstream platforms can sometimes push such individuals
from an open community . . . into fringe environments . . . that foster acceptability of dangerous views. . . . Individuals
can easily find sanction and reinforcement online for their extreme opinions or actions, in some cases neatly packaged
alongside bomb-making instructions.”).
46 When it comes to the internet, in “actual practice . . . national boundaries are highly permeable.” Yahoo! Inc. v. La
Ligue Contre Le Racisme, 433 F.3d 1199, 1202 (9th Cir. 2006) (en banc) (per curiam); see Facebook, Inc. v. Superior
Court, 417 P.3d 725, 748 (Cal. 2018) (“[W]hen, for example, a Facebook user configures a post as public, that
communication becomes both (a) available to all two billion registered Facebook users, and (b) . . . ‘readily accessible
to the general public’ via . . . search engines. The result is that . . . a public communication is available to ‘everyone in
the world’—even to those who are not registered Facebook users, but who have open access to the Internet.” (internal
citation omitted)).
47 See Haig v. Agee, 453 U.S. 280, 308 (1981) (assuming “arguendo, that First Amendment protections reach beyond
our national boundaries,” in a case involving a U.S. citizen’s conduct and speech abroad); Reid v. Covert, 354 U.S. 1, 5
(1957) (plurality opinion) (rejecting “the idea that when the United States acts against citizens abroad it can do so free
of the Bill of Rights”). Cf. United States v. Verdugo-Urquidez, 494 U.S. 259, 271 (1990) (observing that the Court’s
prior decisions regarding the constitutional rights of non-U.S. citizens “establish only that aliens receive constitutional
protections when they have come within the territory of the United States and developed substantial connections with
the country”).
48 See Hedges v. Obama, 724 F.3d 170, 194 n.140 (2d Cir. 2013) (noting that the “case law regarding extraterritorial
application of constitutional rights is sparse,” and assuming, without deciding, that two non-U.S. citizen plaintiffs
residing abroad could assert First Amendment rights because the “relevant facts . . . were not developed” in the court
below and deciding the question was not necessary to resolving the case).
Congressional Research Service
8
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
the extraterritorial reach of other constitutional protections, as well as lower court decisions
involving the First Amendment rights of foreign nationals, suggest that foreign nationals may face
barriers in claiming First Amendment protections for such speech.
The Supreme Court’s decision in United States v. Verdugo-Urquidez—though it involves the
Fourth Amendment—is instructive.49 In that case, the Court held that the Fourth Amendment,
which “prohibits ‘unreasonable searches and seizures,’” did not extend to the search of a Mexican
citizen’s home in Mexico by U.S. authorities.50 The Court reasoned that in contrast to the Fifth
and Sixth Amendments, which concern trial rights and procedures,51 the Fourth Amendment
applies regardless of the prospect of trial, and “a violation of the Amendment is ‘fully
accomplished’ at the time of an unreasonable governmental intrusion.”52 As such, any violation
would have “occurred solely in Mexico.”53 Four of the five Justices who joined the majority
opinion reasoned that the Fourth Amendment reserves its protections to “the people,” which they
interpreted as a “term of art employed in select parts of the Constitution.”54 In the view of those
Justices, a textual analysis of the Constitution suggested that “‘the people’ protected by the
Fourth Amendment, and by the First and Second Amendments,” meant “a class of persons who
are part of a national community or who have otherwise developed sufficient connection with this
country to be considered part of that community.”55 Because the defendant had “no voluntary
attachment to the United States” at the time of the search, he could not claim the protections of
the Fourth Amendment.56
At least two sitting Supreme Court justices—Justices Clarence Thomas and Brett Kavanaugh—
have suggested that the First Amendment does not apply to foreign nationals abroad, citing to
Verdugo-Urquidez.57 The discussion that more directly addressed the applicability of free speech
protections to foreign nationals came from then-Judge Brett Kavanaugh in a 2014 case involving
49
494 U.S. 259 (1990); see Lamont v. Woods, 948 F.2d 825, 834 (2d Cir. 1991) (“Although Verdugo is a Fourth
Amendment case, it nonetheless provides a helpful analytical framework for determining whether other constitutional
provisions apply to governmental activities having extraterritorial dimensions. Specifically, the Verdugo Court
identified three factors as significant to such a determination: (1) the operation and text of the constitutional provision;
(2) history; and (3) the likely consequences if the provision is construed to restrict the government’s extraterritorial
activities.”).
50 Verdugo-Urquidez, 494 U.S. at 264 (quoting U.S. CONST. amend. IV).
51 See id. The Court previously extended the Fifth and Sixth Amendments—insofar as they require “[t]rial by jury in a
court of law . . . after an indictment by grand jury”—to American citizens detained in other countries. See Reid v.
Covert, 354 U.S. 1, 10, 18-19 (1957) (plurality opinion). In contrast, the Court declined to extend certain Fifth
Amendment protections to “nonresident enemy aliens” detained in other countries. Johnson v. Eisentrager, 339 U.S.
763, 781, 785 (1950)).
52 See Verdugo-Urquidez, 494 U.S. at 264.
53 Id.
54 Id. at 265. Although the Court’s opinion garnered a five-member majority, one member, Justice Kennedy, wrote
separately stating that he could not “place any weight on the reference to ‘the people’ in the Fourth Amendment as a
source of restricting its protections.” Id. at 276 (Kennedy, J., concurring).
55 Id. at 265 (majority opinion). Because the Court did not see its “textual exegesis” as “conclusive,” it also considered
“the history of the drafting of the Fourth Amendment,” which it found to support its conclusion that the Framers did not
intend for the Fourth Amendment to apply abroad. Id. at 266.
56 Id. at 274-75.
57 See Trump v. Hawaii, 138 S. Ct. 2392, 2424 (2018) (Thomas, J., concurring) (stating that apart from their
Establishment Clause claim, “[t]he plaintiffs cannot raise any other First Amendment claim, since the alleged religious
discrimination in this case was directed at aliens abroad” (citing Verdugo-Urquidez, 494 U. S. at 265)); Al Bahlul v.
United States, 767 F.3d 1, 75-76 (D.C. Cir. 2014) (en banc) (Kavanaugh, J., concurring in the judgment in part and
dissenting in part) (reasoning that “non-U.S. citizens have no First Amendment rights abroad in foreign countries”).
Congressional Research Service
9
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
the United States’ prosecution of an Al Qaeda associate.58 In that case, a U.S. military
commission convicted a personal assistant to Osama bin Laden for, among other things,
conspiracy to commit war crimes.59 The defendant60 “claim[ed] that he was unconstitutionally
prosecuted for his political speech, including his production of [an] al Qaeda recruitment video
celebrating the terrorist attack on the U.S.S. Cole.”61 When the case, Al Bahlul v. United States,
first reached the full U.S. Court of Appeals for the District of Columbia Circuit (D.C. Circuit),62
the court ruled on a different legal question and remanded the case for consideration of the First
Amendment challenge.63 However, Judge Kavanaugh authored a separate opinion, in part to
address the First Amendment’s applicability.64 He stated that “although non-U.S. citizens
arguably may have some First Amendment rights at [a U.S. military base in] Guantanamo or in
other U.S. territories for any speech they engage in there, non-U.S. citizens have no First
Amendment rights abroad in foreign countries.”65
The Supreme Court has applied the Constitution to aliens in the United States and in U.S.
territories, but has not extended constitutional rights to aliens in foreign countries. See
Boumediene v. Bush, 553 U.S. 723, 768-71 (2008) (applying Article I, Section 9 to U.S.
Naval base at Guantanamo, which was “[i]n every practical sense . . . not abroad”); United
States v. Verdugo-Urquidez, 494 U.S. 259 (1990) (declining to apply Fourth Amendment
to search and seizure of alien’s property in Mexico); Johnson v. Eisentrager, 339 U.S. 763
(1950) (declining to apply habeas corpus right to U.S.-controlled military prison in
Germany); see also Al Maqaleh v. Hagel, 738 F.3d 312 (D.C. Cir. 2013) (declining to apply
habeas corpus right to U.S. military base in Afghanistan); Al Maqaleh v. Gates, 605 F.3d
84 (D.C. Cir. 2010) (same). Therefore, [the defendant] had no First Amendment rights as
a non-U.S. citizen in Afghanistan when he led bin Laden’s media operation.66
Two years later, the full D.C. Circuit took up the Al Bahlul case again following remand.67 This
time, the court squarely rejected the defendant’s First Amendment challenge, citing the
concurring opinions of Judge Kavanaugh, Judge Patricia Millett, and Judge Robert Wilkins, who
all concluded that the defendant could not avail himself of the First Amendment’s protections.68
58 Al Bahlul v. United States, 767 F.3d 1 (D.C. Cir. 2014) (en banc).
59 Id. at 5.
60 Al Bahlul was tried as an enemy combatant before a military commission. Id. at 6-7. This report uses the term
“defendant” for the sake of internal consistency.
61 Id. at 75-76 (Kavanaugh, J., concurring in the judgment in part and dissenting in part).
62 For purposes of brevity, references to a particular circuit in the body of this report (e.g., the D.C. Circuit) refer to the
U.S. Court of Appeals for that particular circuit.
63 Al Bahlul, 767 F.3d at 31 (en banc).
64 Id. at 75-76 (Kavanaugh, J., concurring in the judgment in part and dissenting in part).
65 Id. at 76. Judge Kavanaugh observed as “an initial matter” that “[the defendant] was convicted of conspiracy based
on his conduct,” not his speech alone. Id.
66 Id. (parallel citations omitted).
67 Al Bahlul v. United States, 840 F.3d 757 (D.C. Cir. 2016) (en banc) (per curiam).
68 Id. at 758-59; see also id. at 770 n.12 (Kavanaugh, J., concurring) (calling the First Amendment challenge “frivolous,
for reasons explained” in his earlier opinion); id. at 797 (Millett, J., concurring) (reasoning, inter alia, that “no
governing precedent extends First Amendment protection to speech undertaken by non-citizens on foreign soil”); id. at
804 (Wilkins, J., concurring) (concurring with the court’s disposition of the First Amendment challenge for the reasons
set forth in Judge Millett’s opinion). Judges Kavanaugh and Millett also reasoned that even if the First Amendment
applied to the defendant’s speech in Afghanistan, “the speech encompassed within the charges against Bahlul—
including a terrorist recruitment video produced on foreign soil that ‘was aimed at inciting viewers to join al Qaeda, to
kill Americans, and to cause destruction’—was not protected speech under the First Amendment.” Al Bahlul, 767 F.3d
at 76 (Kavanaugh, J., concurring in the judgment in part and dissenting in part) (quoting United States v. Al Bahlul,
820 F. Supp. 2d 1141, 1249 (C.M.C.R. 2011) (en banc)); see also Al Bahlul, 840 F.3d at 797 (Millett, J., concurring)
Congressional Research Service
10
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
Thus, as the D.C. Circuit phrased it in a prior decision, “aliens beyond the territorial jurisdiction
of the United States are generally unable to claim the protections of the First Amendment.”69
Nevertheless, “[i]n a variety of contexts th[e] Court has referred to a First Amendment right to
‘receive information and ideas.’”70 In order to preserve this right, the Court has largely rejected
governmental attempts to control information because of how the government views that
information.71 For example, in Lamont v. Postmaster General, the Court held that a federal statute
requiring the Postal Service to withhold foreign mailings classified as “communist political
propaganda” from addressees unless they requested delivery of the mailings in writing amounted
“to an unconstitutional abridgment of the addressee’s First Amendment rights.”72 The Court
concluded that the “regime of this Act is at war with the ‘uninhibited, robust, and wide-open’
debate and discussion that are contemplated by the First Amendment.”73
Whether or when the government must observe the First Amendment in its interactions with
foreign nationals in order to preserve the rights of U.S. citizens is uncertain. As then–D.C. Circuit
Judge Ruth Bader Ginsburg noted in a dissenting opinion, “[t]he [F]irst [A]mendment secures to
persons in the United States the respect of our government for their right to communicate and
associate with foreign individuals and organizations.”74 Referring to Lamont, she observed that
“the first federal law the Supreme Court ever held violative of the [F]irst [A]mendment involved
a condition on international correspondence—a restraint on delivery of mail from abroad.”75
While finding it unnecessary to decide in that case “whether the [F]irst [A]mendment limits the
actions of U.S. officials in their dealings with foreign parties,” Judge Ginsburg noted the
following principle from the Restatement (Third) of Foreign Relations:
The provisions of the United States Constitution safeguarding individual rights generally
control the United States government in the conduct of its foreign relations as well as in
domestic matters, and generally limit governmental authority whether it is exercised in the
(“What is settled, moreover, is that the First Amendment offers no shield to speech like Bahlul’s that is ‘directed to
inciting or producing imminent lawless action and . . . [is] likely to incite or produce such action.’” (citations omitted)).
