The Oral Argument in Cisco
April 30, 2026
Editor’s Note: This article also appears in Just Security.
On Tuesday, the Supreme Court heard oral argument in Cisco Systems, Inc. v. Doe, a case testing whether claims for aiding and abetting human rights violations may be brought in federal court under the Alien Tort Statute (ATS) and the Torture Victim Protection Act (TVPA). The defendant Cisco, a U.S. company, allegedly designed and built a surveillance system for the Chinese government knowing that it would be used to persecute practitioners of Falun Gong, a religion banned in China. Practitioners sued Cisco under the ATS, and its former CEO under the TVPA, for aiding and abetting the Chinese government in acts of torture, extrajudicial killing, and other serious human rights violations. (Disclosure: I wrote an amicus brief supporting the plaintiffs in Cisco, though the oral argument did not touch on the issues addressed in that brief.)
The ATS gives federal courts subject matter jurisdiction over “any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States.” It was part of the Judiciary Act of 1789 that established the federal courts. In Sosa v. Alvarez-Machain (2004), the Supreme Court recognized an implied cause of action for claims “based on the present-day law of nations” if those claims “rest on a norm of international character accepted by the civilized world and defined with a specificity comparable to the features of the 18th-century paradigms” the First Congress was familiar with. Those paradigm torts are violations of safe conducts, infringement of the rights of ambassadors, and piracy—referred to at oral argument as “the Blackstone three” (p.6) because William Blackstone identified them in his Commentaries as the principal offenses against the law of nations. Although the claim alleged in Sosa (arbitrary detention) failed the test, the Court cited with approval lower court decisions recognizing claims for torture and extrajudicial killing under the ATS.
The TVPA, enacted in 1992, creates an express cause of action against anyone who “subjects an individual” to torture or extrajudicial killing under color of foreign law. The Supreme Court has held that the TVPA cannot be used to sue corporations. But one of the plaintiffs in Cisco—a U.S. citizen unable to sue under the ATS—brought a TVPA claim against Cisco’s former CEO John Chambers.
The oral argument ran for nearly two hours and focused mostly on the ATS. Some Justices focused on whether the Court should overrule Sosa by limiting the ATS cause of action to the Blackstone three. Others seemed concerned about adopting a categorical rule and wanted to know if the aiding and abetting analysis should instead proceed norm-by-norm. The Justices also probed whether aiding and abetting in general, and the claims against Cisco in particular, raise foreign policy concerns.
Chief Justice Roberts and Justice Barrett seemed reluctant to overrule Sosa, but it is hard to predict just where the Court will land. With the Justices expressing so many cross-cutting views, writing an opinion that five can join may be difficult.
Whether to Overrule Sosa
Justice Thomas began the questioning by asking Kannon Shanmugam, counsel for Cisco, whether there is “any cause of action that Sosa would permit” (p.5). Mr. Shanmugam replied that violations of the Blackstone three would be actionable, as well as any cause of action subsequently recognized by Congress.
“So you’re asking us basically to overrule Sosa,” Justice Sotomayor interjected a few minutes later (p.16). Mr. Shanmugam suggested Sosa’s language about leaving the door ajar for new claims was dictum, so that the Court would not technically have to overrule that decision. But Cisco’s brief does indeed ask the Court to overrule Sosa if “the Court view[s] Sosa’s statements about the possibility of new causes of action as rising to the level of a holding.”
Chief Justice Roberts seemed skeptical. “[U]sually when we overrule a past decision it’s because we think it was wrong,” he observed to Curtis Gannon, who argued for the United States as amicus curiae supporting petitioners. “[A]nd we’re not saying that the Sosa decision was wrong when it interpreted the intent of the First Congress” (p.55). Mr. Gannon responded that it was only the part of Sosa that left open the possibility of recognizing new causes of action that was questionable, though he conceded that “in the abstract, Congress might have expected that the Law of Nations would evolve” (p.56). (I have written about that expectation here.)
Justice Gorsuch, on the other hand, was not sure “why we would need to overrule anything about Sosa.” “We said there that we had no basis to suspect Congress has any examples in mind beyond those three Blackstone torts,” he recalled. “[A]nd we said that future developments might preclude federal courts from recognizing new causes of action” (p.66). Mr. Gannon agreed that the Court’s subsequent holdings in ATS and Bivens cases could constitute such preclusive developments.
