Published by The Lawfare Institute
in Cooperation With
Less than three months after oral argument, the Supreme Court issued unanimous opinions in Twitter v. Taamneh and Gonzalez v. Google, a pair of Anti-Terrorism Act cases arising from the U.S. Court of Appeals for the Ninth Circuit. The plaintiffs in those cases alleged that social media companies—including Facebook, Google, and Twitter—are liable under the Justice Against Sponsors of Terrorism Act (JASTA) for aiding and abetting ISIS in relation to the 2015 Paris attacks and the 2017 Reina nightclub attack in Turkey. The Court took a comparatively narrow view of JASTA’s aiding and abetting liability, reversing the Ninth Circuit’s decision in Taamneh and vacating Gonzalez for the circuit to reconsider in light of Taamneh. In doing so, the Court effectively redrafted JASTA to minimize its core mechanism—a D.C. Circuit case called Halberstam v. Welch intended to guide the inquiry into liability. The new standard instead follows the common law and requires “conscious, voluntary, and culpable participation in another’s wrongdoing.” This modification will add limited clarity to JASTA litigation and may have been a more appropriate task for Congress.
Many expected the Court to focus on Section 230 of the Communications Decency Act, which lower courts have interpreted to provide social media companies with broad immunity from liability for content on their platforms. The breadth of this immunity represented a key issue in Gonzalez, as the Ninth Circuit rejected most of the plaintiffs’ claims based on Section 230. The relevant portion of the text reads that “[n]o provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” The Gonzalez plaintiffs argued below that Google’s failure to remove ISIS content created liability regardless of its status as a publisher, but the plaintiffs abandoned this route before the Supreme Court, focusing instead on the company’s recommendation algorithms. This more technical line of argument, taking major issue with Google’s production of video thumbnails, perplexed the justices. Justice Samuel Alito expressed to counsel, for example, that he was “completely confused by whatever argument you’re making at the present time.” The Court may have felt unprepared to base its decision on the Section 230 issue, moving on to JASTA instead.
I argued recently on these pages that JASTA is a troubled statute that has confused appellate courts and produced at least three circuit splits. Halberstam provides acceptable guidance for cases against aficionados of petty crime, but its three elements and six subfactors have left companies and nongovernmental organizations (NGOs) operating in high-risk areas with limited guidance on how to comply with the law. Some courts even determined that arm’s-length transactions with the general population could support liability under the statute. I suggested that Congress revisit JASTA to nix Halberstam and supply factors specific to acts of international terrorism. The Supreme Court tried to do the job itself in Taamneh. Although Congress identified Halberstam as providing the “proper legal framework” to evaluate aiding and abetting in terrorism cases, the Supreme Court crafted a new standard, effectively rewriting the statute to allow for modification of the Halberstam framework and consideration of other factors sourced from the common law.
The Taamneh Opinion
As the Supreme Court’s opinion recounted, the Taamneh case arose from the 2017 terrorist attack at Istanbul’s Reina nightclub. A gunman killed 39 people, including Nawras Alassaf, whose family filed suit against Facebook, Google, and Twitter, alleging that the companies’ services played a central role in “the explosive growth of ISIS” and that they did not sufficiently combat the organization’s extensive presence on the platforms. Knowingly allowing and supporting ISIS’s use of these services constituted aiding and abetting, the plaintiffs alleged. YouTube’s advertisement revenue-sharing system also sent money to ISIS, according to the plaintiffs, representing a separate source of substantial assistance. The district court determined that the plaintiffs failed to state a claim. The Ninth Circuit reversed, reasoning that Alassaf’s family members sufficiently pleaded that the defendant companies’ services “were central to ISIS’s growth and expansion, and that this assistance was provided over many years.” The Supreme Court took up the case to decide whether the use of services available to the public at large and unconnected to the attack itself could support JASTA liability.