69 DKT Mem’l Fund v. Agency for Int’l Dev., 887 F.2d 275, 284 (D.C. Cir. 1989). But cf. RESTATEMENT (THIRD) OF
FOREIGN RELATIONS LAW OF THE UNITED STATES § 722 cmt. m (1987) (“Although the matter has not been
authoritatively adjudicated, at least some actions by the United States in respect of foreign nationals outside the country
are also subject to constitutional limitations. Thus, trial of an alien under United States authority has been held subject
to constitutional safeguards. Similarly, the taking by United States authorities abroad for public use of property of a
foreign national may give the former owner a right to just compensation under the Fifth Amendment.” (internal citation
omitted)).
70 Kleindienst v. Mandel, 408 U.S. 753, 762 (1972).
71 See Riley v. Nat’l Fed’n of the Blind, 487 U.S. 781, 791 (1988) (“‘The very purpose of the First Amendment is to
foreclose public authority from assuming a guardianship of the public mind through regulating the press, speech, and
religion.’ To this end, the government, even with the purest of motives, may not substitute its judgment as to how best
to speak for that of speakers and listeners; free and robust debate cannot thrive if directed by the government.” (internal
citation omitted) (quoting Thomas v. Collins, 323 U.S. 516, 545 (1945) (Jackson, J., concurring))).
72 381 U.S. 301, 307 (1965).
73 Id. (quoting N.Y. Times Co. v. Sullivan, 376 U.S. 254, 270 (1964)); see also United States v. Playboy Entm’t Grp.,
529 U.S. 803, 813 (2000) (“Our precedents teach these principles. Where the designed benefit of a content-based
speech restriction is to shield the sensibilities of listeners, the general rule is that the right of expression prevails, even
where no less restrictive alternative exists. We are expected to protect our own sensibilities ‘simply by averting [our]
eyes.’” (citation omitted)); Va. State Bd. of Pharmacy v. Va. Citizens Consumer Council, 425 U.S. 748, 765 (1976)
(“[E]ven if the First Amendment were thought to be primarily an instrument to enlighten public decisionmaking in a
democracy, we could not say that the free flow of information does not serve that goal.”).
74 DKT Mem’l Fund v. Agency for Int’l Dev., 887 F.2d 275, 303 (D.C. Cir. 1989) (Ginsburg, J., concurring in part and
dissenting in part).
75 Id.
Congressional Research Service
11
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
United States or abroad, and whether such authority is exercised unilaterally or by
international agreement.76
Judge Ginsburg concluded by stating that she “would hesitate long before holding that in a United
States-foreign citizen encounter, the amendment we prize as ‘first’ has no force in court.”77
Based on these principles and decisions, there may be some cases involving social media content
posted by foreign nationals that implicate the free speech rights of U.S. citizens, who may be
members of the particular online forum or otherwise have access to it.78 However, there could be
prudential limitations on a foreign national’s ability to assert those rights.79 And, even if a court
concludes that a particular foreign national has standing in a given case, it is not clear that the
First Amendment would protect his or her online activities based solely on the purported interests
of other Internet users.80
In contrast, the Supreme Court has recognized that U.S. citizens regularly exercise First
Amendment rights when communicating online, so free speech protections are more directly in
play when considering the United States’ regulation of its own citizens’ online speech.81
Accordingly, the remainder of this report discusses the principles that bear on the government’s
ability to regulate online content promoting terrorism or violence when there is no dispute about
the First Amendment’s applicability.
76 Id. at 307-08 (quoting RESTATEMENT (THIRD) OF FOREIGN RELATIONS LAW OF THE UNITED STATES § 721 (1987)).
77 Id. at 308.
78 The U.S. Court of Military Commission Review in United States v. Al Bahlul specifically considered and rejected the
defendant’s argument that his prosecution unconstitutionally “chill[ed] the dissemination of information available to
U.S. citizens.” 820 F. Supp. 2d 1141, 1242, 1250 (C.M.C.R. 2011) (en banc) (“[The defendant’s] prosecution does not
adversely affect the rights of U.S. citizens to receive such information. The Video is readily available on the Internet
and in numerous foreign languages. Possession or viewing of The Video is not criminalized . . . .”), vacated in part on
other grounds, 767 F.3d 1, 31 (D.C. Cir. 2014).
79 See Valley Forge Christian Coll. v. Ams. United for Separation of Church & State, 454 U.S. 464, 474 (1982)
(“Beyond the constitutional requirements, the federal judiciary has also adhered to a set of prudential principles that
bear on the question of standing. Thus, this Court has held that ‘the plaintiff generally must assert his own legal rights
and interests, and cannot rest his claim to relief on the legal rights or interests of third parties.’” (quoting Warth v.
Seldin, 422 U.S. 490, 499 (1975))); e.g., DKT Mem’l Fund., 887 F.2d at 283-85 (stating that “[b]efore we can decide
that [an exercise of governmental authority] violates constitutional rights, we must find that a plaintiff with standing to
assert the violation of rights is properly before the court” and affirming the district court’s decision that foreign
nongovernmental organizations did not have standing to challenge the U.S. government’s policy on First Amendment
grounds). But see Fallon, supra note 22, at 863 (“Against the background of the ordinary rule that no one can challenge
a statute on the ground that it would be unconstitutional as applied to someone else, a First Amendment exception has
emerged. When speech or expressive activity forms a significant part of a law’s target, the law is subject to facial
challenge and invalidation if: (i) it is ‘substantially overbroad’—that is, if its illegitimate applications are too numerous
‘judged in relation to the statute’s plainly legitimate sweep,’ and (ii) no constitutionally adequate narrowing
construction suggests itself.” (footnote omitted)).
80 Cf. Chevron Corp. v. Donziger, 325 F. Supp. 3d 371, 386 n.52 (S.D.N.Y. 2018) (stating that the defendant’s “First
Amendment argument . . . appears in significant part to be an attempt to assert First Amendment rights on the behalf of
non-citizens living outside the United States and so would fail on the merits even if he had standing”), appeal filed, No.
18-2191 (2d Cir. 2018). But see Artem M. Joukov & Samantha M. Caspar, Comrades or Foes: Did the Russians Break
the Law or New Ground for the First Amendment?, 39 PACE L. REV. 44, 70-76 (2018) (arguing that the First
Amendment should apply to foreign nationals’ online speech in prosecutions involving U.S. laws).
81 See Packingham v. North Carolina, 137 S. Ct. 1730, 1735-36 (2017) (“[S]ocial media users employ these websites to
engage in a wide array of protected First Amendment activity on topics ‘as diverse as human thought.’” (quoting Reno
v. ACLU, 521 U.S. 844, 870 (1997))).
Congressional Research Service
12
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
The Conduct Versus Speech Distinction
A key initial consideration in evaluating whether a law or a particular application of that law
comports with the First Amendment is whether the law at issue regulates conduct or speech. The
distinction is sometimes elusive because speech may occur during a course of conduct, and
actions themselves can sometimes be inherently expressive or “symbolic” speech protected by the
First Amendment.82 As a potential starting point, a law typically regulates conduct if it dictates
what the regulated persons or entities “must [or must not] do . . . not what they may or may not
say.”83 But the touchstone for deciding whether such a law implicates the First Amendment
appears to be whether the law targets expression.84 To determine whether a law targets expression
and depending on the facts of the case, a court might consider (1) the express terms (i.e., “the
face”) of the law, (2) the purpose of the law, or (3) its practical application to see whether the law
is directed at certain content or speakers or applies to the challenger’s activities solely as a result
of what that party seeks to communicate.85
If a law does not target expression, the First Amendment extends the government more leeway to
regulate that activity even if the regulation incidentally burdens speech. As the Supreme Court
has explained,
[R]estrictions on protected expression are distinct from restrictions on economic activity
or, more generally, on nonexpressive conduct. . . . [T]he First Amendment does not prevent
restrictions directed at commerce or conduct from imposing incidental burdens on speech.
That is why . . . “an ordinance against outdoor fires” might forbid “burning a flag” . . . .86
82 See Nat’l Inst. of Family & Life Advocates v. Becerra, 138 S. Ct. 2361, 2373 (2018) (noting that “[w]hile drawing
the line between speech and conduct can be difficult, this Court’s precedents have long drawn it”); Rumsfeld v. Forum
for Acad. & Institutional Rights, Inc., 547 U.S. 47, 65 (2006) (stating that the First Amendment’s protections for
“symbolic speech” concern conduct that is “inherently expressive”); e.g., Texas v. Johnson, 491 U.S. 397, 406 (1989)
(holding that when the defendant burned a flag in public protest, the “burning of the flag was conduct ‘sufficiently
imbued with elements of communication,’ to implicate the First Amendment” (internal citation omitted)).
83 See Rumsfeld, 547 U.S. at 60.
84 See Arcara v. Cloud Books, Inc., 478 U.S. 697, 703 (1986) (noting that the Court has not subjected all criminal and
civil penalties that affect the defendant’s First Amendment activities to scrutiny, “only [those] where it was conduct
with a significant expressive element that drew the legal remedy in the first place . . . or where a statute based on a
nonexpressive activity has the inevitable effect of singling out those engaged in expressive activity”).
85 See Sorrell v. IMS Health Inc., 564 U.S. 552, 557, 567 (2011) (reasoning that a state law prohibiting pharmaceutical
manufacturers from using certain pharmacy records for marketing purposes “[b]oth on its face and in its practical
operation . . . impose[d] a burden based on the content of speech and the identity of the speaker”); Minneapolis Star &
Tribune Co. v. Minn. Comm’r of Revenue, 460 U.S. 575-76, 581 (1983) (reasoning that although the government can
impose “generally applicable economic regulations” on the press, the state’s special use tax on ink and paper was not of
the same character because it singled out certain publications and subjected them to the tax). See generally CRS Report
R45700, Assessing Commercial Disclosure Requirements under the First Amendment 6-10, by Valerie C. Brannon
(discussing the conduct-speech distinction in the commercial disclosure context).
86 Sorrell, 564 U.S. at 567 (citation omitted). Even when the government seeks to regulate conduct that has an
expressive component, the Court has held that a “sufficiently important governmental interest in regulating the
nonspeech element can justify incidental limitations on First Amendment freedoms.” United States v. O’Brien, 391
U.S. 367, 376 (1968). In general, a government regulation of this nature “is sufficiently justified if it is within the
constitutional power of the Government; if it furthers an important or substantial governmental interest; if the
governmental interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First
Amendment freedoms is no greater than is essential to the furtherance of that interest.” Id. at 377. But see Texas v.
Johnson, 491 U.S. at 406-07 (holding that “O’Brien’s relatively lenient standard” is limited “to those cases in which the
governmental interest is unrelated to the suppression of free expression,” not situations in which a law is “directed at
the communicative nature of conduct,” which must “be justified by the substantial showing of need that the First
Amendment requires” (internal quotation marks and citations omitted)).
