In its recently released decision in Al Shimari v. CACI International (4th Cir. 2014), the Fourth Circuit followed a recent trend that has emerged in alien tort statute (“ATS”) cases, post-Kiobel. Like other courts grappling with questions of subject matter jurisdiction under the ATS, the Fourth Circuit purported to apply the presumption against extraterritoriality (“PAE”) in assessing whether it had jurisdiction over the plaintiff’s ATS claims. Also like other courts, however, the Fourth Circuit’s invocation of the PAE was pretense; it instead applied an entirely different doctrine which has, at best, only a passing connection to the PAE, or at least the PAE as it existed pre-Kiobel.
The plaintiffs in Al Shimari are four Iraqi citizens who allege that CACI, a U.S. government contractor providing “interrogation services” to the Department of the Interior, violated international law by torturing and mistreating prisoners at Abu Ghraib. Following the Supreme Court’s decision in Kiobel, however, the district court dismissed their claims under the alien tort statute (“ATS”), concluding that, under Kiobel’s newly issued guidance, there was no subject matter jurisdiction to hear the case, as the alleged torts all took place extraterritorially in Iraq. The plaintiffs appealed.
Constrained by Kiobel’s dictates, the Fourth Circuit analyzed its jurisdiction to hear the suit by applying a test that it called the PAE. After reviewing CACI’s “ties to the territory of the United States,” the court concluded that the plaintiffs’ ATS claims “touch[ed] and concern[ed] the territory of the United States . . . with sufficient force to displace the presumption against extraterritorial application.”
But the judicial doctrine that the Fourth Circuit applied was the PAE in name only. The analysis it performed was a lengthy balancing test, and, absent a single, perfunctory reference to congressional intent in enacting the TVPA, contains nothing that could be described as statutory construction (slip op., at 31).
This is hard to reconcile with the court’s claim that it was applying the PAE. The PAE is a longstanding canon of construction, with a well-developed pedigree, in which courts presume that a stature regulates domestic conduct unless otherwise specified. In Kiobel, however, the Supreme Court announced that, in interpreting the jurisdictional scope of the ATS, the PAE required the Court to construe the statute in a manner that precluded jurisdiction over a foreign plaintiff’s claim against a foreign defendant for foreign conduct. The First Congress had not intended for the ATS to regulate non-domestic conduct (or so SCOTUS claimed – as I’ve discussed in prior posts, this claim is not necessarily supported by history), and, as a result, when a plaintiff brings a case in which “all the relevant conduct took place outside of the United States,” the ATS does not provide the federal courts with jurisdiction to hear it.
The true crux of Kiobel, however, is not its specific holding, but rather the Court’s deliberately ambiguous bit of dicta, in which it noted that “even where [a plaintiff’s] claims touch and concern the territory of the United States,” in order for there to be jurisdiction under the ATS the claims “must do so with sufficient force to displace the presumption against extraterritorial application.”
What “touch and concern” actually means is anyone’s guess. That include the Court itself – Kiobel was a unanimous opinion, but of the nine justices who adopted Roberts’ majority, were a minimum of four different (and contradictory) opinions on what the phrase was intended to signify.fn1
But the lower courts, left with the unenviable task of trying to apply this cipher, have largely done so by simply discarding Kiobel’s purported concern with territory, and replacing it instead with a concern for state responsibility for internationally wrongful acts. In effect, recent ATS cases have been reading Kiobel to say, “[E]ven where [an ATS claim] touch[es] and concern[s] the [international responsibility] of the United States, [it] must do so with sufficient force” to render the case justiciable in a U.S. court.
This re-written version of Kiobel was applied by the Fourth Circuit in Al Shimari to conclude that when a plaintiff sues a U.S. government contractor, alleging that the government contractor committed acts of torture in a foreign territory, the plaintiff’s claim is not extraterritorial at all. To explain this incongruous result, the Fourth Circuit found that the PAE contains the additional caveat — previously overlooked by all other courts — which provides that the PAE should only be applied when doing so serves the policies that the PAE was intended to promote:
[W]e observe that mechanically applying the presumption to bar these ATS claims would not advance the purposes of the presumption. A basic premise of the presumption against extraterritorial application is that United States courts must be wary of ‘international discord’ resulting from ‘unintended clashes between our laws and those of other nations.’
But that’s not how the PAE works. The PAE does not say “a U.S. statute can apply extraterritoriality so long as it doesn’t cause the discord with other states.” It is not a case-by-case balancing test.