The Court’s opinion, authored by Justice Clarence Thomas, reviewed the background and structure of the Anti-Terrorism Act and JASTA. The Court saw JASTA liability as comprising four components. First, there must be an act of international terrorism; second, the act of international terrorism must be committed, planned, or authorized by a terrorist organization; third, the terrorist organization must be designated as a foreign terrorist organization (FTO) at the time of the attack; and fourth, the defendant must have “aided and abetted.”
The nature and object of this “aiding and abetting” represented the core issue in Taamneh and, as I covered previously, the subject of a circuit split. The D.C. Circuit and the Ninth Circuit in Taamneh held that assistance to the terrorist organization is sufficient to incur liability, while the Second Circuit specified that the defendant must knowingly support the injurious attack itself. Courts have also varied greatly in their application of JASTA’s aiding and abetting standard, even between Taamneh and Gonzalez, again as discussed in the previous article. The Court’s opinion broke the “aiding and abetting” inquiry into two parts.
First, it considered what “aiding and abetting” entails under JASTA. Congress included within JASTA’s findings and purpose that a D.C. Circuit case—Halberstam v. Welch—“provides the proper legal framework for how [aiding and abetting] liability should function in the context of” terrorism cases. In that case, the D.C. Circuit determined that the live-in companion of a burglar could be held liable for aiding and abetting the murder of a burglary victim even though she was neither present at the time of the crime nor aware that he planned to burglarize the victim’s home. Halberstam breaks the inquiry neatly into three elements: “(1) the party whom the defendant aids must perform a wrongful act that causes an injury; (2) the defendant must be generally aware of his role as part of an overall illegal or tortious activity at the time that he provides the assistance; [and] (3) the defendant must knowingly and substantially assist the principal violation.” The third element is then split into six subfactors to determine whether any assistance is substantial. For the past seven years, lower courts have followed Congress’s apparent instruction, applying the Halberstam elements and factors to examine whether assistance was substantial. I have criticized applications of Halberstam to JASTA cases, as it is poorly tailored to international terrorism fact patterns.
The Court recounted Halberstam but observed that its framework should be adapted to fit new cases, noting that the test “may not be entirely adequate to resolve these new facts.” The opinion pivoted to the common law roots of criminal aiding and abetting liability, emphasizing its limits. The Court opined that “if aiding-and-abetting liability were taken too far, then ordinary merchants could become liable for any misuse of their goods and services, no matter how attenuated their relationship with the wrong-doer.” Critically, the opinion relied on Nye & Nissen v. United States, a 1949 Supreme Court case, quoting Learned Hand’s observation that a defendant must “in some sort associate himself with the venture, that he participate in it as in something that he wishes to bring about, that he seek by his action to make it succeed.”
The opinion next turned to the sparse federal case law on civil aiding and abetting liability, with some cases indicating that the defendant’s assistance “must be calculated and intended to produce” the harm, that a degree of “culpability” is required, and that “inaction” does not create liability. The Court also asserted that the “knowing and substantial assistance” to the primary tortfeasor—the third element in Halberstam’s test—actually “work[s] in tandem, with a lesser showing of one demanding a greater showing of the other.” “[A]ny approach that too rigidly focuses on Halberstam’s facts, or its exact phraseology risks missing the mark,” the Court stated. In conclusion, the Court determined that “[t]he phrase ‘aids and abets’ in §2333(d)(2), as elsewhere, refers to a conscious, voluntary, and culpable participation in another’s wrongdoing.” JASTA thus no longer requires strict adherence to the three-element, six-subfactor Halberstam test to determine whether assistance was knowing and substantial.
Second, the Court turned to the issue of exactly what a JASTA defendant must aid or abet. Need the object of support be international terrorism generally or the attack itself? The statute’s text is not crystal clear on this point, assigning liability to anyone “who aids and abets, by knowingly providing substantial assistance, or who conspires with the person who committed such an act of international terrorism.” Yet the Court passed on textual interpretation, reaching again for the common law, where “aiding and abetting is inherently a rule of secondary liability for specific wrongful acts.” And under the common law, “a defendant must have aided and abetted (by knowingly providing substantial assistance) another person in the commission of the actionable wrong—here, an act of international terrorism.”