Congressional Research Service
13
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
However, the conduct-focused nature of a law does not necessarily preclude First Amendment
review where the government seeks to penalize a person under that law because of ideas or
messages that person communicated.87
The Supreme Court applied these principles in its 2010 decision in Holder v. Humanitarian Law
Project, which involved the constitutionality of a federal statute concerning the provision of
material support to U.S.-designated foreign terrorist organizations (FTOs).88 The statute imposes
criminal penalties on anyone who “knowingly provides material support or resources to [an
FTO], or attempts or conspires to do so.”89 It defines “material support or resources” in relevant
part as “any property, tangible or intangible, or service, including . . . training, expert advice or
assistance, . . . false documentation or identification, communications equipment, . . . personnel
(1 or more individuals who may be or include oneself) . . . , except medicine or religious
materials.”90 In Humanitarian Law Project, a group of U.S. citizens and domestic organizations
brought a preenforcement challenge to the law, arguing that it would be unconstitutional to punish
them for the types of support that they wished to provide to two FTOs.91 Specifically, the
plaintiffs sought to (1) train members of one FTO on how to use humanitarian and international
law in peaceful dispute resolution; (2) teach that FTO’s members how to petition international
organizations for relief; and (3) engage in political advocacy for the rights of certain groups,
including by supporting one of the FTOs “as a political organization” for this purpose.92 After
concluding that most of these activities clearly constituted “training” or “expert advice or
assistance” under the law,93 the Court proceeded to address the First Amendment implications of
applying the statute to the plaintiffs’ activities.94
The Humanitarian Law Project Court rejected the “extreme positions” advanced by both sides.95
On the one hand, it rejected the plaintiffs’ contention that the statute banned their “pure political
87 For example, in Cohen v. California, the defendant was convicted under a law that prohibited disturbing the peace by
“offensive conduct” because he wore a jacket that displayed a “four-letter word . . . in relation to the draft” in a
municipal courthouse. 403 U.S. 15-16, 20 (1971). The Supreme Court held that his conviction was based “solely upon
‘speech’” because it “quite clearly rest[ed] upon the asserted offensiveness of the words [the defendant] used to convey
his message to the public” and the “only ‘conduct’ which the State sought to punish [was] the fact of communication.”
Id. at 18.
88 561 U.S. 1 (2010).
89 18 U.S.C. § 2339B(a)(1). A companion statute—§ 2339A—makes it a criminal offense to “provide[] material
support or resources or conceal[] or disguise[] the nature, location, source, or ownership of material support or
resources, knowing or intending that they are to be used in preparation for, or in carrying out, a violation of [certain
criminal offenses] . . . or in preparation for, or in carrying out, the concealment of an escape from the commission of
any such violation,” and also criminalizes attempt and conspiracy to commit such acts. The two statutes share the same
definition of “material support or resources.” See id. § 2339B(g)(4) (defining “material support or resources” by
reference to § 2339A).
90 Id. § 2339B(g)(4); see also id. § 2339A(b)(1).
91 Humanitarian Law Project, 561 U.S. at 9-10, 14.
92 Id. at 14-15 (citation omitted).
93 Id. at 21. The Court made this determination as part of its analysis as to whether terms such as “training” and “expert
advice or assistance” were unconstitutionally vague. In this regard, the Court concluded that while “the scope of the
material-support statute may not be clear in every application,” the “dispositive point here is that the statutory terms are
clear in their application to plaintiffs’ proposed conduct, which means that plaintiffs’ vagueness challenge must fail.”
Id. The Court further concluded that to the extent the plaintiffs sought to engage in independent advocacy, their
activities clearly would not be prohibited as the provision of “personnel,” which, as defined, is limited to work under
the FTO’s “direction or control,” or “service[s],” which the Court interpreted to refer to “concerted activity.” Id. at 2324 (citations omitted).
94 Id. at 25.
95 Id.
Congressional Research Service
14
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
speech,” noting that the law did not prevent the plaintiffs from becoming members of the FTOs,
speaking and writing freely about these organizations, or engaging in independent advocacy.96
The Court reasoned that “Congress has prohibited ‘material support,’ which most often does not
take the form of speech at all. And when it does, the statute is carefully drawn to cover only a
narrow category of speech to, under the direction of, or in coordination with foreign groups that
the speaker knows to be terrorist organizations.”97 On the other hand, the Court rejected the
government’s position that applying the law to the plaintiffs’ activities implicated only conduct,
not speech.98 It reasoned that the law itself “regulates speech on the basis of its content” because
whether plaintiffs are subject to prosecution “depends on what they say.”99 Referring to the
statutory definitions of “training” and “expert advice or assistance,” the Court noted that if the
plaintiffs’ speech to the organizations “imparts a ‘specific skill’ or communicates advice derived
from ‘specialized knowledge’—for example, training on the use of international law or advice on
petitioning the United Nations—then it is barred.”100 Even if the law “generally functions as a
regulation of conduct,” the Court reasoned, in these circumstances, “as applied to plaintiffs the
conduct triggering coverage under the statute consists of communicating a message.”101 As
discussed in more detail infra,102 the Court ultimately held that the material support statute did not
violate the First Amendment as applied to the plaintiffs’ proposed activities because the
challenged statutory prohibitions were necessary to further the government’s asserted interests in
combating terrorism.103
Unlike a law involving “material support,” which, in the Court’s view, “most often” takes the
form of conduct rather than speech,104 a law that expressly prohibits or restricts, for example,
social media posts promoting terrorism or violence would more clearly involve speech because of
its central focus on communications. Although the Supreme Court has not had many occasions to
consider laws that expressly restrict online content, in a First Amendment challenge to a federal
law that restricted the online transmission of certain “indecent” and “patently offensive” content,
both the parties and the Court evaluated the law as regulating speech, not conduct.105
Content-Based Laws
Once it is established that a law regulates speech, the next consideration is whether it does so on
the basis of content. First Amendment law historically has distinguished between laws that restrict
96 Id. at 25-26 (citation omitted).
97 Id. at 26.
98 Id.
99 Id. at 27.
100 Id.
101 Id. at 27-28.
102 See infra section entitled “Deference to Political Branches.”
103 Humanitarian Law Project, 561 U.S. at 28-39.
104 Id. at 26.
105 See Reno v. ACLU, 521 U.S. 844, 849, 867-68 (1997) (evaluating two provisions of the Communications Decency
Act “enacted to protect minors from ‘indecent’ and ‘patently offensive’ communications on the Internet” as restrictions
on speech); cf. United States v. Stevens, 559 U.S. 460, 464-65, 468 (2010) (analyzing a federal statute banning the
commercial creation, sale, and possession of certain depictions of animal cruelty, which was enacted to address the
“interstate market for ‘crush videos’ . . . featur[ing] the intentional torture and killing of helpless animals,” as a
regulation of speech); United States v. Williams, 553 U.S. 285, 288, 293 (2008) (analyzing a federal law that
“criminalizes, in certain specified circumstances, the pandering or solicitation of child pornography” as a law
prohibiting “the collateral speech that introduces such material into the child-pornography distribution network”).
Congressional Research Service
15
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
speech because of its content or viewpoint and those that do not draw content-based distinctions
or that have a content-neutral justification.106 Although the Justices of the Court have sometimes
disagreed over whether a particular law is content-based for First Amendment purposes,107 the
Court has largely settled on the following definition:
Government regulation of speech is content based if a law applies to particular speech
because of the topic discussed or the idea or message expressed. This commonsense
meaning of the phrase “content based” requires a court to consider whether a regulation of
speech “on its face” draws distinctions based on the message a speaker conveys. Some
facial distinctions based on a message are obvious, defining regulated speech by particular
subject matter, and others are more subtle, defining regulated speech by its function or
purpose.108
Courts scrutinize content-based distinctions because of the potential for the government to silence
speech with which it disagrees by prohibiting or imposing special burdens on an entire category
of speech: “content discrimination ‘raises the specter that the Government may effectively drive
certain ideas or viewpoints from the marketplace.’”109 Justice Anthony Kennedy wrote in a case
involving federal restrictions on the transmission of sexually explicit cable programming: “The
history of the law of free expression is one of vindication in cases involving speech that many
citizens may find shabby, offensive, or even ugly. It follows that all content-based restrictions on
speech must give us more than a moment’s pause.”110 The Supreme Court has repeatedly stated
that content-based laws are “presumptively unconstitutional” and subject to the Court’s most
stringent review, at least insofar as they involve fully protected speech.111 In current First
106 See Nat’l Inst. of Family & Life Advocates v. Becerra, 138 S. Ct. 2361, 2371 (2018) (“[O]ur precedents distinguish
between content-based and content-neutral regulations of speech.”); Genevieve Lakier, Reed v. Town of Gilbert,
Arizona, and the Rise of the Anticlassificatory First Amendment, 2016 SUP. CT. REV. 233, 233-234 (2016) (“For
decades now, the Supreme Court has insisted that content-based laws—laws that restrict speech because of its ideas or
messages or subject matter—are presumptively unconstitutional, and will be sustained only if they can satisfy strict
scrutiny. In contrast, content-neutral laws—laws that regulate speech for some reason other than its content—are
reviewed under a lesser, and often quite deferential, standard. Whether a law is found to be content-based or contentneutral therefore determines, in many cases, whether a First Amendment challenge to it succeeds.” (footnotes
omitted)).
107 Compare Renton v. Playtime Theatres, 475 U.S. 41, 47-48 (1986) (holding that zoning ordinance restricting the
locations of “adult” movie theatres was “aimed not at the content of the films shown [there], but rather at the secondary
effects of such theaters on the surrounding community,” and thus was “completely consistent with [the Court’s]
definition of ‘content-neutral’ speech regulations as those that ‘are justified without reference to the content of the
regulated speech’” (citations omitted)), with id. at 56-57 (Brennan, J., dissenting) (arguing that the “fact that adult
movie theaters may cause harmful ‘secondary’ land-use effects may arguably give [the city] a compelling reason to
regulate such establishments; it does not mean, however, that such regulations are content neutral” and stating that
“[b]ecause the ordinance imposes special restrictions on certain kinds of speech on the basis of content, I cannot simply
accept, as the Court does, [the city’s] claim that the ordinance was not designed to suppress the content of adult
movies”).
108 Reed v. Town of Gilbert, 135 S. Ct. 2218, 2227 (2015) (internal citations omitted).
109 R.A.V. v. City of St. Paul, 505 U.S. 377, 387 (1992) (quoting Simon & Schuster, Inc. v. Members of N.Y. State
Crime Victims Bd., 502 U.S. 105, 116 (1991)).
110 United States v. Playboy Entm’t Grp., 529 U.S. 803, 826 (2000).
111 Reed, 135 S. Ct. at 2226. But cf. id. at 2237-38 (Kagan, J., concurring in the judgment) (explaining that the Court
applies “strict scrutiny to facially content-based regulations of speech . . . when there is any ‘realistic possibility that
official suppression of ideas is afoot’” but arguing that “when that is not realistically possible, we may do well to relax
our guard so that ‘entirely reasonable’ laws imperiled by strict scrutiny can survive”). See also Playboy Entm’t Grp.,
529 U.S. at 813 (“Since § 505 is a content-based speech restriction, it can stand only if it satisfies strict scrutiny. If a
statute regulates speech based on its content, it must be narrowly tailored to promote a compelling Government interest.
If a less restrictive alternative would serve the Government’s purpose, the legislature must use that alternative.”
(internal citations omitted)); R.A.V., 505 U.S. at 382 (“Content-based regulations are presumptively invalid.”).
Congressional Research Service
16
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
Amendment parlance, such laws must survive “strict scrutiny,” meaning that the government
must demonstrate that they are narrowly tailored to serve compelling governmental interests. 112
Under the Court’s current formulation, a law that expressly regulates what topics can be
discussed in a social media post would likely be considered a “content-based” restriction on
speech, as the regulation “applies to particular speech because of the topic discussed or the idea
or message expressed.”113 As such, as a general matter, such a law would likely be subject to strict
scrutiny and presumptively invalid.114 However, this level of scrutiny may not apply if the
regulated content falls within a category of speech that the Court has said is not fully protected, as
discussed in the next section.115
Protected and Unprotected Speech
If a law regulates speech, the next consideration in the First Amendment analysis is whether that
speech is considered protected (sometimes referred to as “fully protected”116) or instead falls
within one of the narrow categories of so-called “unprotected” speech (sometimes referred to as
the First Amendment’s “exceptions”117) recognized by the Supreme Court.118 Such categories are
not determinative of whether a law is constitutional, but the government generally has greater
leeway to regulate unprotected speech based on its content.119
Cf. Sorrell v. IMS Health Inc., 564 U.S. 552, 565 (2011) (applying “heightened judicial scrutiny” to a content-based
regulation involving commercial speech).