Or at least it wasn’t prior to Kiobel. The PAE, as it was understood for the bulk of the 20th century, is used in interpreting how a statute applies to all cases, without consideration for the particularized facts of any individual suit. Although the PAE exists, in part, as a prophylaxis against international discord, its underlying policy considerations are not incorporated into its operation: a statute reaches extraterritorial conduct based on whether congress affirmatively indicated an intent for it to do so, without regard to whether the statute does or does not pose a risk of international discord. As the Supreme Court held in Morrison v. National Australia Bank, Ltd. (2009), the PAE “applies regardless of whether there is a risk of conflict between the American statute and a foreign law.”
The PAE is used to establish a statute’s geographical area of effect. Once this determination is made, the statute’s territorial boundaries are fixed; they do not flex and bend to accommodate the unique policy interests raised by an individual lawsuit. If a person commits in an act within that territorial boundary, the statute applies. If that same person commits an identical act in a different location, outside that territorial boundary, then the statute does not.
But this isn’t what the recent ATS cases have done. Post-Kiobel, the ATS’s geographical scope could not be drawn on a map, because it is given a new delimitation for every case. Just look at Al Shimari – there, the Fourth Circuit concluded that the ATS’s jurisdiction extends to torture committed in Iraq because of the following factors:
[P]laintiffs’ ATS claims “touch and concern” the territory of the United States with sufficient force to displace the presumption against extraterritorial application based on: (1) CACI’s status as a United States corporation; (2) the United States citizenship of CACI’s employees, upon whose conduct the ATS claims are based; (3) the facts in the record showing that CACI’s contract to perform interrogation services in Iraq was issued in the United States by the United States Department of the Interior, and that the contract required CACI’s employees to obtain security clearances from the United States Department of Defense; (4) the allegations that CACI’s managers in the United States gave tacit approval to the acts of torture committed by CACI employees at the Abu Ghraib prison, attempted to “cover up” the misconduct, and “implicitly, if not expressly, encouraged” it; and (5) the expressed intent of Congress, through enactment of the TVPA and 18 U.S.C. § 2340A, to provide aliens access to United States courts and to hold citizens of the United States accountable for acts of torture committed abroad.
But only one of these factors – the allegation that CACI’s U.S.-based management tacitly approved of the conduct – is concerned with territorial location in any meaningful way. And that factor provides exceedingly weak support to the Fourth Circuit’s conclusion that ATS jurisdiction does extend to the plaintiffs’ claims in Al Shimari. Under the traditional PAE, as it was understood pre-Kiobel, extraterritorial conduct is not transformed into territorial conduct on the basis of corporate activity in the United States that supported or planned the foreign acts. For instance, in EEOC v. Aramco (1991), the Supreme Court invoked the PAE to conclude that Title VII does not extend to a U.S. corporation that recruited a U.S. citizen through recruitment efforts that occurred in the U.S., when the U.S. citizen is employed to work abroad and is fired abroad. And in Morrison, the plaintiffs “contended that they seek no more than domestic application [of the securities statute], since Florida is where [the defendant] and its senior executives engaged in the deceptive conduct of manipulating [its] financial models.” But the fact that U.S.-based corporate management planned the foreign conduct was not sufficient to escape the Court’s application of the PAE to the security statute, because the critical factor “is not upon the place where the [planning of the tortious conduct] originated, but upon [where the tortious conduct occurred].”
So if the courts aren’t applying the PAE, what are they applying?
This new doctrine, the PAE-for-ATS, is a PAE in name only. Its closest doctrinal kin are not canons of construction, which are concerned with statutory interpretation, but avoidance doctrines, which are concerned with justiciability. And as used by the lower courts so far, the PAE-for-ATS has had nothing to do with interpreting the text of the ATS, and everything to do with allowing the federal court system to avoid cases which are not appropriate for it to decide. The PAE-for-ATS therefore bears a much closer relationship with the political question doctrine (“PQD”) than it does to the PAE as it was known prior to Kiobel.
The PQD, like the PAE-for-ATS, is a case-specific balancing test which categorizes cases as justiciable or non-justiciable based on the existence of another governmental body’s competing and superior interest in the case’s subject matter. In applying the PQD to cases involving military operations (such as in Al Shimari, which, in addition to finding jurisdiction under the ATS, also remanded the case to the district court to further develop the PQD issue), the courts do so by “gaug[ing] the degree to which national defense interests may be implicated in a judicial assessment” of a claim, and then decline to adjudicate those cases which are fundamentally political, not legal. Taylor v. Kellogg Brown & Root Services (4th Cir. 2011). Through use of the PQD, would-be plaintiffs in cases concerning national defense are thereby deprived of the ability to “reshape the national response to threats of hostile [ ] attack through the mechanism of tort law.” Tiffany v. United States (4th Cir. 1991). The PQD recognizes that “some questions beyond judicial competence. Where the performance of a ‘duty’ is left to the discretion and good judgment of an executive officer [or the legislative branch], the judiciary will not compel the exercise of his discretion one way or the other, for to do so would be to take over the office.” Baker v. Carr (1962).