The Court next sought to fit this principle into Halberstam’s fact pattern, where the defendant acted as “banker, bookkeeper, recordkeeper, and secretary” for the career burglar but did not accompany him on his criminal escapades. The Court determined that her assistance “was so intentional and systematic that she assisted each and every burglary.” In the context of JASTA, this means that a defendant that does not support an individual attack may still be liable if assistance to the terrorist entity is sufficiently “intentional and systematic.”
The opinion then qualified its rule even further by rejecting the need for a strict nexus between the aiding and abetting activity and the attack. A defendant need not know, for example, “all the particulars of the primary actor’s plan” nor the total scope of the eventual injury. “[R]emote support can still constitute aiding and abetting in the right case,” and the Court reiterated that “in appropriate circumstances, a secondary defendant’s role in an illicit enterprise can be so systemic that the secondary defendant is aiding and abetting every wrongful act committed by that enterprise.”
In summary, JASTA now requires that the defendant support an attack directly, but exceptions risk swallowing the rule.
Finally, the Court applied this new reasoning to the facts at hand. Returning to Halberstam, the Court concluded that the plaintiffs’ allegations satisfied the test’s first two elements—ISIS committed a tort, and the companies were generally aware of their role in that tort. But the allegations “fall short” when it comes to knowing and substantial assistance.
The opinion noted that ISIS used Facebook, YouTube, and Twitter just like everyone else, and the plaintiffs never alleged that the Reina nightclub gunman used those services. The companies did not associate themselves with the attack, wish to bring it about, or seek to make it succeed. Mere creation of the platforms did not create culpability, just like any internet or cell company that provides “services to the public writ large.” The Court rejected the argument that algorithms set the social media companies apart, as they “appear agnostic as to the nature of the content, matching any content (including ISIS’ content) with any user who is more likely to view that content.” And the platforms do not incur liability for failing to cleanse their services of ISIS content.
Given this “passive nonfeasance,” the Court continued, plaintiffs would need to make “a strong showing of assistance and scienter.” This appeared to implement the new approach where less substantial assistance requires greater evidence of knowledge and conscious participation. The Court noted the existence of a “highly attenuated ... arm’s length, passive, and largely indifferent” relationship between the social media companies and ISIS. It also pointed to the “distance” in causation between the defendants’ acts and the Reina attack, given that the companies did not encourage, solicit, or advise the commission of the attack.
In addition, the allegations did not amount to systemic and pervasive assistance that would obviate the need for a close nexus. The Court rested this conclusion in part on policy concerns, observing that “a contrary holding would effectively hold any sort of communication provider liable for any sort of wrongdoing merely for knowing that the wrongdoers were using its services and failing to stop them.”
Next, the Court marched through the Ninth Circuit’s mistakes. The Ninth Circuit erred, according to the Court, by framing the substantial assistance issue as turning on the defendant’s generalized assistance to ISIS rather than assistance to the gunman. It erred by separating out “knowledge” from “substantial assistance,” as those two components should be considered together to determine whether the conduct was culpable. And it erred by considering the six Halberstam factors individually rather than as part of an interwoven effort to determine whether “participation in another’s wrongdoing ... is both significant and culpable enough to justify attributing the principal wrong-doing to the aider and abettor.” The “[f]undamental question,” the Court repeated, is “[d]id defendants consciously, voluntarily, and culpably participate in or support the relevant wrong-doing?”
Finally, in two short paragraphs at the end of the opinion, the Court affirmed the Ninth Circuit’s rejection of the plaintiff’s claim that advertising revenue-sharing amounted to substantial assistance, agreeing with the lower court’s reasoning that the complaint did not share how much money Google sent to ISIS. For all we know, the opinion surmised, it “could be the case that Google approved only one ISIS-related video and shared only $50 with someone affiliated with ISIS.” Plaintiffs, thwarted by Twombly and Iqbal’s plausibility pleading standard, may have been unable to provide precise dollar amounts without the benefit of discovery.