112 See, e.g., Reed, 135 S. Ct. at 2231 (majority opinion) (stating that a town had the “burden to demonstrate that [its
code’s] differentiation between temporary directional signs and other types of signs, such as political signs and
ideological signs, furthers a compelling governmental interest and is narrowly tailored to that end”); Simon & Schuster,
Inc., 502 U.S. at 118 (holding that because the state’s “Son of Sam law” established “a financial disincentive to create
or publish works with a particular content” (i.e., works describing an author’s crime), “[i]n order to justify such
differential treatment, the State must show that its regulation is necessary to serve a compelling state interest and is
narrowly drawn to achieve that end” (internal quotation marks and citation omitted)). Cf. Buckley v. Valeo, 424 U.S. 1,
44-45 (1976) (per curiam) (reasoning that the constitutionality of the statute’s limitation on political expenditures
“turns on whether the governmental interests advanced in its support satisfy the exacting scrutiny applicable to
limitations on core First Amendment rights of political expression”).
113 Reed, 135 S. Ct. at 2227. See, e.g., United States v. Stevens, 559 U.S. 460, 468 (2010) (“Section 48 explicitly
regulates expression based on content: The statute restricts ‘visual [and] auditory depiction[s],’ such as photographs,
videos, or sound recordings, depending on whether they depict conduct in which a living animal is intentionally
harmed.”). Setting aside for a moment whether they are ultimately constitutional, laws that prohibit advocacy of a
certain type would likely be considered content-based. See Wilson R. Huhn, Assessing the Constitutionality of Laws
That Are Both Content-Based and Content-Neutral: The Emerging Constitutional Calculus, 79 IND. L.J. 801, 806
(2004) (stating that “a law that makes it illegal to advocate the violent overthrow of the government is purely contentbased”).
114 Cf. Reed, 135 S. Ct. at 2226; Playboy Entm’t Grp., 529 U.S. at 813, 818.
115 See R.A.V., 505 U.S. at 387 (“Even the prohibition against content discrimination that we assert the First
Amendment requires is not absolute. It applies differently in the context of proscribable speech than in the area of fully
protected speech.”).
116 E.g., R.A.V., 505 U.S. at 387.
117 E.g., Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786, 792 (2011) (referring to “the obscenity exception to the First
Amendment”). But see JESSE H. CHOPER ET AL., CONSTITUTIONAL LAW: CASES, COMMENTS, AND QUESTIONS 788
(12th ed. 2015) (observing that categories of unprotected speech “are often described as ‘exceptions’ to the First
Amendment,” but “once we realize just how much human communication remains untouched by the First
Amendment—contract law, the law of wills, prosecution for perjury and blackmail, and much else—the language of
‘exceptions’ seems a bit misleading”).
118 See Stevens, 559 U.S. at 470 (discussing the “historically unprotected categories of speech”).
119 The Supreme Court has explained that the First Amendment imposes some limitations on the government even
Congressional Research Service
17
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
The Supreme Court has long considered political and ideological speech to be at the “core” of the
First Amendment and the ability to exchange ideas to be integral to a functioning democracy.120
Our cases have often noted the close connection between our Nation’s commitment to selfgovernment and the rights protected by the First Amendment. . . .
The First Amendment creates “an open marketplace” in which differing ideas about
political, economic, and social issues can compete freely for public acceptance without
improper government interference. The government may not prohibit the dissemination of
ideas that it disfavors, nor compel the endorsement of ideas that it approves.121
These principles extend even to speech that many would consider to be deeply offensive or
hateful.122 Accordingly, a law that restricts speech concerning “politics, nationalism, religion, or
other matters of opinion”123 generally receives strict scrutiny.124
But the First Amendment does not just protect core political and ideological speech. Even in cases
involving speech historically considered to have lower “social value,”125 the government
when it is regulating unprotected speech.
We have sometimes said that [certain] categories of expression are “not within the area of
constitutionally protected speech,” or that the “protection of the First Amendment does not extend”
to them. Such statements must be taken in context, however . . . . What they mean is that these areas
of speech can, consistently with the First Amendment, be regulated because of their constitutionally
proscribable content (obscenity, defamation, etc.)—not that they are categories of speech entirely
invisible to the Constitution, so that they may be made the vehicles for content discrimination
unrelated to their distinctively proscribable content. Thus, the government may proscribe libel; but it
may not make the further content discrimination of proscribing only libel critical of the government.
R.A.V., 505 U.S. at 383-84 (internal citations omitted).
120 See Williams v. Rhodes, 393 U.S. 23, 32 (1968) (“Competition in ideas and governmental policies is at the core of
our electoral process and of the First Amendment freedoms.”); see also Buckley v. Valeo, 424 U.S. 1, 93 n.127 (1976)
(per curiam) (noting that the “the central purpose of the Speech and Press Clauses was to assure a society in which
‘uninhibited, robust, and wide-open’ public debate concerning matters of public interest would thrive, for only in such a
society can a healthy representative democracy flourish” (quoting N.Y. Times Co. v. Sullivan, 376 U.S. 254, 270
(1964))).
121 Knox v. SEIU, Local 1000, 567 U.S. 298, 308-09 (2012) (internal citations omitted).
122 See Snyder v. Phelps, 562 U.S. 443, 460-61 (2011) (“Speech is powerful. It can stir people to action, move them to
tears of both joy and sorrow, and—as it did here—inflict great pain. On the facts before us, we cannot react to that pain
by punishing the speaker. As a Nation we have chosen a different course—to protect even hurtful speech on public
issues to ensure that we do not stifle public debate.”).
123 W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624, 642 (1943).
124 See Citizens United v. FEC, 558 U.S. 310, 340 (2010) (“Laws that burden political speech are ‘subject to strict
scrutiny,’ which requires the Government to prove that the restriction ‘furthers a compelling interest and is narrowly
tailored to achieve that interest.’ While it might be maintained that political speech simply cannot be banned or
restricted as a categorical matter, the [strict scrutiny standard] provides a sufficient framework for protecting the
relevant First Amendment interests in this case.” (internal citations omitted)); Wash. Post v. McManus, No. PWG-182527, 2019 U.S. Dist. LEXIS 3073, at *22 (D. Md. Jan. 3, 2019) (“The test for [content-based] regulations mirrors the
test for laws that burden political speech, in that both are subject to strict scrutiny.”), appeal filed, No. 19-1132 (4th Cir.
Feb. 4, 2019).
125 See United States v. Stevens, 559 U.S. 460, 470-71 (2010) (acknowledging that the Supreme Court “has often
described historically unprotected categories of speech as being ‘of such slight social value as a step to truth that any
benefit that may be derived from them is clearly outweighed by the social interest in order and morality,’” but
explaining that when the Court has “identified categories of speech as fully outside the protection of the First
Amendment, it has not been on the basis of a simple cost-benefit analysis” (citation omitted)); United States v. Playboy
Entm’t Grp., 529 U.S. 803, 806, 826 (2000) (holding that “[b]asic speech principles” were at stake in a case involving a
federal law limiting the transmission of “sexually-oriented” cable programming and reasoning that the Court “cannot
be influenced . . . by the perception that the regulation in question is not a major one because the speech is not very
important”); Cincinnati v. Discovery Network, 507 U.S. 410, 428 (1993) (expressing unwillingness to recognize the
Congressional Research Service
18
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
generally “has no power to restrict expression because of its message, its ideas, its subject matter,
or its content.”126 The same is generally true for speech that the legislature considers “too harmful
to be tolerated,”127 as the government generally may not proscribe speech based on its content
unless that speech falls within one of the narrow categories of unprotected speech recognized by
the Supreme Court.128 Three of those categories are of particular relevance in the context of
online speech that promotes terrorism or violence.129
(1) Brandenburg’s Incitement Standard
In 1969, in Brandenburg v. Ohio, the Supreme Court considered a state law that prohibited
“advocat[ing] . . . the duty, necessity, or propriety of crime, sabotage, violence, or unlawful
methods of terrorism as a means of accomplishing industrial or political reform.”130 The state
convicted a Ku Klux Klan leader of violating the statute based on films of a Klan rally that
showed, among other things, hooded figures carrying firearms burning a cross and included, in
the Court’s words, “scattered phrases . . . that were derogatory of Negroes and, in one instance, of
city’s “bare assertion that the ‘low value’ of commercial speech is a sufficient justification for [the city’s] selective and
categorical ban on newsracks dispensing ‘commercial handbills’”).
126 Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786, 790-91 (2011) (quoting Ashcroft v. ACLU, 535 U.S. 564, 573
(2002)); see generally CRS In Focus IF11072, The First Amendment: Categories of Speech, by Victoria L. Killion.
127 Brown, 564 U.S. at 791. A plurality of the Court has alluded to a “historic and traditional” First Amendment
exception involving “speech presenting some grave and imminent threat the government has the power to prevent.” See
United States v. Alvarez, 567 U.S. 709, 717 (2012) (plurality opinion). However, the Justices acknowledged that “a
restriction under [this] category is most difficult to sustain,” citing a case where the Court held that the government did
not meet its burden to enjoin the publication of a classified study in the New York Times and the Washington Post.
Id. (citing N.Y. Times Co. v. United States, 403 U.S. 713 (1971) (per curiam)).
128 See Brown, 564 U.S. at 791 (“Last Term, in Stevens, we held that new categories of unprotected speech may not be
added to the list by a legislature that concludes certain speech is too harmful to be tolerated. . . . As in Stevens,
California has tried to make violent-speech regulation look like obscenity regulation by appending a saving clause
required for the latter. That does not suffice. Our cases have been clear that the obscenity exception to the First
Amendment does not cover whatever a legislature finds shocking, but only depictions of ‘sexual conduct.’” (citation
omitted)). But cf. Sherman, supra note 10, at 139 (noting that some legal scholars have “suggested that free speech
rules be modified in cases where the speech in question could be tied to ‘extraordinary’ levels of harm”).
129 See Hoffman, supra note 18, at 240 (noting the “strong likelihood of a First Amendment, ‘as applied’ Brandenburg
challenge” in a prosecution for “online terrorism advocacy”); Tsesis, supra note 18, at 664 (arguing that the
“incitement” and “true threats” doctrines offer “viable legislative approaches for restricting terrorist incitement or
propaganda” on the internet); Sherman, supra note 10, at 141 (positing that “[a]nother possible, though ultimately
problematic, approach [to addressing terrorist recruitment and calls for violence on social media without resort to
Brandenburg] comes from the Supreme Court’s jurisprudence on speech integral to criminal conduct”). Other
categories of unprotected speech could be relevant to online content promoting terrorism or violence depending on the
circumstances. For example, the First Amendment may not be a defense in certain cases involving “fighting words”—
words “which by their very utterance inflict injury or tend to incite an immediate breach of the peace.” Chaplinsky v.
New Hampshire, 315 U.S. 568, 572 (1942). However, the Supreme Court “has construed the fighting words doctrine
narrowly and consistently rejected [governmental] efforts to regulate speech that is merely insulting or offensive.”
Richard E. Levy, The Tweet Hereafter: Social Media and the Free Speech Rights of Kansas Public University
Employees, 24 KAN. J.L. & PUB. POL’Y 78, 105 n.146 (2014) (citing relevant decisions). Moreover, there remains an
open question as to whether online speech can ever qualify as “fighting words” because the speaker is unlikely to be
face-to-face or in close proximity with the recipient when the speech in question is made. See People ex rel. R.D., No.
