The ATS and Extraterritoriality, Part II: Universal Civil Jurisdiction and Choice of Law

The ATS and Extraterritoriality, Part II: Universal Civil Jurisdiction and Choice of Law

[Anthony J. Colangelo is an Assistant Professor of Law at SMU Dedman School of Law]

I summarized in a previous post my arguments that the presumption against extraterritoriality should not apply to the ATS to the extent courts use international law incorporated into U.S. common law as the rule of decision. The presumption was raised explicitly by the brief of the UK and Dutch Governments in Kiobel and will likely be raised again.

This post addresses three discrete but related issues that may arise going forward:

1. Whether the ATS’s jurisdictional character alters the application of the presumption against extraterritoriality;
2. Whether “universal civil jurisdiction” is sufficiently recognized under international law—an issue that seemed to get attention at oral argument based on Chevron’s amicus brief; and
3. Choice of law, including as to corporate liability.

I’ll address each issue in turn, though I’ll say at the outset that I will also try to tie them together to open up what might be a new route for corporate liability grounded in an old legal discipline historically included as part of “the law of nations”; namely, private international law. Some of these preliminary thoughts will be elaborated and bolstered by other arguments in an amicus brief Anthony D’Amato and I intend to file in support of neither side.

1. A Jurisdictional Statute

First, does the ATS’s character as a primarily jurisdictional statute alter the applicability of the presumption against extraterritoriality? My guess is that it does not, at least as to U.S. domestic common law principles. While the presumption traditionally has been applied to substantive statutes, if the ATS is construed to authorize courts to project U.S. domestic common law principles of liability into foreign territory the rationales behind the presumption have equal (if not greater) force. Again, its motivating rationale was to avoid discord resulting from “clashes between our laws and those of other nations” inside foreign territory. Only here it would be courts, not Congress, crafting “our laws”—potentially amplifying arguments for restrictive constructions of judicially (as opposed to politically) generated legal principles. So I would not hang my hat on the jurisdictional distinction in that respect. (In fact, I argue in a forthcoming article that foreign defendants may even have due process objections to the application of purely U.S. domestic law to entirely foreign conduct.)

Yet the statute’s jurisdictional character may prove important for another reason. Unlike, say, securities or antitrust laws, the question is not whether U.S. law applies to the foreign conduct, but instead which law applies to the foreign conduct. In other words, if the Securities Exchange Act or the Sherman Act is construed not to reach foreign conduct at issue in a case, the claim is dismissed. But because the ATS is a jurisdictional statute that draws from the law of nations, which encompasses both public and private international law, it invites choice-of-law analysis. And under such analysis there may still be potential for corporate liability, as I explain in my third point below.

2. Universal Civil Jurisdiction

Before we get there, the second issue is “universal civil jurisdiction.” This is likely to be a major issue in the case. Both the Chevron brief and the UK and Dutch Governments’ brief argue that universal civil jurisdiction—that is, jurisdiction where the plaintiff, the defendant, and the harm are foreign—is contrary to international law.

How this issue is framed will bear heavily on its resolution. An initial question is whether (a) it’s an issue of adjudicative jurisdiction, or (b) it’s an issue of prescriptive jurisdiction. In this respect, the ATS’s location at the intersection of public and private international law is what makes the statute so interesting and complex. However it is framed, I suspect the outcome will turn largely on the default rule the Court adopts, that is, whether the Court views jurisdiction as precluded unless shown otherwise or allowed unless shown otherwise.

(a) Adjudicative jurisdiction. As an issue of adjudicative jurisdiction, there is nothing exceptional about U.S. courts entertaining civil suits involving foreigners for conduct abroad. And it is certainly not a violation of international law. In fact, just the opposite: entertaining such suits was historically viewed as a matter of international justice under private international law. In this connection, the authorities suggest that U.S. courts were required to open themselves to foreigners, even for claims arising abroad. To close themselves to foreigners would have amounted to a denial of access to justice and discrimination contrary to the law of nations.