The New JASTA
Twitter v. Taamneh provides much fodder for attorneys and lower courts to work through in the coming years. While the Supreme Court left most of the statute in place, it revised one of the most actively litigated parts of its scheme—the standard for knowing and substantial assistance. The Court released litigants from the strict bounds of the Halberstam test and introduced additional points for federal courts to consider when evaluating JASTA claims for secondary liability. And it did so by construing Congress’s instruction that Halberstam “provides the proper legal framework” to mean that federal courts should just look to the common law rather than the case’s clearly laid-out test or the text of the statute.
The opinion no doubt raises the pleading standard for JASTA plaintiffs. But the Supreme Court’s new framework leaves much vagueness for future litigation, which represented a key issue in the previous Halberstam scheme for potential defendants attempting to tailor their activity in high-risk areas. So too did the underdetermined standard leave traumatized plaintiffs subject to lengthy litigation that may ultimately prove fruitless. First of all, Taamneh does not present a clear pathway for federal courts to follow when evaluating claims, piling on additional considerations but failing to specify where they might fit in a court’s analysis. Additionally, what does “conscious, voluntary, and culpable participation” really mean, and how does it merge into the test for liability? What form of activity will be so “intentional and systematic” as to overcome the need to show a close nexus to the attack? How broad is the related exception for “remote support”? Plaintiffs, as well as companies and NGOs facing uncertainty under the previous arrangement, will likely continue to feel perplexed by Taamneh’s revisions.
In addition, the Court seemed confident in Taamneh that federal case law on civil aiding and abetting will guide lower courts through the inquiry better than Halberstam can on its own. But the common law tort of aiding and abetting is not well developed within federal courts, particularly when it comes to complicated transnational fact patterns. In its opinion, the Court relied on Central Bank of Denver, which itself cautioned that “[t]he doctrine has been at best uncertain in application.” Halberstam also noted that its reasoning was “only a beginning probe into tort theories.” Federal civil aiding and abetting cases in recent years have largely taken place within JASTA lawsuits, and related case law will now be of limited value in post-Taamneh litigation. Federal courts have been reset to square one with JASTA aiding and abetting case law, tasked with fashioning new landmarks using different terms of art.
While Taamneh opens the door to further litigation on these points, it leaves additional issues unresolved. For example, must the foreign terrorist organization be a central actor in the attack, as the Sixth Circuit demanded, or is it enough that an FTO be tangentially involved, as the D.C. Circuit concluded? While Taamneh addressed the third Halberstam prong regarding knowing and substantial assistance, it did not clarify the second prong evaluating general awareness. What forms this “awareness,” and does it even matter for post-Taamneh litigation?
Taamneh does not mark the end of terrorism litigation. Since Congress enacted JASTA in 2016, plaintiffs have refined and adapted their complaints to follow case law in the lower courts, finding greater success in recent years. There is no reason to believe that plaintiffs will suddenly stop pursuing relief for their terrible injuries, particularly when Anti-Terrorism Act verdicts result in treble damages reaching hundreds of millions of dollars and defendants may be prone to settle. Plaintiffs must now shift their complaints to allege that defendants’ activity was conscious, culpable, and systematic. It is reasonable to believe that the number of filings—29 cases in 2019 and 23 cases in 2020 by my count—may decrease in the coming years. But, horribly, terrorists continue to kill Americans, leaving room for future lawsuits.
Congress should still revisit JASTA to more clearly delineate liability and provide courts with factors bearing greater relevance to transnational terrorism fact patterns. The statute needed renovation, but the Supreme Court has only added vague standards to this high-stakes litigation while wiping the slate of federal case law clean. The Court substituted the common law for Congress’s seemingly clear dictate to use Halberstam’s framework, effectively rewriting the statute. It remains to be seen whether JASTA litigants will be better off following the Supreme Court’s intervention, but this one may have been better left for legislators.