Customary International Law’s Domestic Status: Reflections After Twenty-Five Years

We are grateful to Bill Dodge for highlighting our 1997 article on the domestic legal status of customary international law.  In that article, we critically analyzed what we referred to as the “modern position,” which is the claim made by some academics and the Restatement (Third) of Foreign Relations Law that customary international law has the legal status in the United States of federal common law.  That claim has two related components:  First, that customary international law is “self-executing” and thus, in the words of Louis Henkin, should be applied by courts “without any need for it to be enacted or implemented by Congress”; and, second, that this self-executing customary international law is federal law, which means (among other things) that it can create a basis for federal court jurisdiction, preempt inconsistent state law, and potentially bind the President under the Constitution’s Take Care Clause.

Critique of the Modern Position

Our critique had a number of elements.  It drew on the constitutional text, which in the Supremacy Clause mentions treaties but not customary international law, and which mentions customary international law only in its grant of power to Congress to define and punish offenses against the law of nations.  The critique also relied on constitutional structure to argue that because customary international law, unlike treaty law, is not expressly approved by domestic political actors, giving it direct domestic effect is in tension with principles of both separation of powers and federalism. This is especially so, we explained, in cases involving modern customary international law (such as in the area of human rights), which overlaps with traditional areas of domestic regulation.  We further pointed out that, in areas such as human rights, the political branches had indicated in various ways (such as through non-self-execution declarations to treaties) that they did not want this international law to be directly applied by the courts, and they had demonstrated in statutes like the Torture Victim Protection Act (TVPA) the ability to authorize its application in select instances.  Finally, we explained that the idea that the entire body of customary international law applies automatically in the U.S. legal system as federal law, regardless of its relationship to the policies set forth in enacted law, was inconsistent with the nature and limitations of federal common law as generally understood in the federal courts field.

In addition to these points about constitutional text and structure, we noted that there was essentially no judicial support for the modern position.  Although federal courts applied customary international law in the absence of congressional authorization in some cases before Erie Railroad v. Tompkins, we explained that they treated it during this period as having the status of “general common law,” not federal law.  There was some dispute among scholars on this point when we made that historical claim, but since then most scholars (including Bill) have agreed with us (and the Supreme Court in Sosa v. Alvarez-Machain expressly confirmed the point).  Importantly, general common law was not federal law and thus did not create a basis for federal court jurisdiction or preempt inconsistent state law.  While the Supreme Court since Erie has approved the development of federal common law on select topics, it has never applied customary international law as self-executing federal common law.  In the most famous Supreme Court decision involving federal common law relating to foreign affairs—Banco Nacional de Cuba v. Sabbatino—the Court pointedly declined an invitation to apply customary international law.

Alien Tort Statute Litigation

The modern position was aspirational when we critiqued it, and it has grown even more aspirational in the intervening quarter century.  Starting with the Second Circuit’s 1980 decision in Filartiga v. Pena-Irala, a few lower courts did say, in cases brought under the Alien Tort Statute (ATS), that cases involving customary international law arose under federal common law for purposes of Article III “arising under” jurisdiction.  Even those courts never embraced other aspects of the modern position, and the Supreme Court’s subsequent ATS decisions—which tied the authority to apply customary international law in ATS cases to the ATS itself, and not to the automatic incorporation of customary international law—rendered this line of decisions moot.  In part for that reason, and in part because Congress has selectively incorporated customary international law in discrete contexts, the debate over the modern position, which was once very heated, has largely expired.

Bill suggests that the Supreme Court, in agreeing to review suits between aliens under the ATS (and then disallowing them on various grounds), implicitly accepted the modern position.  This suggestion is unpersuasive.  In Sosa, the Supreme Court reasoned that the ATS “was meant to underwrite a narrow set of common law actions” and that it should be construed as authorizing the federal courts to develop causes of action for a “modest number” of modern customary international law violations.   In other words, the federal common law applied in ATS cases is supported by a congressional enactment—a position that, as we explained in our original article, is consistent with how federal common law normally works, and is a far cry from the claim of the modern position that customary international law operates as self-executing federal common law in the absence of congressional authorization.  (Lest there be any doubt on this point, the Court subsequently explained in Kiobel v. Royal Dutch Petroleum that the issue was whether Congress “intended federal common law under the ATS.”)

As we noted in a 2007 article with David Moore, the Supreme Court’s reliance in Sosa on statutory authorization for applying customary international law in the ATS context—and its emphasis on the many separation of powers reasons for “caution” in implementing causes of action under the ATS—should be seen as a repudiation of the modern position, which, again, contends that no statutory authorization is needed.  So too should the Court’s explanation that federal common law is limited to “interstitial areas” and generally depends on “legislative guidance.”  Consistent with that view, the Court cited the U.S. non-self-execution declaration to the International Covenant on Civil and Political Rights as a reason not to rely on that treaty in developing federal common law.  It also stated that the “affirmative authority” for a precisely defined cause of action in the TVPA “is confined to specific subject matter” and would not support “judicial creativity” beyond its terms.  And the Court noted the possibility of “case-specific deference to the political branches” in deciding whether to allow suits under the ATS.  All of these observations are inconsistent with the modern position.