(b) Prescriptive jurisdiction. The main concern is really about prescriptive jurisdiction. All agree that international law does not disallow universal jurisdiction. The idea being, on my view anyway, that it is not really extraterritorial jurisdiction but the decentralized enforcement by states of an international law that covers the globe. In private international law terms, there is a “false conflict” of laws. As discussed already by John Knox and Jordan Paust, this tends to erase or dilute the reasonableness constraint. According to Empagran, which involved the application of U.S. domestic antitrust (not international) law, reasonableness “helps the potentially conflicting laws of different nations work together in harmony.” But if the laws don’t conflict because universal jurisdiction violations are prohibited everywhere, reasonableness loses much of its bite. In any case, since all agree that universal criminal jurisdiction exists in international law, the argument against “universal civil jurisdiction” hinges principally on “civil.”

There are at least two weaknesses in the argument against universal civil jurisdiction, one based in international law and the other in the prevailing methodology for discerning the law of nations under the ATS. The first is that the distinction between civil and criminal jurisdiction does not appear in other bases of international jurisdiction, and indeed states generally have broader jurisdiction in civil cases under choice-of-law methodologies than in criminal cases, which tend to anchor jurisdiction more closely to geographic territory. I also suspect that extraterritorial criminal jurisdiction is generally regarded as more of an interference with other nations’ sovereignty than civil jurisdiction (hence states will not apply each other’s penal or tax laws). Carving out a separate requirement especially for universal civil jurisdiction thus appears inconsistent with the rest of international law of jurisdiction. One could come back and say that there is state practice for civil jurisdiction under other heads. But I’m not sure that’s always true or required, or that demanding it for universal jurisdiction now doesn’t invent a brand new requirement that conveniently insulates defendants from liability for the worst of the worst “tort[s] … in violation of the law of nations.”

As noted, a lot is likely to depend on what the default rule is, and the default that universal jurisdiction should be treated like other bases of jurisdiction, which generally don’t distinguish between criminal and civil, seems equally if not more reasonable than a contrary default rule that would uniquely require a showing for civil cases in particular. Indeed, this might be especially so in the ATS context given that torts and crimes were historically linked and the access to justice requirements noted above, not to mention modern human rights guarantees to an effective remedy. Nor should complaints by foreign governments that naturally don’t want to see their domiciliary defendants sued in U.S. courts be dispositive; they complain about extraterritorial jurisdiction even when it’s not universal and such complaints just as easily can extend to criminal cases as well. Finally, the argument that criminal jurisdiction is different because there is prosecutorial discretion is not so much about foreign sovereign interference as it is about domestic separation of powers, and I address it in that context below.

The second weakness is that singling out the “civil” aspect of universal jurisdiction is inconsistent with the current methodology as it appears to have been blessed by the Supreme Court for discerning the law of nations under the ATS. Modern ATS cases rely overwhelmingly on international criminal law for the law of nations under the statute. The Supreme Court itself cited a criminal case about piracy to demonstrate the “historical paradigms” that inform ATS inquiries under “the present-day law of nations.” As the Chevron brief explains: “international criminal treaties are a primary basis for ATS civil causes of action. Without their underpinning, the remaining sources typically relied on in ATS cases … could not plausibly suffice to support causes of action related to the international criminal law treaties’ subject matter.” (citing Sosa).

Of course, the criminal law sources themselves generally don’t provide for civil liability. But ATS cases rely on them. Again, a reasonable default is that the methodology should be consistent for international jurisdictional law, which clearly establishes universal jurisdiction on “the current state of international law,” to quote Sosa. The treaties are plain in this regard. Sosa also supports this view. It cited not only a criminal case about piracy, the original universal jurisdiction offense, but also Filartiga’s famous statement that “[F]or purposes of civil liability, the torturer has become—like the pirate and slave trader before him—hostis humani generis—an enemy of all mankind.” (emphasis added). In one paragraph, Sosa thus connected criminal and civil universal jurisdiction. Factually, Filartiga involved a foreign plaintiff, foreign defendant, and foreign harm. Legally, the invocation of an “enemy of all mankind” encompasses both a substantive prohibition and a jurisdictional principle that all states can enforce the prohibition. (Indeed the jurisdictional component may have been more important than the substantive component at the time of the ATS’s drafting.)