14-CA-1800, 2016 COA 186, ¶ 19 (Colo. Ct. App. Dec. 29, 2016) (observing that “a number of state[’s] [courts] have
concluded that ‘[t]he potential to elicit an immediate violent response exists only where the communication occurs
face-to-face or in close physical proximity,’” but noting one decision in which the court upheld an injunction that
prohibited an individual from sending certain communications, including emails, to certain other individuals based on
the fighting words doctrine (citation omitted)), cert. granted, No. 17SC116, 2017 Colo. LEXIS 770 (Colo. Sept. 5,
2017).
130 395 U.S. 444, 444-45 (1969) (per curiam).
Congressional Research Service
19
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
Jews.”131 During a speech, the defendant stated, “We’re not a revengent organization, but if our
President, our Congress, our Supreme Court, continues to suppress the white, Caucasian race, it’s
possible that there might have to be some revengeance taken.”132 The Supreme Court reversed the
defendant’s conviction, concluding that the statute, by punishing “mere advocacy and
[forbidding], on pain of criminal punishment, assembly with others merely to advocate the
described type of action” violated the First Amendment.133 Effectively establishing a three-part
test, the Court held that “the constitutional guarantees of free speech and free press do not permit
a State to forbid or proscribe advocacy of the use of force or of law violation except where such
advocacy is [1] directed to inciting or producing [2] imminent lawless action and [3] is likely to
incite or produce such action.”134 The Court reiterated that “the mere abstract teaching . . . of the
moral propriety or even moral necessity for a resort to force and violence, is not the same as
preparing a group for violent action and steeling it to such action.”135
Supreme Court cases since Brandenburg have helped to elucidate its “directed to,” “imminence,”
and “likelihood” requirements to some degree—though not in the specific context of internet
speech. Hess v. Indiana involved a conviction for disorderly conduct stemming from an anti-war
rally at which the defendant shouted, “We’ll take the [expletive] street later.”136 The Court
overturned the defendant’s conviction because his statement, though made to a crowd of people,
“was not directed to any person or group of persons” and “amounted to nothing more than
advocacy of illegal action at some indefinite future time.”137 In the Court’s words, “there was no
evidence, or rational inference from the import of the language, that his words were intended to
produce, and likely to produce, imminent disorder.”138
Some argue that the Hess decision suggests that the Court views the “imminence” requirement to
mean that violence must be likely to occur immediately as a result of the speech at issue.139 State
and federal courts have not always applied Hess or the imminence requirement of Brandenburg
131 Id. at 445-46.
132 Id. at 446.
133 Id. at 449. Decisions prior to Brandenburg had allowed the government to ban the advocacy of violence under less
stringent standards. See, e.g., Whitney v. California, 274 U.S. 357, 371 (1927) (giving “great weight” to a state’s
determination that to “assist in organizing an association to advocate, teach or aid and abet the commission of crimes or
unlawful acts of force, violence or terrorism as a means of accomplishing industrial or political changes, involves such
danger to the public peace and the security of the State, that these acts should be penalized in the exercise of its police
power,” noting that “[e]very presumption is to be indulged in favor of the validity of the statute,” and holding that “it
may not be declared unconstitutional unless it is . . . arbitrary or unreasonable”), overruled by Brandenburg, 395 U.S. at
449; Dennis v. United States, 341 U.S. 494, 516-17 (1951) (upholding the defendants’ convictions because they
conspired “to organize the Communist Party and to teach and advocate the overthrow of the Government of the United
States by force and violence” and thereby “created a ‘clear and present danger’ of an attempt to overthrow the
Government by force and violence”). Cf. Dennis, 341 U.S. at 580 (Black, J., dissenting) (“I cannot agree that the First
Amendment permits us to sustain laws suppressing freedom of speech and press on the basis of Congress’ or our own
notions of mere ‘reasonableness.’”).
134 Brandenburg, 395 U.S. at 447 (emphasis added).
135 Id. at 448 (quoting Noto v. United States, 367 U.S. 290, 297-98 (1961)).
136 414 U.S. 105, 106-07 (1973) (per curiam).
137 Id. at 108-09.
138 Id. at 109.
139 See Michal Buchhandler-Raphael, Overcriminalizing Speech, 36 CARDOZO L. REV. 1667, 1677 (2015) (“In light of
Hess, imminent means nothing but immediate action, which is an almost impossible burden to satisfy.”). Cf. Marc
Rohr, Grand Illusion? The Brandenburg Test and Speech That Encourages or Facilitates Criminal Acts, 38
WILLAMETTE L. REV. 1, 18-19 (2002) (“[I]n Hess, the Court did appear to require that the interval between speech and
called-for response must be quite brief. Beyond that, uncertainty remains.”).
Congressional Research Service
20
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
so strictly.140 For example, in People v. Rubin, a California state court ruling, the defendant was
charged with solicitation of murder.141 During a press conference to protest an upcoming march
by the American Nazi Party through Skokie, IL, the defendant offered money to anyone who
“kills, maims, or seriously injures a member of the American Nazi Party.”142 He added, “This is
not said in jest, we are deadly serious.”143 The trial court concluded that his speech was protected
by the First Amendment, but the state appeals court reversed in a split decision with one judge
dissenting.144 Analogizing criminal solicitation to incitement, the appeals court applied
Brandenburg’s imminence and likelihood requirements and concluded that both were satisfied
even though the march in Skokie was not scheduled to take place until five weeks after the
defendant had spoken.145 The court wrote that “time is a relative dimension and imminence a
relative term, and the imminence of an event is related to its nature. . . . We think solicitation of
murder in connection with a public event of this notoriety, even though five weeks away, can
qualify as incitement to imminent lawless action.”146
Supreme Court decisions after Hess suggest that whether strong rhetoric is directed to inciting or
producing imminent lawless action depends on the context in which the statements at issue were
made—and to some degree, whether violence actually resulted. In NAACP v. Claiborne
Hardware Co., “17 white merchants” filed suit against the NAACP, its Field Secretary in
Mississippi, and over a hundred other individuals involved in a “boycott of white merchants in
Claiborne County, [Mississippi],” that was organized to protest racial discrimination, alleging
tortious interference with trade.147 In relevant part, at issue were claims arising from certain
speeches that the Field Secretary, Charles Evers, made during the middle of the boycott.148 In the
wake of the shooting and killing of “a young black man . . . during an encounter with two Port
Gibson police officers,” which led to mounting “[t]ension in the community” and “sporadic acts
of violence,” Evers allegedly stated that “boycott violators would be ‘disciplined’” and that if
anyone was caught entering the boycotted stores, “we’re gonna break your damn neck.”149
The Claiborne Hardware Court held that Evers was not liable to the boycotted store owners for
their economic losses because his speech was protected under the First Amendment.150 The Court
acknowledged that “[i]n the passionate atmosphere in which the speeches were delivered, they
might have been understood as inviting an unlawful form of discipline or, at least, as intending to
create a fear of violence whether or not improper discipline was specifically intended.”151 Still,
the Court held, “[t]he emotionally charged rhetoric . . . did not transcend the bounds of protected
speech set forth in Brandenburg” because the “strong language” used was part of “lengthy
addresses” that “generally contained an impassioned plea for black citizens to unify, to support
140 Thomas Healy, Brandenburg in a Time of Terror, 84 NOTRE DAME L. REV. 655, 670-72 (2009) (discussing cases
that, in the author’s view, arguably misconstrued Brandenburg by not adhering to its imminence requirement).
141 158 Cal. Rptr. 488, 489 (Cal. Ct. App. 1979). See Healy, supra note 140, at 673-74 (citing People v. Rubin as an
example that lower courts have not strictly applied Brandenburg’s imminence requirement).
142 Rubin, 158 Cal. Rptr. at 488.
143 Id. at 489.
144 Id. at 489, 494; see also id. at 494-500 (Roth, J., dissenting) (viewing the statements in context as hyperbole).
145 Id. at 492-93 (majority opinion).
146 Id.
147 458 U.S. 886, 888-90, 891 n.7, 915 (1982).
148 Id. at 902, 926.
149 Id. at 902.
150 Id. at 929.
151 Id. at 927.
Congressional Research Service
21
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
and respect each other, and to realize the political and economic power available to them.”152 Of
potential significance, the Court noted that “a substantial question [as to the defendant’s liability]
would be presented” if that language “had been followed by acts of violence,” but there was no
evidence of violence occurring after the challenged statements.153
In a later case, Texas v. Johnson, involving a criminal defendant’s First Amendment challenge to
his prosecution for burning a flag during a political protest, the Supreme Court ruled for the
defendant, finding it notable that despite the allegedly “disruptive behavior of the protestors
during their march . . . no actual breach of the peace occurred at the time of the flagburning or in
response to the flagburning.”154 The Court added that in such circumstances, the “likel[ihood]” of
imminent lawless action under Brandenburg cannot be inferred merely because an audience may
take “serious offense” to particular expression.155 The Court explained,
[A] principal “function of free speech under our system of government is to invite dispute.
It may indeed best serve its high purpose when it induces a condition of unrest, creates
dissatisfaction with conditions as they are, or even stirs people to anger.” Terminiello v.
Chicago, 337 U.S. 1, 4 (1949). It would be odd indeed to conclude both that “if it is the
speaker’s opinion that gives offense, that consequence is a reason for according it
constitutional protection,” FCC v. Pacifica Foundation, 438 U.S. 726, 745 (1978) (opinion
of Stevens, J.), and that the government may ban the expression of certain disagreeable
ideas on the unsupported presumption that their very disagreeableness will provoke
violence.156
Thus, the Court reasoned, to equate the potential for violence with speech “directed to” and
“likely to” incite or produce such action would be to “eviscerate [the Court’s] holding in
Brandenburg.”157
Scholars and commentators have noted the limits of the Brandenburg incitement doctrine when it
comes to regulating content on social media.158 In particular, many have questioned what
constitutes “imminence” when speech is made in an online forum rather than in connection with a
specific event where violence or unrest might be anticipated.159 Others have questioned how the
152 Id. at 928 (“An advocate must be free to stimulate his audience with spontaneous and emotional appeals for unity
and action in a common cause. When such appeals do not incite lawless action, they must be regarded as protected
speech. To rule otherwise would ignore the ‘profound national commitment’ that ‘debate on public issues should be
uninhibited, robust, and wide-open.’” (quoting N.Y. Times Co. v. Sullivan, 376 U.S. 254, 270 (1964))).
153 Id.
154 491 U.S. 397, 408 (1989) (internal quotation marks and citation omitted).
155 Id. at 408-09.
156 Id. (certain internal citations omitted).
157 Id. at 409. See generally United States v. Williams, 553 U.S. 285, 321-22 (2008) (Souter & Ginsburg, JJ.,
dissenting) (stating that the Brandenburg standard replaced “the rule that dominated the First World War sedition and
espionage cases, allowing suppression of speech for its tendency and the intent behind it” by “unmistakably insist[ing]
that any limit on speech be grounded in a realistic, factual assessment of harm”). Cf. Zieper v. Metzinger, 392 F. Supp.
2d 516, 518, 524-25 (S.D.N.Y. 2005) (reasoning that a video posted to a website approximately two months before the
New Year’s Eve celebration leading up to the year 2000 that showed locations in New York City’s Times Square and
included narration instructing various “teams” to have people “running for their lives” by midnight “can be seen as
provoking thought,” was “not a call to imminent riotous actions,” and was not stripped of First Amendment protection
merely because “a reasonable person viewing the video for this first time” may have thought “that someone had
planned to undertake activities that are unlawful”), aff’d on other grounds, 474 F.3d 60 (2d Cir. 2007).
158 See, e.g., John P. Cronan, The Next Challenge for the First Amendment: The Framework for an Internet Incitement
Standard, 51 CATH. U. L. REV. 425, 449 (2002) (noting that “[t]he Brandenburg standard was not created for
cyberspace and later courts interpreting the standard had no need to address the demands of the Internet”).
