The following contribution to our Kiobel v. Royal Dutch Petroleum symposium is written by Eugene Kontorovich, a professor at Northwestern University School of Law, where he teaches courses in constitutional law, federal courts, and international law. This year he is a fellow at the Institute for Advanced Study and the recipient of the Federalist Society’s Bator Award. He has written extensively about the ATS, extraterritorial jurisdiction, and related constitutional issues, including the forthcoming Discretion, Delegation, and Defining in the Constitution’s Law of Nations Clause, and Implementing Sosa v. Alvarez-Machain: What Piracy Teaches About the Limits of the Alien Tort Statute. [Lyle published an introduction to the issues in Kiobel last week.]
Normally, statutes are not thought to apply in foreign territory, unless the language or context of the law demands extraterritorially application. The Supreme Court has strongly reaffirmed this “presumption against extraterritoriality,” in Morrison v. National Australia Bank LLC (2010). One would think that, if anything, the presumption would be stronger in ATS cases, many of which have no connection whatsoever to the U.S., as the presumption typically applies to laws governing conduct that has significant effects on the U.S., but merely takes place abroad. However, the Kiobel plaintiffs argue that this presumption should not apply to the ATS. Rather, the statute’s reference to “the law of nations” should be taken as an explicit rebuttal of the presumption. Since many violations of the law of nations take place extraterritorially (by definition, in the case of piracy), their argument goes, the presumption cannot apply.
Two reasons support the application of the canon of interpretation against extraterritoriality to the ATS. First, precedent and policy show that the presumption applies in full to international law offenses – even to piracy. Second, there is another jurisdictional presumption that applies to the ATS, though not one that has thus far featured in the litigation. Federal courts generally read broad jurisdictional statutes like the ATS as creating less jurisdiction than the maximum the text would support. This is true even when the jurisdictional statutes involve offenses against the law of nations.
The Supreme Court has made clear that the fact that a statute deals with matters of international law does not mean that it automatically up ends presumptions against extraterritoriality. (I develop this point at greater length a forthcoming essay, A Tort Statute, With Aliens and Pirates.) For example, in United States v. Palmer (1818), Chief Justice Marshall read a statute criminalizing “piracy” by “any person” as requiring a U.S. nexus, even though it was clear that Congress could constitutionally apply it universally. A presumption is only valuable if it is usually right. The anti-extraterritoriality rule stops making sense if one thinks that by invoking international law norms, Congress usually intends to use the full extent of the jurisdiction allowed to it by international law. The evidence goes in the opposite direction. Recent statutes of the United States do not extend universal jurisdiction to genocide or war crimes though they could do so under international law. They do apply extraterritorial jurisdiction to torture, but only by making the foreign application explicit. In short, the universal cognizability of an offense is relevant to Congress’s constitutional ability to regulate it, but not necessarily relevant to whether that power has been maximally exercised.
One might suggest that the policies behind the anti-extraterritorial presumption – such as avoiding conflicts with foreign laws – do not apply to the ATS. Under the ATS, it is international law that is to be applied. This is the same everywhere – what can be the conflict? First, international law is silent as to the method of its enforcement, and particularly on the question of penalties. With the exception of the U.S., no country has civil remedies for extraterritorial torts, and several have filed briefs protesting such litigation under the ATS. Even in international criminal law, punishment varies massively across jurisdictions even for identical conduct. Moreover, ATS cases are not pure international law. Certainly the punitive damages conflict with other legal regimes comes not just from penalizing what they choose to legalize, but also from penalizing to different degrees. (Indeed, this was the essence of the Court’s argument for federal preemption of Arizona’s immigration laws.) The Kiobel plaintiffs urge the Court to borrow important rules of decision – such as corporate liability – from U.S. domestic law, not international law.