In sum, on the prevailing methodology for discerning the law of nations under the ATS, which relies overwhelmingly on international criminal law, the ATS must encompass universal jurisdiction. If we’re to be consistent, the argument that universal civil jurisdiction is precluded because it is unique is in tension with current ATS methodology; the whole ATS is unique.

This may not persuade a number of the Justices, but as a matter of principled decision-making it ought to be taken into account. At the very least, the burden should be on those who would urge the Court to abandon the established methodology for discerning the law of nations under the ATS instead of on those who might use precisely that methodology to show universal jurisdiction exists over especially serious violations of international law.

As alluded to, the distinction likely to be used will be to say that criminal jurisdiction is different because the state, as prosecutor, brings the case (something that is not always true outside the United States). Thus, there is executive discretion. See e.g., Chevron brief (“Civil actions are broadly enforceable by individuals rather than by the government, which exercises political discretion in enforcement.”). In a sense, this just restates the basic question in the case: did the political branches, in passing the ATS, authorize U.S. courts to entertain these types of suits? Yet even if we were to assume the criminal-civil distinction makes a difference in this context, the argument may do itself in. Its concession that universal jurisdiction is acceptable if the executive endorses it suggests that if the Obama administration endorses the extraterritoriality of the ATS—perhaps uncomfortable, but certainly not impossible—these arguments against universal jurisdiction vanish on their own terms.

3. Choice of Law

If the Court permits ATS suits for conduct involving foreigners abroad, it raises the follow-up question of what happens when the law of nations runs out? And more specifically, can there be corporate liability (assuming international law itself doesn’t provide for such liability)? If U.S. courts apply U.S. domestic law principles to foreign corporations inside foreign nations, there could be a conflict of laws or exorbitant jurisdiction, defeating the rationale for allowing the suit.

An option here would be resort to private international law or conflict of laws—a field especially germane for ATS cases since it has been dealing with foreign torts for centuries. This field was considered part of the “law of nations” in early Supreme Court decisions and, as noted above, was designed to ensure access to justice. Joseph Story explained that “The common law of both [the United States and England] has been expanded to meet the exigencies of the times, as they have arisen; and so far as the practice of nations, or the jus gentium privatum, has been supposed to furnish any general principle, it has been followed out with a wise and manly liberty.” The question is whether the common law should continue to “expand[] to meet the exigencies of the times” to permit redress for human rights abuses abroad.

Through a species of what is called dépeçage, or the splitting up of the claim, international law creates the prohibition and liability generally, and when it runs out, the law of the place of the harm or the party’s domicile provides liability specifically as to the corporate entity. Another way to put this in conflict terms would be that international law supplies the “conduct regulating” rule while the foreign law of the party’s domicile supplies the “loss-allocating” rule, an “established principle” or technique of the modern “jus gentium privatum.” Thus instead of using forum (U.S. domestic) law, the court would use foreign legal principles governing the corporation. On that scission of the claim, we can still be relatively sure there’s no true conflict of laws. Moreover, foreign nations’ complaints lose force if it’s their law that’s being applied—again, something that’s been happening in both U.S. and foreign courts for centuries. To tie the point back to the traditional justice requirement that our courts be open to foreigners, it cannot be that U.S. courts are precluded from holding foreign actors liable for foreign harms under foreign legal principles. U.S. courts have been doing so since the founding as the inveterate field of conflict of laws manifestly demonstrates.