The PAE-for-ATS operates in the same way, except that where the PQD’s concern is with decisions that have been exclusively committed to a separate branch of the federal government, the PAE-for-ATS’s concern is with decisions that have been exclusively committed under international law to a non-domestic entity. And since the U.S. has no business in adjudicating a foreign sovereign’s responsibility to an alien arising from a breach of the foreign sovereign’s international obligations, the PAE-for-ATS quite appropriately excluded the federal courts from deciding the plaintiffs’ claims in Kiobel. U.S. courts should not be deciding questions which, under international law, they lack the adjudicative jurisdiction to consider.
But a state has every right (and often, in fact, an obligation) to evaluate its responsibility for a breach of its own international obligations. And that dovetails perfectly with the reason the ATS was enacted in the first place: to ensure that the federal government had the ability to redress injuries of foreign nationals that were caused by a breach of the U.S.’s responsibilities under international law. And a state also has every right to evaluate a foreign actor’s violations of its international obligations to the adjudicating state – such as in the case of pirates, who violate the rights of all nations.
Thus, lower courts have been implicitly concluding that, under the PAE-for-ATS, cases are sorted as justiciable or nonjusticiable based on whether they “touch and concern” the international rights or responsibilities of the United States with “sufficient force” to support U.S. adjudicative jurisdiction. The court have found, for instance, that they can properly hear an ATS claim brought against a U.S. citizen who assisted, managed, and advised a foreign nation’s violations of international law (Sexual Minorities Uganda v. Lively (D. Mass. 2013)); a claim brought against U.S. citizens and a U.S. organization who organized and funded terrorist bombings in Sri Lanka (Krishanti v. Rajaratnam (D.N.J. 2014)); a claim brought by foreign nationals injured in a foreign terrorist attack against a U.S. embassy in a foreign state (Mwani v. Laden (D.D.C. 2013)); or a claim against a U.S. company and citizen that created technology used by China “to identify and torture dissidents” (Daobin v. Cisco (D. Md. 2014) (assumied without deciding)).fn1
On the other hand, the courts have found that a claim against a U.S. corporation whose foreign subsidiaries aid-and-abet a foreign state’s breaches of international law, by passively doing business in that state, is not a case which should be heard in a U.S. court, or at least not in the absence of some additional set of facts that might create a compelling link implicating U.S. responsibility (Balintulo v. Daimler (2d Cir. 2013)). Nor should federal courts be adjudicating a claim brought by a Bangladeshi plaintiff against a Bangladeshi business and citizen for torture occurring in Bangladesh also does not implicate U.S. responsibility (Chowdhury v. Worldtel (2d Cir. 2014)); a claim brought by Chinese citizens and residents against a Chinese media executive who promoted the torture of Falun Gong practioners in China (Gang v. Zhizhen (D. Conn. 2013)); a claim brought by Israeli and Canadian citizens for terrorist attacks “allegedly funded by Iran, launched from Lebanon, and target[ing] Israel” (Kaplan v. Central Bank of Islamic Rep. of Iran (D.D.C. 2013)); a claim brought by Indian citizens against an Indian political party for organizing acts of genocide, rape, torture, summary execution, and extrajudicial killing, all of which took place in India (Sikhs For Justice Inc. V. Indian National Congress Party (S.D.N.Y. 2014)); a claim of extrajudicial killing and torture brought by relatives of an Iranian citizen against the Islamic Republic of Iran and the Revolutionary Guard (Mohammadi v. Islamic Republic of Iran (D.D.C. 2013)); or a claim brought by a former Ukrainian prime minister against a Swiss corporation that allegedly bribed Ukrainian officials to have her “prosecuted on unfounded, politically-motivated criminal charges,” as revenge against the prime minister for enactment of a policy that caused the corporation commercial harm (Tymoshenko v. Firtash (11th Cir. 2014)).
Two things stand out from these cases. First, it is simply not possible to predict whether a court will find jurisdiction in an ATS case by looking at the GPS coordinate of the jail cell where alleged acts of torture occurred. Whatever canon of construction is being applied here, it is not one in which geography plays a dispositive role. It does, however, appear to be possible to predict whether a court will find jurisdiction based on whether or not the claim is based on a breach of the United States’ obligations or rights under international law.