But avoiding conflicts is only one of the reasons for the anti-extraterritoriality presumption. Another is avoiding the application of U.S. jurisdiction to conduct that only marginally involves U.S interests. On this score, reading a U.S. territory requirement into the ATS is consistent with its purpose. The statute was designed to give an avenue of redress to aliens aggrieved by law of nations violations for which the U.S. might be held responsible by foreign powers. The ATS’s reference to international law does not untether it from American interests. Consider by analogy the Ambassadors Clause in Article III of the Constitution, which provides that the Court will have original jurisdiction over suits involving ambassadors, and does not textually limit the class of ambassadors involved. Yet when a U.S. Ambassador – that is, an emissary from Washington, not to it – sought a trial before the Court, he was dismissed out of hand, in one short paragraph. The clause (or perhaps just the similarly worded statute conferring the jurisdiction) only applies to foreign ambassadors because that is the purpose of the jurisdiction: to avoid offending other countries.
Besides the presumption against extraterritoriality, an additional canon of statutory construction applies to the ATS, and indeed applies with even greater force. The ATS is a jurisdictional statute. Federal courts have traditionally interpreted jurisdictional statutes narrowly, excluding many cases that fall within the plain reading of the text. This is so even when the statute seems to recapitulate the relevant constitutional provision, which one might think would suggest that Congress wanted to confer jurisdiction to the constitutional maximum. Famous examples include the diversity statute, which like the corresponding Article III provision, simply requires a plaintiff to be from a different state than the defendant. Yet in 1806, the Supreme Court in Strawbridge v. Curtis read into the statute a requirement of complete diversity, where all plaintiffs must be diverse from all defendants. Some scholars have suggested that the complete diversity requirement actually excludes the great majority of otherwise proper diversity suits. Similarly, the statute giving federal courts jurisdiction of any suits “arising under” federal law parallels the relevant provision in Article III of the Constitution. Yet the Court has consistently interpreted the statute as being narrower than the constitutional provision, with the statute generally requiring a federal cause of action while Article III only requires some federal “ingredient.”
The presumption of narrowly construing statutes applies in full and perhaps a fortiori to jurisdictional grants incorporating international law or touching on foreign relations. “Cases affecting Ambassadors” fall not just within federal jurisdiction, but also within the original jurisdiction of the Supreme Court. This suggests the Framers regarded these cases as of the greatest importance, and did not want Congress to pawn them off on inferior tribunals. Moreover, who qualifies as an “ambassador” is a question governed at least in part by international law.
Yet jurisdiction over such cases has been interpreted narrowly. Take one example greatly relevant to the ATS. Offenses against ambassadors were one of the three paradigm offenses for which the drafters of the ATS sought to provide a tort remedy. Moreover, the Constitution gives the Supreme Court jurisdiction over all suits “affecting Ambassadors.” Yet the Supreme Court has held in United States v. Ortega, that even an assault on an ambassador, though in gross violation of international law and well within the literal meaning of the original jurisdiction statute, does not “affect ambassadors” within the meaning of the jurisdictional statute. Nor do federal courts have any jurisdiction over divorce suits by ambassadors, though few things could “affect” them more. Thus, jurisdictional statutes are read narrowly even when they involve the very offenses that the Court in Sosa said lay at the heart of the ATS.
The reasons for narrowly interpreting even the most important jurisdictional statutes apply in full force to the ATS. One such reason is docket control. Courts tend to interpret the outer bounds of Article III jurisdiction broadly, as mistakes in this regard are hard to reverse. Yet a jurisdictional statute interpreted too narrowly can be corrected by Congress. Moreover, constitutional limits are described in vague terms in accord with the text’s succinctness and generality, and a desire not to exclude any case to which federal power might properly extend.
With jurisdiction, the emphasis switches to precise limits. Certainly the notion of “violation of the law of nations” is as vague as “arising under federal law.” Indeed, the indeterminacy of the former was stipulated at the Constitutional Convention. Interpreting the statute to not extend to any case in the world fits with docket management and prioritization principles. Second, broad interpretations of statutory jurisdiction will require frequent interpretation of the vague Article III limits; this could insert serious constitutional issues into otherwise unimportant cases, where the game is not worth the candle. Finally, while constitutional rules lend themselves to general standards, precision and bright line rules are particularly important for jurisdictional statutes. Litigation on off-the-merits issues like jurisdiction is pure social loss; it is simply throwing deck chairs off the Titanic. Thus a jurisdictional rule that errs on the side of exclusion and does not invite speculative litigation is a good thing.