The question then becomes whether the ATS authorizes such a choice of law. One way would be for the federal common law that gives rise to the cause of action to simply adopt the rule of the foreign jurisdiction. This would be consistent with how the traditional approach to private international law worked under the First Restatement and the idea of transitory torts. And it also fits to some extent within the law of nations’ “more pedestrian element” at the time of the ATS’s drafting which, according to Sosa, “did fall within the judicial sphere, as a body of judge-made law regulating the conduct of individuals situated outside domestic boundaries and consequently carrying an international savor,” albeit as a conflict technique as opposed to, say, the law merchant. To quote Justice Cardozo:

A foreign statute is not law in this state, but it gives rise to an obligation, which, if transitory, ‘follows the person and may be enforced wherever the person may be found’ … ‘No law can exist as such except the law of the land; but … it is a principle of every civilized law that vested rights shall be protected’. … The plaintiff owns something, and we help him to get it.

This would probably require pleading and proving the foreign law, which I leave to another day.

Print Friendly, PDF & Email
North America
Notify of

Anthony: I am in general agreement re: your part 2.  However, on line 2 in your intro you mention “courts use international law incorporated into U.S. common law.”  The ATS incorporates international law by express reference, not common law, as the substantive law that, when violated, produces the wrong. Filartiga, D.Ct.
And more generally international law has never been mere common law.  See, e.g., 14 U.C. davis J. Int’l L. & Pol’y 205, 219 n.42 (2008), available here:  and Paust, International Law as Law of the United States 7-10 (2 ed. 2003).

John C. Dehn

Great post, Anthony, though I suspect you know from our many collegial exchanges that I disagree with you.  For some of my response, please see my comment to Eugene’s excellent contribution immediately below.  To expand on those comments, I do not see criminal and civil jurisdiction as coextensive under international law.  States must implement public international law in their domestic law before they may “punish” or otherwise provide a remedy violations of that law.  Implementation, in this context, can and often does include incorporation by reference (as in the War Crimes Act and Uniform Code of Military Justice UCMJ).  But doing so is an exercise of prescriptive/law-making power unless the relevant body of law provides not only the conduct regulating norm but also the specific remedy.  The concept of universal criminal jurisdiction entails an exercise of both adjudicatory and prescriptive jurisdiction because its implementation necessarily makes conduct an offense against a sovereign.  In my view, the ATS is similar.  As the Sosa and Rio Tinto courts found, it authorizes the creation of a federal remedy for a law of nations violation — one that does not independently exist in international law.  This, too, is therefore an exercise of prescriptive jurisdiction, allowing… Read more »


Ex parte Quirin is much better on what the 1916 Articles of War (now 10 U.S.C. Sec. 818) did — incorporated the laws of war by reference as offenses against the law of the United States — see also In re Yamashita.
And Ex part Qurin basically noted that the 1916 Act was the first, but that the S.Ct. (and others) had used the law of war directly for various purposes.  See also Henfield’s case (1793) and FM 27-10, The Law of Land Warfare paras. 505 e and 507 a (1956).

Anthony Colangelo
Anthony Colangelo

Thanks for these comments. On conceptualizing the relationship between international and U.S. law, the important point for me is that however it is conceptualized, courts use international law as the rule of decision. On the remedy issue, that is not a matter of prescriptive jurisdiction but a matter of forum law, or the lex fori, under the law of nations. 

Benjamin Davis
Benjamin Davis

“Through a species of what is called dépeçage, or the splitting up of the claim, international law creates the prohibition and liability generally, and when it runs out, the law of the place of the harm or the party’s domicile provides liability specifically as to the corporate entity.”

SHould the law of the place of the harm or the party’s domicile be the one providing the liability?  If the law of the place of the harm is domestic law in either case and those domestic laws shield corporations from liability in these settings, it does not seem to me that it follows that the question would end there.  I can imagine a US court finding those approaches repugnant and preferring to see liability as a forum issue.  Alternatively, if an international law rule develops from the “general principles” rules type I-law – wouldn’t that also oust at least corporate liability immunizing rules that might be outliers in the law of the place of the harm or the law of the domicile of the corporation.  We are, after all, not talking about selling widgets here but about the highest level of peremptory norms.

There is depecage and there is depecage.


John C. Dehn

Anthony, while I agree that remedies are forum law in strictly private law cases, I am not sure that is how we should characterize the remedial right and remedy permitted by the ATS.