159 See Cronan, supra note 158, at 449-50 (arguing that Brandenburg’s “imminence” requirement is ambiguous when
Congressional Research Service
22
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
“directed to” or “likelihood” prongs of the test operate when speech is made to an unknown
audience of internet users rather than an assembled group in person.160
In applying Brandenburg to internet speech, a few lower courts have identified certain types of
content that may not constitute incitement but might constitute a true threat, another category of
unprotected speech discussed below.161 For example, the Third Circuit has stated that “merely
posting information on unlawful acts that have already occurred, in the past, does not incite
future, imminent unlawful conduct,” but held that under the circumstances of that case, the
defendants’ use of “past incidents to instill fear in future targets” amounted to unprotected
speech.162 That circuit also observed in dicta in another case that Brandenburg might allow the
government to obtain an injunction to “restrain a website published by a hate group naming
specific groups or individuals as targets, or specifying instructions for committing a crime.”163
(2) “True Threats”
As with incitement of the Brandenburg variety, the government may prohibit some forms of
intimidation such as “true” threats.164 True threats occur when the speaker “means to
communicate a serious expression of an intent to commit an act of unlawful violence to a
particular individual or group of individuals,” even if the speaker does not “actually intend to
carry out the threat.”165 In this way, the doctrine focuses on the harms related to the message the
speaker communicates rather than the possibility that it will stir others to commit violent acts.166
Like the line between incitement and “mere advocacy” that the Court drew in Brandenburg, the
Supreme Court has distinguished true threats from “political hyperbole.”167
In Watts v. United States—the 1969 decision coining the phrase “true threat”—the Court held that
a statute that prohibited any person from “knowingly and willfully . . . [making] any threat to take
applied to communications in cyberspace); Tsesis, supra note 18, at 667 (positing that a statute prohibiting incitement
“could be effective against immediate calls for violence through applications such as Instagram or Snapchat,” but
because of Brandenburg’s imminence requirement, may not supply a remedy for speech that “seeks long-term
indoctrination, mentoring, recruitment, and so on”).
160 See Nat’l Constitution Ctr., Free Speech Online (Stanford): Social Media Platforms and the Future of Democracy,
YOUTUBE, at 51:00 (May 3, 2018), https://www.youtube.com/watch?v=U8RFtagcjKk (remarks of Juniper Downs,
Global Head of Public Policy and Government Relations, YouTube) (“I think there are a couple of things about [the
Brandenburg] standard that don’t carry over well into the digital age. . . . [T]he problem with online speech is you have
speech that is going out to an unknown audience, and so evaluating whether someone saying . . . ‘go kill Americans’ is
likely to provoke such action is really difficult.”). Cf. Sherman, supra note 10, at 138-39 (“[I]t is questionable whether
[Brandenburg’s incitement standard] could be extended to many of the general calls for violence against, for example,
any available American targets. Instead, it would likely be limited to a relatively small number of situations in which a
target could be pinpointed with some degree of specificity from the speech in question.”).
161 See, e.g., United States v. Wheeler, 776 F.3d 736, 745 (10th Cir. 2015) (“Exhorting groups of followers to kill
specific individuals can produce fear in a recipient no less than more traditional forms of threats. Allowing defendants
to seek refuge in the First Amendment simply by phrasing threats as exhortations [under Brandenburg] would . . . leave
the state ‘powerless against the ingenuity of threateners.’” (citation omitted)).
162 United States v. Fullmer, 584 F.3d 132, 155-56 (3d Cir. 2009).
163 United States v. Bell, 414 F.3d 474, 483 n.10 (3d Cir. 2005).
164 See Watts v. United States, 394 U.S. 705, 708 (1969) (per curiam).
165 Virginia v. Black, 538 U.S. 343, 359-60 (2003).
166 See id. at 360 (“[A] prohibition on true threats protects individuals from the fear of violence and from the disruption
that fear engenders, in addition to protecting people from the possibility that the threatened violence will occur.”
(internal quotation marks and citation omitted)).
167 Watts, 394 U.S. at 708.
Congressional Research Service
23
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
the life of or to inflict bodily harm upon the President of the United States” was “constitutional on
its face.”168 But the Court ruled that it was improperly applied to an individual who, in the course
of expressing opposition to the draft during a public rally, stated, “If they ever make me carry a
rifle the first man I want to get in my sights is L.B.J. . . . They are not going to make me kill my
black brothers.”169 The Court reasoned that this “kind of political hyperbole” was not a “true
‘threat’” within the meaning of the statute because “[the Court] must interpret the language
Congress chose ‘against the background of a profound national commitment to the principle that
debate on public issues should be uninhibited, robust, and wide-open, and that it may well include
vehement, caustic, and sometimes unpleasantly sharp attacks on government and public
officials.’”170
Nearly 35 years later, in Virginia v. Black, the Supreme Court applied the true threats doctrine in a
case involving a state law prohibiting cross burning with the intent to intimidate.171 The Court
explained that “[i]ntimidation in the constitutionally proscribable sense of the word is a type of
true threat, where a speaker directs a threat to a person or group of persons with the intent of
placing the victim in fear of bodily harm or death.”172 The Court held that “[t]he First Amendment
permits [a state] to outlaw cross burnings done with the intent to intimidate because burning a
cross is a particularly virulent form of intimidation.”173 A plurality of the Court went on to
conclude that the particular statute before it was unconstitutional insofar as it included a
presumption making cross burning “prima facie evidence of an intent to intimidate a person or
group of persons.”174 They reasoned that such a presumption would likely result in convictions in
any cross-burning case regardless of the purpose of the cross burning and therefore chill protected
speech.175
While some scholars have cited the true threats doctrine as one avenue for the government to
regulate social media content promoting terrorism or violence,176 others have argued that the
Supreme Court’s decisions in this area may provide inadequate guidance for distinguishing
between real threats online and protected expression, particularly in cases where judges do not
share the same linguistic frame of reference as members of a particular online community.177
168 Id. at 705, 707.
169 Id. at 706.
170 Id. at 708 (quoting N.Y. Times Co. v. Sullivan, 376 U.S. 254, 270 (1964)).
171 538 U.S. 343 (2003).
172 Id. at 360.
173 Id. at 363.
174 Id. (quoting VA. CODE ANN. § 18.2-423 (1996)).
175 Id. at 365-66 (plurality opinion). The plurality reasoned that “a burning cross is not always intended to intimidate,”
and may not even be a statement of ideology (e.g., when used in a theatrical performance). Id. at 366. Justice Antonin
Scalia—who joined the plurality in the portion of the opinion holding that a state may prohibit cross burning with the
intent to intimidate—disagreed with his colleagues in the majority insofar as they argued that the possibility of an
unconstitutional conviction rendered the statute unconstitutional. Id. at 368-73 (Scalia, J., concurring in part and
dissenting in part). In Justice Scalia’s view, such convictions “could be challenged on a case-by-case basis” and did not
justify facial invalidation of the statute. Id. at 373.
176 Tsesis, supra note 18, at 670 (citing the “true threats doctrine along with the incitement doctrine” as “two pieces of
the constitutional puzzle lawmakers will need to put together in constructing a statute prohibiting terrorist threats on
social media”).
177 Lidsky & Riedemann Norbut, supra note 10, at 1891 (“Even the Court’s recent Facebook threats case, Elonis v.
United States, failed to explain how to interpret threatening speech in situations in which legal decision-makers do not
share a frame of linguistic reference with the speaker or her audience. This deficit, coupled with the doctrinal
uncertainty regarding the requisite mens rea, leaves legal decision-makers at a loss in separating social media threats
Congressional Research Service
24
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
This year, the Supreme Court declined to review the Pennsylvania Supreme Court’s interpretation
of the true threats doctrine in a case involving alleged “terroristic threats” posted to social
media.178 In Commonwealth v. Knox, the state court considered whether the defendant had a First
Amendment right to publish “a rap-music video containing threatening lyrics directed to named
law enforcement officers,” including two officers who were scheduled to testify against him in a
pending criminal case.179 The defendant argued that he never intended for the video to be
uploaded to social media (in that case, YouTube and Facebook) and that the song was a way to
express himself and was not meant to be taken literally.180 Several organizations filed briefs in
support of the defendant, noting, among other things, the defendant’s status as a semiprofessional
rap artist and the First Amendment protections accorded to speech about violence in other forms
of media.181
The Pennsylvania Supreme Court held that the song constituted a true threat.182 It began by
observing that although “First Amendment freedoms apply broadly to different types of
expression,” and while the government “generally lacks the authority to restrict expression based
on its message, topic, ideas, or content,” speech that “threatens unlawful violence can subject the
speaker to criminal sanction” under the true threats doctrine.183 The court explained that while the
Watts decision suggested that courts should use a “contextual,” “objective” standard in
determining whether speech constituted a “true threat,” courts since Virginia v. Black “have
disagreed over whether the speaker’s subjective intent to intimidate is relevant in a true-threat
analysis.”184 For its part, the Pennsylvania Supreme Court read the opinions in the Black case to
mean that the “First Amendment necessitates an inquiry into the speaker’s mental state,” which
can be discerned from context.185 After reviewing the lyrics in the defendant’s song, the court
concluded that certain “aspects of the song tend to detract from any claim that [the defendant’s]
words were only meant to be understood as an artistic expression of frustration”; namely, that the
song “mentions [two police officers] by name, stating that the lyrics are ‘for’ them,” and
“proceeds to describe in graphic terms how [the defendant] intends to kill those officers.”186 The
court also noted that the lyrics were “tied to interactions which had recently taken place between
from violent, yet protected, free expression.”); see also id. at 1902-13 (discussing the informal and often hyperbolic
nature of social media posts).
178 Commonwealth v. Knox, 190 A.3d 1146, 1150 (Pa. 2018), cert. denied, No. 18-949, 2019 U.S. LEXIS 2793 (Apr.
15, 2019).
179 Id. at 1148-49.
180 Id. at 1153.
181 Id.
182 Id. at 1149, 1160-61.
183 Id. at 1154-55.
184 Id. at 1155-58.
185 Id. at 1157 (“Construing the Court’s discussion [in Black] of the speaker’s intent as pertaining solely to the act of
transmitting the speech appears difficult to harmonize with the assertion that ‘[i]ntimidation in the constitutionally
proscribable sense of the word is a type of true threat, where a speaker directs a threat to a person or group of persons
with the intent of placing the victim in fear of bodily harm or death.’” (quoting Virginia v. Black, 538 U.S. 343, 360
(2003))). The Pennsylvania Supreme Court also cited the Supreme Court’s decision in Elonis v. United States, 135 S.
Ct. 2001 (2015), insofar as the Court held that a federal statute prohibiting the transmission of certain threatening
communications in interstate commerce that was silent as to the speaker’s intent must be interpreted to include an intent
requirement above mere negligence. Knox, 190 A.3d at 1157. Although the facts of Elonis are similar in that the
defendant was convicted of violating a criminal law on the basis of statements he made in “self-styled ‘rap’ lyrics”
posted to Facebook, the Elonis Court did not reach the defendant’s First Amendment challenge, instead overturning his
conviction because the jury was not instructed on the requisite mental state to convict. Elonis, 135 S. Ct. at 2004, 2012.