Justice Barrett, however, seemed to think such an approach would be too facile. In a question to Paul Hoffman, representing the respondents, she observed that “the puzzle of this case” is what to do in “the post-Erie world when the Alien Tort Statute was enacted in the pre-Erie world” (p.99). Courts don’t recognize causes of action based on general common law anymore. What should the Court do, she asked, with “a jurisdictional statute giving federal courts jurisdiction over something that we say courts generally lack the power to do?” (pp.99-100).
Mr. Hoffman answered that Sosa had considered “the exact point that you made about Erie” (p.100). He seemed to be thinking of Justice Souter’s classic line in Sosa: “We think it would be unreasonable to assume that the First Congress would have expected federal courts to lose all capacity to recognize enforceable international norms simply because the common law might lose some metaphysical cachet on the road to modern realism.”
Whatever one thinks of the result in Sosa, it is precedent. And stare decisis is supposed to be strongest in statutory cases because Congress can always amend a statute if it disagrees with the Court’s interpretation. Near the end of the argument, Mr. Hoffman delivered what I think is the strong argument against overruling Sosa: “Congress can withdraw this authority as it was interpreted by this Court anytime it wants” (p.119). Congress has not only declined to do so but has also passed the TVPA to endorse and supplement the ATS.
Categorical Rules
The Court spent a surprising amount of time at oral argument on how to conceptualize aiding and abetting liability. Justice Kagan asked whether aiding and abetting liability might be a “norm-by-norm inquiry” (p.10), available for some violations of international law but not for others, an approach Judge Bumatay suggested when dissenting from the Ninth Circuit’s denial of Cisco’s petition for rehearing en banc. Mr. Hoffman “urges one categorical rule, you urge another categorical rule,” she told Mr. Shanmugam. But “the more natural way of thinking about these questions [is to] take a look at the norm, is the norm covered first? And if the norm is covered, what’s the scope of liability, secondary liability, that’s attached to that norm?” (p.12).
Justice Sotomayor soon picked up “Justice Kagan’s point that what aids and abets an international law [violation] is specific to each one” (p.18). Justice Jackson also suggested that, if “an underlying cause of action meets the Sosa test, then you might have aiding-and-abetting liability depending upon the historical circumstances of aiding and abetting with relationship to that norm” (p.35). To adopt a categorical ban against aiding and abetting claims, she later suggested, would be inconsistent with “the language of the ATS itself” and with the First Congress’s intent, “because aiding-and-abetting liability … was widely recognized as part of the Law of Nations … at around the time of the founding” (pp.78-79). And Justice Barrett thought it was “a little odd” to apply a categorical rule on aiding and abetting to a jurisdictional statute like the ATS (p.31).
Justice Barrett also seemed attuned to other complexities that might preclude a categorical approach. She noted scholarship by Professors Anthony Bellia and Bradford Clark suggesting that the ATS was not limited to the three offenses that Blackstone identified and was intended to cover “violence committed by an American citizen against a foreign national” (p.30). She also asked both Mr. Shanmugam and Mr. Gannon about “other forms of secondary liability, like civil conspiracy” (p.14). The Fourth Circuit recently upheld a $42 million verdict in an ATS case for civil conspiracy, and neither lawyer wanted to take a position on that question.
Both of them maintained, however, that a categorical bar against aiding and abetting liability would be appropriate based on foreign policy concerns. In recent decisions, the Supreme Court has read Sosa to require a two-step analysis—first, whether the human rights norm is as generally accepted and specifically defined as the Blackstone three and, second, whether prudential considerations weigh against recognizing a cause of action. Mr. Shanmugam argued that “a norm-by-norm analysis” was not required at the second step “because the foreign policy concerns … would be equally applicable regardless of the norm that is at issue.” “In each of these cases,” he continued, “you’re talking about a claim that would require a federal court to find a primary violation of international law based on the conduct of a foreign sovereign on its own soil” (p.13). Mr. Gannon agreed, pointing out that the Court had adopted a categorical rule against suing foreign corporations in Jesner v. Arab Bank (2018) based on foreign policy concerns (p.43).
Foreign Policy Concerns
Perhaps unsurprisingly, the lawyer representing the United States emphasized foreign policy concerns. Mr. Gannon said that most aiding and abetting cases involve “allegations that somebody has aided and abetted a primary violation that was happening in a foreign country often by state actors because many of the norms … need to be done by state actors.” As a result, he argued, “the entire case is parasitic on having to prove that foreign government officials engaged in serious human rights violations in their own countries” (p.43).