Furthermore, as Bill notes, the Court in Sosa suggested in a footnote that claims under customary international law could not be brought under the federal question jurisdiction statute, 28 U.S.C. § 1331.  This conclusion, too, is an implicit rejection of the modern position.  If customary international law were in fact federal law for purposes of Article III of the Constitution, as proponents of the modern position claim, it is not clear why it would not also be federal law for purposes of Section 1331, at least when it serves as the basis for the plaintiff’s case.  The Supreme Court, after all, has made clearthat federal law for purposes of Section 1331 includes federal common law.  And, except for the well-pleaded complaint rule, federal question jurisdiction under Article III and Section 1331 largely track each other.

Other Developments that Undercut the Modern Position

Other aspects of Supreme Court and lower court practice are also at odds with the modern position.  For example, on separation of powers grounds the Supreme Court has grown increasingly skeptical about federal common law that is not tethered to enacted law.  For similar reasons, it has been skeptical about implied statutory rights of action, and it has refused for decades to recognize new judicially developed damage claims for constitutional violations.  Even more significantly, despite the fact that treaties are expressly approved by the President and the Senate and are mentioned in the Supremacy Clause, the Supreme Court has been very strict about the conditions under which treaties are to be applied as self-executing federal law.  It is implausible to suggest that, even though treaties are self-executing only in limited circumstances, all of customary international law—which is often based on, and substantially overlaps with, treaties, but which lacks their democratic sanction—is self-executing federal law.

Notably, in the context in which it would probably be easiest to justify applying customary international law as federal common law—foreign official immunity—courts have not done so.  In its 2010 decision in Samantar v. Yousuf, the Supreme Court held that this immunity is not regulated by the Foreign Sovereign Immunities Act (FSIA) but that it may exist as a matter of “common law.”  The Court did not suggest that this common law should involve the application of customary international law.  And in the twelve years since that decision, lower courts have not said that, in applying foreign official immunity, they should directly apply customary international law.  Many courts have instead simply applied the views of the executive branch about whether to confer immunity without even discussing international law, an approach arguably endorsed by the Court’s statement in Samantar that Congress had not shown an intent in the FSIA to eliminate “the State Department’s role in determinations regarding individual official immunity.”  As we have explained elsewhere, judicial reliance on executive branch positions in this context is inconsistent with the modern position.  Importantly, the executive branch itself treats customary international law as simply one of many considerations in deciding whether to suggest immunity, and (as Bill appears to concede) it has not indicated that it views the customary international law in this area as binding federal law.

It is not difficult to understand why presidential administrations from both parties have resisted endorsing the modern position.  Customary international law evolves and develops through state practice and argumentation, and its precise content is often uncertain and contested.  Deeming it to be directly enforceable federal law would limit the flexibility that the U.S. government normally enjoys in this realm to develop U.S. positions with respect to customary international law and also, in some instances, not to take a position.  As the Supreme Court observed in Sabbatino, “when articulating principles of international law in its relations with other states, the Executive Branch speaks not only as an interpreter of generally accepted and traditional rules, as would the courts, but also as an advocate of standards it believes desirable for the community of nations and protective of national concerns.”

In short, the case against the modern position is stronger today than when we wrote our initial article.

The Continued Relevance of Customary International Law

Although there is no affirmative evidence that federal courts (or any other government actors) have accepted the modern position, and a great deal of evidence that they have not, it is worth remembering, as we pointed out in our 2007 article, that the implausibility of the modern position does not mean that customary international law is irrelevant to domestic litigation.  Congress can authorize its application, and it has done so in various ways, such as in the criminal piracy statute and in aspects of the FSIA.  In addition, courts have long applied the Charming Betsy canon of construction, pursuant to which they attempt to construe statutes to avoid violations of international law, including customary international law.  Moreover, when courts develop select rules of federal common law, it may be appropriate for them to take account of customary international law along with other materials and considerations, including most notably the policies reflected in enacted federal law.  As we have noted, to the extent that courts have considered customary international law in the official immunity context, that is how they have done so.


Again, we are grateful to Bill for engaging with our article.  A lot has happened in the twenty-five years since we published it, and we think our arguments in that piece have been vindicated.  We have also been gratified to see how much richer the debates have become in the foreign relations law area, on this and other topics, since we wrote that article.  To the extent that we played some role in helping to spur this development, we view that as the more important contribution of the piece.