186 Knox, 190 A.3d at 1159.
Congressional Research Service
25
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
[the defendant and the named officers],” and that the named officers responded to hearing them
by taking measures to enhance their safety.187 With respect to the song’s publication online, the
court stated that “although the song was not communicated directly to the police and a third party
uploaded it to YouTube, this factor does not negate an intent on [the defendant’s] part that the
song be heard by the officers.”188 The court accepted the lower courts’ findings that the defendant
“either intended for the song to be published or knew publication was inevitable,” particularly
where a link to the YouTube video was later posted on a Facebook page thought to belong to a
codefendant who helped to write and record the song.189
The case law to date on online communications demonstrates that not all forms of alleged
intimidation are analyzed under the true threats doctrine. In a lower court case involving state tort
claims for invasion of privacy and intentional infliction of emotional distress, a federal district
court in Montana rejected the First Amendment arguments of the defendant, an “alt-right website”
publisher accused of launching an anti-Semitic “troll storm”190 against the plaintiff after
publishing several articles accusing her of “extortion” in business discussions with the “mother of
[a] prominent neo-Nazi.”191 The defendant allegedly published the plaintiff’s “phone numbers,
email addresses, and social media profiles, as well as those of her husband, twelve-year-old son,
friends, and colleagues” after which she and her family “received more than 700 disparaging
and/or threatening messages over phone calls, voicemails, text messages, emails, letters, social
media comments, and Christmas cards.”192 According to the court, the defendant had “called for
‘confrontation’ and ‘action,’ but he also told readers to avoid illegal activity.”193 In denying the
defendant’s motion to dismiss, the court did not analyze the defendant’s speech as a true threat or
any other category of unprotected speech, reasoning that “there is no categorical exception to the
First Amendment for harassing or offensive speech.”194 Instead, it considered whether it was clear
from the pleadings that the defendant was speaking on a matter of public concern, which might
give rise to a First Amendment defense under Supreme Court precedent.195 The court concluded
that the plaintiff had “made a plausible claim that [the defendant’s] speech involved a matter of
strictly private concern.”196 The court reasoned that although the defendant “drew heavily on his
readers’ hatred and fear of ethnic Jews, rousing their political sympathies, there is more than a
colorable claim that he did so strictly to further his campaign to harass [the plaintiff]” because of
“a perceived conflict” between the plaintiff and his friend’s mother.197
187 Id.
188 Id. at 1160.
189 Id. at 1149, 1160.
190 Merriam-Webster defines the verb “troll” as “to antagonize (others) online by deliberately posting inflammatory,
irrelevant, or offensive comments or other disruptive content.” Troll, MERRIAM-WEBSTER (2019),
https://www.merriam-webster.com/dictionary/troll.
191 Gersh v. Anglin, 353 F. Supp. 3d 958, 962-63 (D. Mont. 2018).
192 Id.
193 Id. at 963 (internal citation omitted).
194 Id. (quoting United States v. Osinger, 753 F.3d 939, 953 (9th Cir. 2014)). The court also noted that the plaintiff did
not argue that the defendant’s speech was categorically unprotected. Id.
195 Id. at 964 (explaining that the First Amendment “can serve as a defense in state tort suits,” but where “‘matters of
purely private significance are at issue, First Amendment protections are often less rigorous’” (quoting Snyder v.
Phelps, 562 U.S. 443, 452 (2011))).
196 Id. at 965.
197 Id. at 966.
Congressional Research Service
26
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
(3) Speech Integral to Criminal Conduct
Like incitement and true threats, speech that is integral to criminal conduct is considered
unprotected under the Court’s First Amendment jurisprudence.198 This exception has sometimes
been used to explain why the government may proscribe so-called “inchoate crimes—acts
looking toward the commission of another crime” such as conspiracy, solicitation, and attempt.199
Although these acts typically involve speech, the speech is “intended to induce or commence
illegal activities” and thus is “undeserving of First Amendment protection.”200 Once again, the
Supreme Court has distinguished in this context between “proposal[s] to engage in illegal
activity” and “the abstract advocacy of illegality,” extending the First Amendment’s protections
to the latter type of speech.201 However, as one district court has observed, the “line between
advocacy and action” can be “a hazy one,” particularly in the case of inchoate offenses where the
“action” that the law criminalizes “may be minor or look benign.”202
Some courts have cited the speech integral to criminal conduct doctrine in rejecting First
Amendment challenges to criminal convictions based on online communications.203 However, the
scope of the doctrine, particularly as it applies in the internet context, is not clear. One scholar has
suggested that “[h]aving to show that something on social media is ‘integral’ to criminal activity
seems like a tall order” except perhaps in “the case of a Tweet or Facebook posting along the
lines of ‘the bomb is in location x. Here is what you need to do to detonate it in location y at time
z.’”204 In contrast, another scholar has suggested that it is too easy, particularly in the context of
federal terrorism statutes, to prosecute someone for conspiracy based on online speech and
thereby avoid Brandenburg’s requirements.205 More broadly, others have proposed ways to cabin
198 See Giboney v. Empire Storage & Ice Co., 336 U.S. 490, 498 (1949) (“It rarely has been suggested that the
constitutional freedom for speech and press extends its immunity to speech or writing used as an integral part of
conduct in violation of a valid criminal statute. We reject the contention now.”).
199 United States v. Williams, 553 U.S. 285, 300 (2008). A related, alternative theory is that these offenses prohibit
conduct, not speech. See, e.g., United States v. Daly, 756 F.2d 1076, 1081-82 (5th Cir. 1985) (holding that statute
prohibiting conspiracy to defraud the United States prohibited conduct, not speech).
200 Williams, 553 U.S. at 298; e.g., United States v. Rahman, 189 F.3d 88, 117 (2d Cir. 1999) (discussing whether the
First Amendment protected defendant’s statements, including his advice to another individual to “find a plan to destroy
or to bomb or to . . . inflict damage to the American Army,” and reasoning that “[w]ords of this nature—ones that
instruct, solicit, or persuade others to commit crimes of violence—violate the law and may be properly prosecuted
regardless of whether they are uttered in private, or in a public speech”).
201 Williams, 553 U.S. at 298-99. See generally Eugene Volokh, The “Speech Integral to Criminal Conduct” Exception,
101 CORNELL L. REV. 981, 1030 (2016) (interpreting the “zone of unprotected speech” involving attempt to commit a
crime as “limited to speech that is preparatory to the actor’s (or the actor’s confederates’) commission of some other
nonspeech act” and positing that “a restriction on speech that angers listeners or creates a disturbance could not be
upheld on the grounds that the speech brings about a criminally prohibited result (anger or disturbance)”).
202 United States v. Nagi, 254 F. Supp. 3d 548, 558 (W.D.N.Y. 2017).
203 See, e.g., United States v. Ackell, No. 15-CR-123-01, 2017 U.S. Dist. LEXIS 105095, at *2-4, *13, *31 (D.N.H.
July 7, 2017) (rejecting defendant’s argument that the First Amendment barred the government from punishing his
online conduct, which included threatening to expose partially nude photographs of a 16-year-old girl (“R.R.”) unless
she continued to engage in a “dominant/submissive relationship” online with him, on the grounds that his
“communications with R.R. were integral to a course of criminal conduct and, therefore, unprotected insofar as they
were part and parcel of his ‘extortionate threats to harass and intimidate [R.R] if she terminated their . . . relationship’”
(quoting United States v. Sayer, 748 F.3d 425, 434 (1st Cir. 2014))), aff’d on other grounds, 907 F.3d 67 (1st Cir.
2018), petition for cert. filed, No. 18-7613 (Jan. 22, 2019); United States v. Hobgood, 868 F.3d 744, 747-48 (8th Cir.
2017) (holding that defendant’s violation of the federal stalking statute also constituted extortion, which is unprotected
speech integral to criminal conduct under the First Amendment).
204 Sherman, supra note 10, at 142.
205 Buchhandler-Raphael, supra note 139, at 1686 (arguing that “substantive criminal law—via its generous conspiracy
Congressional Research Service
27
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
the doctrine’s reach so that it is not used in a circular fashion to criminalize speech because of its
content without “serious First Amendment analysis.”206 In view of these considerations, this
category of unprotected speech appears to be less well-defined (and thus susceptible to broader
application) than the standards for incitement and true threats.207
Strict Scrutiny and Overbreadth
Whether the government seeks to regulate protected speech, unprotected speech, or both, the First
Amendment imposes some limitations on the means the government may use to achieve its
regulatory objectives. When a law regulates protected speech, courts generally apply strict
scrutiny and require the government to show that the law is narrowly tailored to achieve a
compelling governmental interest.208 In contrast, laws that primarily regulate nonexpressive
conduct or unprotected speech normally are not subjected to strict scrutiny.209 However, such laws
can sometimes be challenged as unduly overbroad on the grounds that a “substantial number of
[their] applications are unconstitutional, judged in relation to the statute’s plainly legitimate
sweep.”210 This section examines the strict scrutiny test and the overbreadth doctrine, as well as
applications of those doctrines in the context of terrorism-related offenses and criminal
prosecutions involving online speech, as those standards and precedents are likely to affect how
the government may regulate online speech promoting terrorism or violence.
Strict Scrutiny
As previously noted, when a law regulates protected speech—especially on the basis of its
content—courts generally require the government to prove: (1) a compelling governmental
interest, and (2) that the law is narrowly tailored to achieve that interest.211 As discussed below,
not all of the government’s interests in regulating speech promoting violence may rise to the level
doctrines—provides the government with ample measures to limit the spread of allegedly terrorist messages” because
conspiracy law “functions as an effective means for the government to bypass First Amendment jurisprudence’s strict
requirements for prohibiting advocacy of violence, resulting in a direct collision between these two areas of law”).
206 Volokh, supra note 201, at 987-88 (arguing that the “doctrine can’t justify treating speech as ‘integral to illegal
conduct’ simply because the speech is illegal under the law that is being challenged,” stating that this “should be
obvious, since the whole point of modern First Amendment doctrine is to protect speech against many laws that make
such speech illegal,” but observing that the doctrine has “become, at times, a tool for avoiding serious First
Amendment analysis—a way to uphold speech restrictions as supposedly fitting within an established exception,
without a real explanation of how the upheld restrictions differ from other restrictions that would be struck down”).
207 See United States v. Matusiewicz, 84 F. Supp. 3d 363, 369 (D. Del. 2015) (“Under the broadest interpretation, if the
government criminalized any type of speech, then anyone engaging in that speech could be punished because the
speech would automatically be integral to committing the offense. That interpretation would clearly be inconsistent
with the First Amendment . . . .”), aff’d sub nom. United States v. Gonzalez, 905 F.3d 165 (3d Cir. 2018).
208 See supra note 124. There are, however, some exceptions. For example, laws that regulate commercial speech
generally receive intermediate scrutiny, as do laws that impose content-neutral time, place, or manner restrictions on
speech. See generally Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm’n, 447 U.S. 557, 566 (1980); Ward v. Rock
Against Racism, 491 U.S. 781, 798-800 (1989).
209 As previously noted, however, the First Amendment still places other limitations on regulations involving
unprotected speech. See supra note 119. For example, the “government may not regulate . . . based on hostility—or
favoritism—towards the underlying message expressed.” R.A.V. v. City of St. Paul, 505 U.S. 377, 386 (1992).
210 United States v. Stevens, 559 U.S. 460, 473 (2010).
211 See supra note 112 and associated text.
Congressional Research Service
28
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
of compelling.212 In addition, even if the government can demonstrate a compelling interest, it
would still have to show that the law is sufficiently tailored to that interest.213
Governmental Interests
The Supreme Court has held that the government does not have a compelling interest—or even a
substantial one214—in shielding listeners from messages that they might find offensive.215 In
addition, where the government has deemed certain speech harmful, the Court has in some cases
required the government to demonstrate actual harms that the regulation can redress. 216 In
contrast, the Court has repeatedly recognized the government’s compelling interests in
maintaining national security and combating terrorism.217 It has also acknowledged that Congress
and the executive branch are uniquely positioned—both as a constitutional matter and in terms of
their expertise—to safeguard the nation’s security and regulate foreign affairs.218 Accordingly, it
212 See infra note 215, citing cases in which the Court rejected justifications based on protecting listeners from
offensive or demeaning speech.
213 See Sable Commc’ns of Cal. v. FCC, 492 U.S. 115, 126 (1989) (“It is not enough to show that the Government’s
ends are compelling; the means must be carefully tailored to achieve those ends.”).
214 The standard for intermediate scrutiny, which typically applies to commercial speech, is whether the law directly
advances a substantial governmental interest and is narrowly drawn to achieve that interest. See Cent. Hudson, 447 U.S.
at 564-65.