Mr. Gannon later cited as an example a case in which multinational companies doing business in South Africa were sued for aiding and abetting apartheid. “The South African government expressed concerns,” he said, “that having these cases litigated in the United States … was inconsistent with the Truth and Reconciliation Commission” that country had established (p.72). Mr. Hoffman later rebutted this example. “The South African government was complaining because three of its corporations were being sued,” he noted. “As soon as those [defendants] were dismissed, the South African government filed something with the Court saying we don’t have any problem with you … considering the liability of your own corporations” (pp.95-96).
Hoffman also distinguished Jesner as a so-called “foreign-cubed” case, involving claims by foreign plaintiffs against foreign defendants based on conduct in a foreign country. “[W]e are talking in this case,” he noted, “about the actions of a United States citizen … that has committed actions on U.S. territory that have … led to the harms that are covered by this statute” (pp.102-03).
Some of foreign policy discussion revolved around footnote 21 of Sosa, in which the Court raised the possibility of “case-specific deference to the political branches.” When the U.S. government files a statement of interest objecting to a case on foreign policy grounds, Sosa said, “there is a strong argument that federal courts should give serious weight to the Executive Branch’s view of the case’s impact on foreign policy.” Justice Kavanaugh asked Mr. Gannon whether “lower courts [are] sufficiently paying attention to executive branch statements” (p.68). Mr. Gannon was “not aware of recent cases where we’ve filed what I would call a case-specific concern that has been disregarded by the lower courts.” In Cisco, he noted, the United States “didn’t file individual statements.”
Justice Jackson later picked up on this, questioning the need for a categorical rule against aiding and abetting claim because “the United States can protect its interests if it does make a statement that this particular claim is a problem” (p.82). There might be cases in which a foreign government doesn’t care if a U.S. company is being sued, she suggested. “So why would Cisco be absolved and the plaintiffs here not get a remedy on the basis of some speculation about a foreign policy concern that the United States is not even willing to say in writing right now would actually occur?” (p.83). “Sometimes,” Mr. Gannon replied, “having to take a position about an individual case has its own foreign policy consequences” (p.83).
The TVPA
The Justices spent significantly less time on aiding and abetting liability under the TVPA, and the arguments largely tracked the parties’ briefs. Mr. Shanmugam argued that dictionary definitions of “subjects” “require inflicting some action on the victim” rather than “merely facilitating something” (p.25). Justice Sotomayor asked about “the person who’s in the room with the torturer and hands him the instruments” (p.25). Mr. Shanmugam responded that such a person could be held “primarily liable” (p.25), but Justice Sotomayor seemed unconvinced.
Mr. Shanmugam also relied heavily on the Supreme Court’s decision in Central Bank of Denver v. First Interstate Bank of Denver (1994), which rejected civil aiding and abetting liability for securities fraud. Central Bank came “pretty close,” he argued, to adopting “a magic words requirement” that Congress must use the words “aiding and abetting” to create civil liability (p.29).
Mr. Hoffman disagreed. “In Central Bank itself,” he noted, “the Court was rejecting a presumption in favor of aiding-and-abetting liability and rejecting it. It wasn’t creating a presumption against aiding-and-abetting liability” (p.105). He also pointed out that Congress “didn’t know about Central Bank in 1992 [when the TVPA was passed] because Central Bank wasn’t decided” (p.106).
Finally, Mr. Hoffman pointed to a passage in the Senate Report stating that the TVPA provides a cause of action “against persons who ordered, abetted, or assisted in the torture.” “You can’t get any plainer than that from Congress” he observed (p.106).
Conclusion
As I have written previously, there appear to be four votes on the Court to overrule Sosa and limit the ATS cause of action to the Blackstone three. Justices Thomas, Gorsuch, and Kavanaugh said as much in Nestlé U.S.A. v. Doe (2021), and Justice Alito seemed to be of the same mind. Nothing in the oral argument makes me think that any of them will change position.
But Chief Justice Roberts and Justice Barrett did not seem similarly inclined. Perhaps the Court will embrace as a compromise the norm-by-norm analysis that Justice Kagan championed and that Justices Sotomayor, Barrett, and Jackson seemed open to. Such a compromise could avoid “giv[ing] the green light to U.S. corporations acting from the United States to help foreign governments commit torture or extrajudicial killing,” as Mr. Hoffman put it in his opening statement (p.86).
I expect Cisco to be one of the last decisions handed down this Term. The issues are complex, and multiple separate opinions are likely. As Steve Vladeck recently pointed out, hearing cases in April leaves the Court just two months to finish their work. There is a lot to sort through in Cisco and precious little time for sorting.