Second, although there are only a limited number of post-Kiobel cases so far, they are already difficult (or impossible) to reconcile with one another, if they are understood to all be applying the same PAE. For instance, Balintulo — finding that ATS jurisdiction did not extend to conduct by a U.S. corporation’s foreign act — cannot be squared with either Al Shimari — finding that foreign torts are not extraterritorial where, inter alia, managers at the U.S.-based corporate headquarters “attempted to ‘cover up’ the misconduct” — or with Krishanti — finding that jurisdiction exists against a corporation that raised funds in the U.S. to support a terrorist bombing in Sri Lanka — if you assume that these cases are applying the same standard as Aramco and Morrison.
These cases are consistent with one another, however, when considered under the newly created PAE-for-ATS, as the defendants in Krishanti and Al Shimari took actions which could implicate U.S. responsibility on the international plane, while the Balintulo defendants did not.
For now, the federal courts are bound by Kiobel’s pretense that it applied the PAE, and must continue to awkwardly retrofit a PAE analysis onto a decision based on international responsibility rather than territory. But if the Court were to announce that the ATS’s version of the PAE is not the same as the traditional version, what should the standards of this new doctrine look like?
The standards applied in PQD cases would be a good place to start. Under the PQD, there are six primary factors which courts consider in determining whether a case is nonjusticiable:
Prominent on the surface of any case held to involve a political question is found  a textually demonstrable constitutional commitment of the issue to a coordinate political department; or  a lack of judicially discoverable and manageable standards for resolving it; or  the impossibility of deciding without an initial policy determination of a kind clearly for nonjudicial discretion; or  the impossibility of a court’s undertaking independent resolution without expressing lack of the respect due coordinate branches of government; or  an unusual need for unquestioning adherence to a political decision already made; or  the potentiality of embarrassment from multifarious pronouncements by various departments on one question.
Because the PQD and PAE-for-ATS serve the same principles (but differ in that first doctrine addresses only an intra-governmental division of power, while the latter addresses both intra-governmental and inter-governmental divisions), these factors would work well for evaluating ATS cases. Under the PAE-for-ATS, courts would find that:
Prominent on the surface of any case held to involve a question of foreign international responsibility not involving U.S. interests, and therefore not appropriate for a U.S. court to adjudicate, are  an international norm committing the issue to a foreign sovereign or IGO; or  a lack of judicially discoverable and manageable standards for resolving it; or  the impossibility of deciding without a foreign policy determination of a kind clearly for nonjudicial discretion; or  the impossibility of a court’s undertaking independent resolution without expressing lack of the respect to a foreign sovereign; or  a need for adherence to an international obligation of the United States which would prohibit the U.S. from adjudicating and enforcing a particular case; or  the potentiality of embarrassment from multifarious pronouncements by various branches of the U.S. government on one question.
Finally, in addition to an adapted version of the Baker factors, courts applying the PAE-for-ATS should also look at factors that support the case’s adjudication in a U.S. court. These factors are: (1) the need for adherence to an international obligation of the United States which would require (or strongly urge) the U.S. to adjudicate and enforce a particular case; (2) whether the plaintiff’s claims allege conduct by the defendants for which the U.S. could bear international responsibility; (3) whether the foreign sovereign(s) whose rights(s) are implicated by the suit have indicated their support for the adjudication of the suit by a U.S. court; (4) whether the executive branch has indicated that adjudicating the suit is consistent with the U.S.’s international obligations; or (5) the existence of an internationally wrongful act that violates the U.S.’s rights under international law, for which the U.S. is entitled to seek redress (such as in the case of pirates or terrorist acts against the U.S.).
fn1. Also left unanswered is why, in describing this new test, SCOTUS chose to adopt a phrase which was originally used as a test to evaluate covenants on the use of real property. I’m just waiting for a creative ATS defendant to argue that, by using the phrase “touch and concern,” the Court was subtly implying that ATS cases, like real covenants, can only be upheld if they are “so related to the [United States’] land as to enhance its value and confer a benefit upon it.”
fn2. Although the issue has yet to be decided, post-Kiobel, I suspect that the courts would find that “harboring” cases — cases in which, like Filartiga and Samantar, a foreign official has moved to the U.S. after committing violations of international law in his home state — are also within the reach of the PAE-for-ATS. Offering sanctuary and immunity to a violator of international law could implicate the U.S.’s responsibility under international law, depending on the specific facts of a case. For instance, whether or not the deposed dictator’s home state opposed or supported the U.S.’s adjudication of the dispute would be very much relevant to the determination. And, if the PAE-for-ATS were officially recognized for what it is, the courts could easily take this factor into account, either through letters from the executive branch, or from the foreign state itself.