215 See Matal v. Tam, 137 S. Ct. 1744, 1764 (2017) (plurality opinion) (analyzing the Patent and Trademark Office’s
prohibition against disparaging trademarks under intermediate scrutiny, rejecting the government’s asserted interest in
“preventing ‘underrepresented groups’ from being ‘bombarded with demeaning messages in commercial advertising,’”
and stating that the idea that the government “has an interest in preventing speech expressing ideas that offend . . .
strikes at the heart of the First Amendment” (internal citation omitted)). Cf. Snyder v. Phelps, 562 U.S. 443, 459 (2011)
(considering whether the First Amendment barred recovery for intentional infliction of emotional distress on the facts
presented, and stating, “In most circumstances, ‘the Constitution does not permit the government to decide which types
of otherwise protected speech are sufficiently offensive to require protection for the unwilling listener or viewer.
Rather, . . . the burden normally falls upon the viewer to avoid further bombardment of [his] sensibilities simply by
averting [his] eyes.’ As a result, ‘[t]he ability of government, consonant with the Constitution, to shut off discourse
solely to protect others from hearing it is . . . dependent upon a showing that substantial privacy interests are being
invaded in an essentially intolerable manner.’” (internal citations omitted)); Robinson v. Hunt Cty., No. 18-10238,
2019 U.S. App. LEXIS 11013, at *9 (5th Cir. Apr. 15, 2019) (stating, in a case involving a county’s alleged removal of
“inappropriate” posts from its official Facebook page, that “[o]fficial censorship based on a state actor’s subjective
judgment that the content of protected speech is offensive or inappropriate is viewpoint discrimination”).
216 See Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786, 799-801 (2011) (stating that the government “must specifically
identify an ‘actual problem’ in need of solving, and the curtailment of free speech must be actually necessary to the
solution” and holding that the state did not meet that standard when it sought to prohibit the sale of violent video games
to minors because it could not “show a direct causal link between violent video games and harm to minors,” the
research the state proffered showed at best a correlation between the two, and any effects such games had on children
were “both small and indistinguishable from effects produced by other media” (internal citations omitted)).
217 See Holder v. Humanitarian Law Project, 561 U.S. 1, 28 (2010) (“Everyone agrees that the Government’s interest in
combating terrorism is an urgent objective of the highest order.”); Haig v. Agee, 453 U.S. 280, 307 (1981) (“It is
‘obvious and unarguable’ that no governmental interest is more compelling than the security of the Nation.” (citation
omitted)).
218 See Trump v. Hawaii, 138 S. Ct. 2392, 2422 (2018) (stating that “‘[j]udicial inquiry into the national-security realm
raises concerns for the separation of powers’ by intruding on the President’s constitutional responsibilities in the area of
foreign affairs” and that “‘when it comes to collecting evidence and drawing inferences’ on questions of national
security, ‘the lack of competence on the part of the courts is marked.’” (quoting Ziglar v. Abbasi, 137 S. Ct. 1843
(2017) and Humanitarian Law Project, 561 U.S. at 34)); Aptheker v. Sec’y of State, 378 U.S. 500, 509 (1964) (“That
Congress under the Constitution has power to safeguard our Nation’s security is obvious and unarguable.”); Kennedy v.
Mendoza-Martinez, 372 U.S. 144, 160 (1963) (“Congress has broad power under the Necessary and Proper Clause to
enact legislation for the regulation of foreign affairs. Latitude in this area is necessary to ensure effectuation of this
indispensable function of government.”).
Congressional Research Service
29
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
has recognized some situations in which the First Amendment must yield to the “exclusive[]”
authority of the political branches to “maintain[] normal international relations and defend[] the
country against foreign encroachments and dangers.”219
Narrow Tailoring
Once the government has established a compelling interest, the focus of the First Amendment
inquiry is on whether the law is sufficiently tailored to achieve that interest. Under a strict
scrutiny standard, narrow tailoring typically means that the law has to be the least speechrestrictive means of advancing the government’s interest.220 For example, in Reno v. ACLU, the
Supreme Court struck down two provisions of the Communications Decency Act of 1996 (CDA)
that banned the knowing transmission of “indecent” messages, and the knowing sending or
display of “patently offensive” messages, to minors over the internet.221 The CDA contained
affirmative defenses for persons who took “good faith, reasonable, effective, and appropriate
actions” to restrict minors’ access to the prohibited communications or who restricted access
through certain age-verification measures such as requesting a verified credit card.222 The Court
first concluded that the terms “indecent” and “patently offensive” were vague and thus threatened
to “silence[] some speakers whose messages would be entitled to constitutional protection.”223
Moving to the narrow tailoring analysis, the Court then explained that the CDA’s “burden on
protected speech cannot be justified if it could be avoided by a more carefully drafted statute.”224
The Court recounted the district court’s findings regarding the technological limitations of
restricting minors’ access to such content and the cost-prohibitive nature of age-verification
solutions.225 In contrast, the district court had found that available software for parental controls
could curtail minors’ access in a reasonably effective way.226 The Court concluded that the
government failed to “explain why a less restrictive provision would not be as effective as the
CDA,” and thus the challenged provisions were not narrowly tailored.227
219 Kleindienst v. Mandel, 408 U.S. 753, 765 (1972) (internal quotation marks and citation omitted); see also id. at 769-
70 (declining to second-guess the Executive’s decision to exclude a foreign scholar “on the basis of a facially legitimate
and bona fide reason” or to “test [that decision] by balancing its justification against the First Amendment interests of
those [U.S. persons] who [sought] personal communication with [him]” in view of the “plenary congressional power to
make policies and rules for exclusion of aliens,” which Congress delegated for this purpose to the Executive).
220 See Sable Commc’ns of Cal. v. FCC, 492 U.S. 115, 126 (1989) (stating that the government may “regulate the
content of constitutionally protected speech in order to promote a compelling interest if it chooses the least restrictive
means to further the articulated interest”).
221 521 U.S. 844, 859-60 (1997) (quoting 47 U.S.C. § 223(a) and (d) (Supp. 1997)).
222 Id. at 860-61 (quoting 47 U.S.C. § 223(e)(5)(A) and citing § 223(e)(5)(B)).
223 Id. at 874.
224 Id. (“In order to deny minors access to potentially harmful speech, the CDA effectively suppresses a large amount of
speech that adults have a constitutional right to receive and to address to one another. That burden on adult speech is
unacceptable if less restrictive alternatives would be at least as effective in achieving the legitimate purpose that the
statute was enacted to serve.”).
225 Id. at 876-77.
226 Id. at 877.
227 Id. at 879 (“The arguments in this Court have referred to possible alternatives such as requiring that indecent
material be ‘tagged’ in a way that facilitates parental control of material coming into their homes, making exceptions
for messages with artistic or educational value, providing some tolerance for parental choice, and regulating some
portions of the Internet—such as commercial web sites—differently than others, such as chat rooms. Particularly in the
light of the absence of any detailed findings by the Congress, or even hearings addressing the special problems of the
CDA, we are persuaded that the CDA is not narrowly tailored if that requirement has any meaning at all.”).
Congressional Research Service
30
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
Deference to Political Branches
While a court may examine the availability and effectiveness of less speech-restrictive
alternatives under some circumstances,228 it may exercise some deference in evaluating the policy
judgments of Congress and the Executive in certain areas such as national security and foreign
affairs.229 The Humanitarian Law Project decision discussed above is one case in which the Court
applied stringent scrutiny but deferred in some measure to the means that Congress chose to
combat terrorism.230 In that case, the Supreme Court construed the federal material support statute
to ban “only material support coordinated with or under the direction of” an FTO, not
“[i]ndependent advocacy that might be viewed as promoting the group’s legitimacy.”231 It
concluded that the statute was “carefully drawn” insofar as it reached speech rather than
conduct.232 The Court then considered whether the law, as applied to the plaintiffs’ proposed
activities, was “necessary to further” the government’s compelling interest in combating
terrorism.233 In this regard, the Court examined the plaintiffs’ contention that they sought to
advance only the “legitimate activities of the [FTOs], not their terrorism.”234 Reasoning that
whether FTOs “meaningfully segregate support of their legitimate activities from support of
terrorism is an empirical question,” the Court gave “significant weight” to “the considered
judgment of Congress and the Executive that providing material support to [an FTO]—even
seemingly benign support—bolsters the terrorist activities of that organization.”235 The Court then
provided examples of how the plaintiffs’ support could potentially further the terrorist objectives
of the organizations.236
In upholding the statute as applied to the plaintiffs’ activities, the Court cabined its decision in
three respects.237 First, it held that it was not opining on the constitutionality of “any future
228 In Reno, the Court explained that the “breadth” and “content-based” nature of the challenged CDA provisions
“impose[d] an especially heavy burden on the Government to explain why a less restrictive provision would not be as
effective as the CDA.” Id. at 879.
229 See Trump v. Hawaii, 138 S. Ct. 2392, 2422 (2018) (“While we of course ‘do not defer to the Government’s reading
of the First Amendment,’ the Executive’s evaluation of the underlying facts is entitled to appropriate weight,
particularly in the context of litigation involving ‘sensitive and weighty interests of national security and foreign
affairs.’” (quoting Holder v. Humanitarian Law Project, 561 U.S. 1, 33-34 (2010))); e.g., Al Haramain Islamic Found.,
Inc. v. United States Dep’t of the Treasury, 660 F.3d 1019, 1033 (9th Cir. 2011) (“[T]he government’s interest in
national security cannot be understated. We owe unique deference to the executive branch’s determination [in the
executive order] that we face ‘an unusual and extraordinary threat to the national security’ of the United States.”).
230 The dissenting Justices observed that while the majority rejected the application of the “intermediate First
Amendment standard that applies to conduct,” the majority did not expressly adopt a strict scrutiny standard, stating
only that a “more demanding standard” was required. Humanitarian Law Project, 561 U.S. at 45 (Breyer, J.,
dissenting) (internal quotation marks omitted) (“Indeed, where, as here, a statute applies criminal penalties and at least
arguably does so on the basis of content-based distinctions, I should think we would scrutinize the statute and
justifications ‘strictly’—to determine whether the prohibition is justified by a ‘compelling’ need that cannot be ‘less
restrictively’ accommodated.” (citations omitted)).
231 Humanitarian Law Project, 561 U.S. at 31-32 (majority opinion).
232 Id. at 26.
233 Id. at 28 (citation omitted).
234 Id. at 29.
235 Id. at 29, 36. The Court also noted that “persuasive evidence” in the record supported the judgment of the two
political branches. Id. at 36.
236 Id. at 30, 36-38 (reasoning, for example, that material support “frees up other resources within the organization that
may be put to violent ends” and “helps lend legitimacy to foreign terrorist groups . . . that makes it easier for those
groups to persist, to recruit members, and to raise funds”).
237 Id. at 39.
Congressional Research Service
31
Terrorism, Violent Extremism, and the Internet: Free Speech Considerations
applications of the material-support statute to speech or advocacy” or “any other statute relating
to speech and terrorism.”238 Second, it suggested that “a regulation of independent speech” may
not “pass constitutional muster, even if the Government were to show that such speech benefits
[FTOs].”239 And third, it expressly disclaimed any suggestion “that Congress could extend the
same prohibition on material support at issue here to domestic organizations.”240
The Humanitarian Law Project decision was criticized by several dissenting Justices and some
outside commentators for a perceived departure from settled First Amendment standards. The
three dissenting Justices agreed that the government has a “compelling” interest “in protecting the
security of the United States” by “denying [FTOs] financial and other fungible resources.”241
However, they argued that the government failed to show how applying the statute to the
plaintiffs’ activities would “help achieve that important security-related end,” which is typically
required under any level of heightened scrutiny.242 Legal scholars have disagreed as to whether
the decision is consistent with the protections accorded to political advocacy in other contexts
such as campaign finance.243 Others have argued that the Court erred in prohibiting “mere
advocacy” without asking
This text is long and has been trimmed here. Open the source document for the complete record.
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.